Tracking the position of rolled paper

Abstract

A method for facilitating web position tracking is disclosed. The method includes receiving a representation of a plurality of first stage anomaly locations representative of locations of anomalies detected on the web at a first web processing stage, receiving a representation of a plurality of second stage anomaly locations representative of locations of anomalies detected on the web at a second web processing stage, and for each of a plurality of different candidate position offsets, comparing the candidate position offset to the plurality of first stage anomaly locations and the plurality of second stage anomaly locations to determine a differential representation associated with the candidate position offset, associating the differential representation with the candidate position offset, and identifying a determined position offset from the candidate position offset based at least in part on the differential representation. Other methods, systems, and computer readable media are disclosed.

Description

Cross-reference of related applications

[0001] This application claims the benefit of U.S. Provisional Patent Application No. 63 / 287,675, filed on 9 December 2021, entitled “WEB POSITION TRACKING,” which is incorporated herein by reference in its entirety. [Technical Field]

[0002] Embodiments of this disclosure relate to position tracking, and more particularly to facilitating the position tracking of rolled paper. [Background technology]

[0003] Tracking the position of rolls of paper in paper mills or cardboard factories is crucial for roll processing. For example, accurate tracking of the roll's position during unwinding and rewinding on a rewinding machine can help identify and repair defects in the roll during processing. Several known systems for roll position tracking can be used by attaching codes to the roll at regular intervals and then using ink or laser markers to read the codes as representing the position on the roll. [Overview of the Initiative] [Problems that the invention aims to solve]

[0004] However, such systems can be cumbersome, expensive, potentially dangerous, and / or require ongoing maintenance. [Means for solving the problem]

[0005] According to various embodiments, a method is provided for facilitating roll position tracking, the method comprising: receiving a plurality of representations of first-stage anomaly locations representing the locations of anomalies detected on the roll in a first roll processing stage; receiving a plurality of representations of second-stage anomaly locations representing the locations of anomalies detected on the roll in a second roll processing stage; for each of a plurality of different candidate position offsets, comparing the candidate position offset with a plurality of first-stage anomaly locations and a plurality of second-stage anomaly locations to determine a difference representation associated with the candidate position offset; and associating the difference representation with the candidate position offset. The method further comprises identifying a position offset determined from the candidate position offset, at least in part on the difference representation. Comparing a candidate position offset with multiple first-stage anomaly positions and multiple second-stage anomaly positions may involve applying the candidate position offset to multiple first-stage anomaly positions to determine multiple offset first-stage anomaly positions, and determining the difference between the multiple offset first-stage anomaly positions and multiple second-stage anomaly positions. Determining the difference may involve determining the offset difference for each of the multiple offset first-stage anomaly locations.

[0006] Applying a candidate position offset may involve determining, for each of several first-stage anomaly positions, at least one candidate offset distance adjusted based on the position of the first-stage anomaly position, and adding at least one candidate offset distance to the first-stage anomaly position. Comparing a candidate position offset with multiple first-stage anomaly positions and multiple second-stage anomaly positions may involve applying the candidate position offset to multiple second-stage anomaly positions to determine multiple offset second-stage anomaly positions, and determining the difference between the multiple offset second-stage anomaly positions and multiple first-stage anomaly positions. Determining the difference may involve determining the offset difference for each of the multiple offset second-stage anomaly locations. Applying a candidate position offset may involve determining, for each of several second-stage anomaly locations, at least one candidate offset distance adjusted based on the location of the second-stage anomaly, and adding at least one candidate offset distance to the second-stage anomaly location. Comparing a candidate position offset with multiple first-stage anomaly positions and multiple second-stage anomaly positions may involve determining, for each offset difference, the respective weighted offset difference based on the offset difference and the anomaly weight associated with the offset difference. The method may include determining the associated anomaly weight for each offset difference based on the severity of the anomaly for which the associated offset difference was determined.

[0007] Receiving multiple representations of second-stage anomalies may include receiving multiple representations of candidate second-stage anomalies and determining multiple second-stage anomalies as a subset of multiple candidate second-stage anomalies. Determining multiple Stage II anomalies may involve ranking multiple candidate Stage II anomalies and selecting a subset of one or more top-ranked candidate Stage II anomalies. Ranking multiple candidate second-stage anomaly locations may involve ranking each of the multiple candidate second-stage anomaly locations based at least in part on the proximity of the second-stage anomaly location to the position of the paper roll in the second paper roll processing stage. Ranking multiple candidate stage 2 anomaly locations may involve ranking each of the multiple candidate stage 2 anomaly locations based at least in part on the severity of the anomaly associated with that location. The second paper winding processing stage may be downstream of the first paper winding processing stage.

[0008] Identifying the determined position offset may involve identifying a candidate position offset among several candidate position offsets associated with the smallest of the difference representations. Receiving representations of multiple first-stage anomaly locations may include receiving one or more sets of first-stage images of the rolled paper in a first rolled paper processing step, and determining multiple first-stage anomaly locations, at least partially based on the application of at least one first-stage anomaly identification sensitivity to one or more sets of first-stage images. One or more sets of first-stage images may include a first set of first-stage images and a second set of first-stage images, and at least one first-stage anomaly detection sensitivity may include a first first-stage anomaly detection sensitivity and a second first-stage anomaly detection sensitivity, the second first-stage anomaly detection sensitivity being different from the first first-stage anomaly detection sensitivity. Determining multiple first-stage anomaly locations may include determining a first set of first-stage anomaly locations based at least partially on the application of a first first-stage anomaly detection sensitivity to a first set of first-stage images, and determining a second set of first-stage anomaly locations based at least partially on the application of a second first-stage anomaly detection sensitivity to a second set of first-stage images. Each of the first and second first-stage anomaly detection sensitivities may include multiple anomaly detection thresholds, each associated with a specific pixel location.

[0009] The method may include determining at least one anomaly density associated with a first first-stage anomaly recognition sensitivity, and determining a second first-stage anomaly recognition sensitivity based at least one anomaly density associated with the first first-stage anomaly recognition sensitivity and at least in part on the first first-stage anomaly recognition sensitivity. Determining at least one anomaly density may include determining the number of anomalies represented by a first set of multiple first-stage anomaly locations. Determining at least one anomaly density may involve determining at least one number of anomaly pixels included in an anomaly represented by a first set of multiple first-stage anomaly locations. The method may include determining at least one difference between at least one anomaly density associated with a first first-stage anomaly recognition sensitivity and a desired first-stage anomaly density, and determining a second first-stage anomaly recognition sensitivity may include determining a second first-stage anomaly recognition sensitivity based at least in part on the determined at least one difference.

[0010] Receiving multiple representations of second-stage anomaly locations may include receiving one or more sets of second-stage images of the rolled paper in a second rolled paper processing stage, and determining multiple second-stage anomaly locations based at least partially on the application of at least one second-stage anomaly recognition sensitivity to one or more sets of second-stage images. One or more sets of second-stage images may include a first set of second-stage images and a second set of second-stage images, and at least one second-stage anomaly recognition sensitivity may include a first second-stage anomaly recognition sensitivity and a second second-stage anomaly recognition sensitivity. Determining multiple second-stage anomaly locations may include determining a first set of second-stage anomaly locations based at least partially on the application of a first second-stage anomaly recognition sensitivity to a first set of second-stage images, and determining a second set of second-stage anomaly locations based at least partially on the application of a second second-stage anomaly recognition sensitivity to a second set of second-stage images. The method may include determining at least one anomaly density associated with a first second-stage anomaly recognition sensitivity, and determining a second second-stage anomaly recognition sensitivity based at least partially on the at least one anomaly density associated with the first second-stage anomaly recognition sensitivity and the first second-stage anomaly recognition sensitivity. The method may also include determining the difference between the at least one anomaly density associated with the first second-stage anomaly recognition sensitivity and a desired second-stage anomaly density, where the desired second-stage anomaly density is less than 90% of the desired first-stage anomaly density. Determining the second second-stage anomaly recognition sensitivity may include determining the second second-stage anomaly recognition sensitivity based at least partially on the determined difference between the at least one anomaly density associated with the first second-stage anomaly recognition sensitivity and the desired second-stage anomaly recognition sensitivity.

[0011] Determining at least one anomaly density associated with a first second-stage anomaly recognition sensitivity may involve determining the number of anomaly pixels included in an anomaly represented by a first set of multiple second-stage anomaly locations. Determining at least one anomaly density associated with a first second-stage anomaly discrimination sensitivity may include determining the number of anomalies represented by a first set of a plurality of second-stage anomaly positions. Each of the first and second second-stage anomaly discrimination sensitivities may include a plurality of anomaly discrimination thresholds each associated with respective pixel positions. Receiving a representation of a plurality of second-stage anomaly positions may include receiving one or more sets of second-stage images of the web in a second web handling stage and determining a plurality of second-stage anomaly positions based at least in part on the application of at least one second-stage anomaly discrimination sensitivity to the one or more sets of second-stage images. The one or more sets of second-stage images may include a first set of second-stage images and a second set of second-stage images, the at least one second-stage anomaly discrimination sensitivity may include a first second-stage anomaly discrimination sensitivity and a second second-stage anomaly discrimination sensitivity, and the second second-stage anomaly discrimination sensitivity is different from the first second-stage anomaly discrimination sensitivity. Determining a plurality of second-stage anomaly positions may include determining a first set of second-stage anomaly positions based at least in part on the application of the first second-stage anomaly discrimination sensitivity to the first set of second-stage images and determining a second set of second-stage anomaly positions based at least in part on the application of the second second-stage anomaly discrimination sensitivity to the second set of second-stage images.

[0012] Determining at least one anomaly density associated with a first second-stage anomaly discrimination sensitivity and determining a second second-stage anomaly discrimination sensitivity based at least in part on the at least one anomaly density associated with the first second-stage anomaly discrimination sensitivity and the first second-stage anomaly discrimination sensitivity. Determining at least one anomaly density may include determining the number of anomalies represented by a first set of a plurality of second-stage anomaly positions. Determining at least one anomaly density may involve determining at least one number of anomaly pixels included in an anomaly represented by a first set of multiple second-stage anomaly locations.

[0013] The method may include determining at least one difference between at least one anomaly density associated with a first second-stage anomaly discrimination sensitivity and a desired second-stage anomaly density, and determining a second second-stage anomaly discrimination sensitivity may include determining a second second-stage anomaly discrimination sensitivity based at least in part on the determined at least one difference. Each of the first and second first-stage anomaly detection sensitivities may include multiple anomaly detection thresholds, each associated with a specific pixel location. The method may include receiving a calibration set of first-stage images of the rolled paper in a first rolled paper processing step; determining at least one first-stage calibration anomaly density based at least in part on the application of first-stage calibration anomaly detection sensitivity to the calibration set of first-stage images; and determining a calibration-based first-stage anomaly detection sensitivity based at least in part on the first-stage calibration anomaly detection sensitivity and at least one first-stage calibration anomaly density. Determining at least one first-stage calibration anomaly density may include determining a first-stage calibration set of anomaly locations, at least in part, based on the application of calibration first-stage anomaly discrimination sensitivity to a calibration set of first-stage images, and determining the number of anomalies represented by the first-stage calibration set of anomaly locations.

[0014] Determining at least one first-stage calibration anomaly density may involve determining the number of anomaly pixels included in the calibration set of first-stage images. Each of the first-stage calibrated anomaly detection sensitivity and the calibration-based first-stage anomaly detection sensitivity may include multiple anomaly detection thresholds, each associated with a specific pixel location. Determining the calibration-based first-stage anomaly discrimination sensitivity may include determining at least one difference between at least one first-stage calibrated anomaly density and a desired first-stage calibrated anomaly density, and determining the calibration-based first-stage anomaly discrimination sensitivity at least partially based on the determined at least one difference. The first set of first-stage images may include at least one of the calibration sets of first-stage images. The first set of first-stage images and the calibration set of first-stage images may be the same images.

[0015] The method may include receiving a representation of one or more detected first-stage defect locations representing the location of a defect detected on the roll paper in a first roll paper processing stage, receiving a representation of a detected second-stage location of the roll paper in a second roll paper processing stage, wherein the detected second-stage location represents the current position of the roll paper in the second roll paper processing stage, determining the defect proximity of the detected second-stage location to at least one of the defects based on the determined position offset, the detected second-stage location of the roll paper, and the one or more detected first-stage defect locations, and generating a signal to adjust the processing in the second roll paper processing stage if the determined defect proximity meets a threshold criterion. Determining the defect proximity may include applying the determined positional offset to one or more detected first-stage defect locations to determine one or more predicted second-stage defect locations that represent the predicted defect locations for the roll paper in the second roll paper processing stage, and comparing one or more predicted second-stage defect locations with the detected second-stage locations. Determining the defect proximity may involve applying the determined positional offset to the detected second-stage position to determine the offset detection position, and comparing one or more detected first-stage defect positions to the offset detection position.

[0016] According to various embodiments, a system is provided for facilitating roll paper position tracking, the system including at least one processor configured to perform one of the methods described above. According to various embodiments, a non-temporary computer-readable medium is provided that, when executed by at least one processor, stores code causing at least one processor to perform one of the methods described above. Other aspects and features of embodiments of this disclosure will become apparent to those skilled in the art when the following description of a particular embodiment of this disclosure is considered in conjunction with the accompanying drawings. [Brief explanation of the drawing]

[0017] [Figure 1] This is a schematic diagram of a system for facilitating the tracking of the position of rolled paper, according to various embodiments. [Figure 2] Figure 1 is a schematic diagram of a position tracker system, including a processor circuit, according to various embodiments. [Figure 3] This is a schematic diagram of the first anomaly detector of the system shown in Figure 1, including a processor circuit, according to various embodiments. [Figure 4] This is a schematic diagram of a second anomaly detector of the system shown in Figure 1, including a processor circuit, according to various embodiments. [Figure 5] This flowchart illustrates blocks of code for instructing the position tracker shown in Figure 2, in order to facilitate the roll paper position tracking function in various embodiments. [Figure 6] This flowchart illustrates blocks of code for instructing the first anomaly detector shown in Figure 3, in order to facilitate anomaly detection for the roll paper position tracking function in various embodiments. [Figure 7] Figure 3 shows first-stage images received by the first anomaly detector according to various embodiments. [Figure 8] This is a diagram of a first-stage abnormal position recording that can be used in the system shown in Figure 1, according to various embodiments. [Figure 9] This flowchart illustrates the code blocks that may be included in the flowchart shown in Figure 6, according to various embodiments. [Figure 10] This flowchart illustrates blocks of code for instructing the second anomaly detector shown in Figure 3, in order to facilitate anomaly detection for the roll paper position tracking function in various embodiments. [Figure 11] Figure 4 shows a second-stage image received by the second anomaly detector, according to various embodiments. [Figure 12] This is a diagram of a second-stage abnormal position recording that can be used in the system shown in Figure 1, according to various embodiments. [Figure 13] This is a diagram of a second-stage abnormal position recording that can be used in the system shown in Figure 1, according to various embodiments. [Figure 14] This is a diagram of candidate position offset records that can be used in the system shown in Figure 1, according to various embodiments. [Figure 15] This flowchart illustrates the code blocks that may be included in the flowchart shown in Figure 5, according to various embodiments. [Figure 16] This is a diagram of an offset first-stage abnormal position record that can be used in the system shown in Figure 1, according to various embodiments. [Figure 17] This figure shows a candidate position offset difference record that can be used in the system shown in Figure 1, according to various embodiments. [Figure 18] This is a diagram of candidate position offset records that can be used in the system shown in Figure 1, according to various embodiments. [Figure 19] This is a diagram of candidate position offset records that can be used in the system shown in Figure 1, according to various embodiments. [Figure 20] This is a diagram of a determined positional offset record that can be used in the system shown in Figure 1, according to various embodiments. [Figure 21]This is a diagram of a determined positional offset record that can be used in the system shown in Figure 1, according to various embodiments. [Figure 22] This flowchart illustrates blocks of code that may be executed by the location tracker after or during the execution of the flowchart shown in Figure 5, according to various embodiments. [Figure 23] This is a diagram of a detected defect location record that can be used in the system shown in Figure 1, according to various embodiments. [Figure 24] This is a diagram of a predicted defect location record that can be used in the system shown in Figure 1, according to various embodiments. [Figure 25] This is a flowchart illustrating blocks of code that can be executed by the first anomaly detector shown in Figure 3, according to various embodiments. [Figure 26] This is a diagram of a first-stage calibration anomaly position record that can be used in the system shown in Figure 1 according to various embodiments. [Figure 27] This is a schematic diagram of a system for facilitating the tracking of the position of rolled paper, according to various embodiments. [Figure 28] Figure 27 is a schematic diagram of a position tracker system, including a processor circuit, according to various embodiments. [Figure 29] This is a schematic diagram of the first anomaly detector of the system shown in Figure 27, which includes a processor circuit according to various embodiments. [Figure 30] This is a schematic diagram of the second anomaly detector of the system shown in Figure 27, which includes a processor circuit according to various embodiments. [Figure 31] This flowchart illustrates blocks of code for instructing the position tracker shown in Figure 28, in order to facilitate the roll paper position tracking function in various embodiments. [Figure 32] This flowchart illustrates blocks of code for instructing the first anomaly detector shown in Figure 29, in order to facilitate anomaly detection for the roll paper position tracking function in various embodiments. [Figure 33]This is a diagram of a first-stage abnormal threshold record that can be used in the system shown in Figure 27, according to various embodiments. [Figure 34] This is a diagram of an updated first-stage anomaly density record that can be used in the system shown in Figure 27, according to various embodiments. [Figure 35] This is a diagram of a first-stage abnormal threshold record that can be used in the system shown in Figure 27, according to various embodiments. [Figure 36] This flowchart illustrates blocks of code for instructing the second anomaly detector shown in Figure 29, in order to facilitate anomaly detection for the roll paper position tracking function in various embodiments. [Figure 37] This is a depiction of code blocks that may be included in the flowchart shown in Figure 36, according to various embodiments. [Figure 38] This is a diagram of a second-stage abnormal location severity record that can be used in the system shown in Figure 1, according to various embodiments. [Figure 39] This is a diagram of a second-stage abnormal location severity record that can be used in the system shown in Figure 1, according to various embodiments. [Figure 40] This is a depiction of code blocks that may be included in the flowchart shown in Figure 31, according to various embodiments. [Figure 41] This is a depiction of code blocks that may be included in the flowchart shown in Figure 31, according to various embodiments. [Figure 42] This is a diagram of candidate position offset records that can be used in the system shown in Figure 27, according to various embodiments. [Figure 43] This figure shows an offset second-stage abnormal position record that can be used in the system shown in Figure 27 according to various embodiments. [Figure 44] This is a diagram of candidate position offset difference recording that can be used in the system shown in Figure 27 according to various embodiments. [Figure 45]This figure shows a weighted candidate position offset difference record that can be used in the system shown in Figure 27, according to various embodiments. [Figure 46] This is a diagram of candidate position offset records that can be used in the system shown in Figure 27, according to various embodiments. [Figure 47] This is a diagram of a determined positional offset record that can be used in the system shown in Figure 27, according to various embodiments. [Figure 48] This flowchart illustrates blocks of code that may be executed by the location tracker after or during the execution of the flowchart shown in Figure 31, according to various embodiments. [Modes for carrying out the invention]

[0018] For example, the basic function of a paper roll processing facility such as a paper mill may include processing such as papermaking in a paper machine, where a portion of the paper roll may be wound onto a larger reel, then unwound, re-wound, and cut into smaller rolls that may represent the final product delivered to the customer. In various embodiments, a paper roll inspection system installed in the paper machine may detect defects such as holes and other defects along with their position on the paper roll, which may be important for the final product quality to ensure that the largest holes and other major defects are removed or corrected between the unwinding and rewinding of the paper roll. In various embodiments, to facilitate this, the position of the paper roll on the reel may be tracked during unwinding or conversion so that previously identified defects can be predicted. In various embodiments, accurately tracking the position of the paper roll may be difficult for a variety of reasons, including, for example, when an operator may take an unknown amount of product (e.g., tens or hundreds of meters) from the reel.

[0019] Referring to Figure 1, schematic diagrams of a system 100 for facilitating roll position tracking in various embodiments are provided. The system 100 includes a position tracker 102 communicating with a first anomaly detector 104 and a second anomaly detector 106. In some embodiments, the first anomaly detector 104 and the second anomaly detector 106 may be configured to detect or sense anomalies in a portion of the roll or roll 108 in a first roll processing step 105 and a second roll processing step 107, respectively. In various embodiments, the roll 108 may be a portion of the roll larger than the roll that is wound onto the reel in the first roll processing step 105. In various embodiments, the roll paper 108 is shown for illustrative purposes in both the first roll paper processing step 105 and the second roll paper processing step 107 of Figure 1, but it will be understood that the roll paper 108 may first be in the first roll paper processing step 105, and then, after processing is completed in the first roll paper processing step, the roll paper 108 may move to the second roll paper processing step 107. Thus, the second roll paper processing step may be downstream of the first roll paper processing step. In some embodiments, the anomalies detected in the first roll processing step 105 and the second roll processing step 107 may each represent minor defects in the roll 108, and combined, they may represent a unique fingerprint or map of the location or position of anomalies in the roll 108. In various embodiments, matching the fingerprint or anomaly location map in the first roll processing step 105 and the second roll processing step 107 can facilitate tracking the location in the second roll processing step 107 to the location detected in the first roll processing step.

[0020] In some embodiments, the first anomaly detector 104 can communicate with one or more first cameras 120 configured to image the roll paper 108 while the roll paper is being wound onto a reel during papermaking in the paper machine, which can function as a first roll paper processing stage 105. In various embodiments, the first anomaly detector 104 can communicate with a first position sensor 122 configured to detect the machine-direction position or location of the roll paper associated with each of the images acquired by one or more first cameras 120 of the first anomaly detector (e.g., a rotary encoder configured to detect roller positions, or a device that can estimate the position by integrating speed measurements such as tachometer readings). In various embodiments, the machine-direction position can represent a position along the length of the roll paper 108. In various embodiments, the machine-direction position may generally increase over time during the processing of the roll paper 108 in the first roll paper processing stage 105. In various embodiments, the increasing machine direction position may be periodically reset to zero, for example, when a reel is completed in the paper machine and / or when paper is started to be wound onto a new reel. In various embodiments, the machine direction position from just before the reset may be stored so that it can be used later as an estimate of the total length of the roll of paper 108.

[0021] In some embodiments, the second anomaly detector 106 can communicate with one or more second cameras 124 configured to image the roll of paper 108 during rewinding, which can function as a second roll processing stage 107. In various embodiments, the second anomaly detector 106 can communicate with a second position sensor 126 (which may be substantially similar to, for example, a first position sensor 122 included in the first anomaly detector 104) configured to detect the position or location of the machine-oriented roll of paper associated with each of the images acquired by one or more second cameras 124. In various embodiments, the position tracker 102 can communicate with the second position sensor 126 so that the position tracker 102 can be configured to track the current roll of paper position in the second roll processing stage 107.

[0022] In some embodiments, the system 100 may include a roll processing driver 130 that communicates with a position tracker 102. In various embodiments, the roll processing driver 130 may be configured to control the processing speed or movement of the roll 108 in a second roll processing stage 107. In some embodiments, the roll processing driver 130 may include, for example, a programmable logic controller (PLC).

[0023] In various embodiments, the position tracker 102 may be configured to facilitate the tracking of the roll of paper 108 so that the location of features detected in the first roll processing step 105, such as defects or major defects, can be converted into the expected location of the features in the second roll processing step 107. Thus, in various embodiments, the position tracker 102 can facilitate the prediction of future features during processing in the second roll processing step 107 based on their previous detection in the first roll processing step 105, and this predictive capability can facilitate improved processing of the roll of paper 108 using the system 100. In various embodiments, the anomalies described herein are numerous and may represent minor defects, while the defects described herein may represent less common features that may be relevant from a product quality standpoint, for example, to the grading of the final paper roll. In various embodiments, the defects described herein may include defects that may represent features that require adjustment in processing, such as enabling repair or removal of the defect.

[0024] Referring to Figure 1, in various embodiments, the position tracker 102 may be configured to receive a representation of a first-stage anomaly location, which represents the location of an anomaly detected on the roll paper 108 during the first roll paper processing stage 105. In various embodiments, the position tracker 102 may be configured to receive a representation of a first-stage anomaly location from the first anomaly detector 104 and store the first-stage anomaly location in memory. In some embodiments, the anomaly may be detected during the processing of the roll paper 108 during the first roll paper processing stage 105.

[0025] In some embodiments, the first anomaly detector 104 may be configured to receive one or more sets of first-stage images of the roll paper 108 in a first roll paper processing stage 105 from one or more first cameras 120. In various embodiments, the first anomaly detector 104 may be configured to receive the associated machine direction position for each received image from a first position sensor 122. In various embodiments, the first anomaly detector 104 may be configured to identify anomalies in the images and determine the position for each anomaly. In various embodiments, the position for each anomaly may include a machine direction position and a cross direction position. In various embodiments, the machine direction position may represent a position along the length of the roll paper 108, and the cross direction may represent a position across or along the width of the roll paper 108. In various embodiments, the machine direction and cross direction may be perpendicular to each other. In various embodiments, the machine direction position may be determined based on the pixel position of the anomaly in the image and the machine direction position associated with the image in which the anomaly was detected, for example, by converting the pixel position to real-world units such as meters and then summing them up. In various embodiments, the crossing direction position may be determined based on the pixel position of the anomaly in the image, as well as a calibrated mapping from the (usually horizontal) pixel coordinates to a real-world position (e.g., measured from some fixed zero point and / or from the edge of the roll of paper) (e.g., on a meter or millimeter basis).

[0026] In various embodiments, the position tracker 102 may be configured to receive a representation of a second-stage anomaly location, which represents the location of an anomaly detected on the roll paper 108 during the second roll paper processing stage 107. In various embodiments, the position tracker 102 may be configured to receive a representation of a second-stage anomaly location from the second anomaly detector 106 and store the second-stage anomaly location in memory. In some embodiments, the anomaly may be detected during the processing of the roll paper 108 during the second roll paper processing stage 107.

[0027] In some embodiments, the second anomaly detector 106 may be configured to receive one or more sets of second-stage images of the roll paper 108 in the second roll paper processing stage 107 from one or more second cameras 124, and to receive the associated machine orientation position for each image from the second position sensor 126. In various embodiments, the second anomaly detector 106 may be configured to identify anomalies in the images and determine the location of each anomaly, generally as described above with respect to the first roll paper processing stage 105. In some embodiments, portions of the roll paper 108 are analyzed in reverse order in the second roll paper processing stage 107 (due to this stage involving the unwinding of the reel wound in the first roll paper processing stage 105), but anomalies may appear in reverse order in the second roll paper processing stage, as the machine orientation position may increase over time during imaging in the first roll paper processing stage 105, and the machine orientation position in the second roll paper processing stage may decrease over time during imaging. In various embodiments, the second position sensor 126 may provide an increasing machine direction position as the roll of paper 108 is processed in a second roll processing stage, and the position tracker 102 may be configured to calculate the corresponding decreasing machine direction position by subtracting each increasing machine direction position from an estimate of the total length of the roll of paper 108, which may be provided or detected in advance in the first roll processing stage 105.

[0028] In various embodiments, the position tracker 102 may be configured to compare the footprints or maps of abnormal positions from the first roll processing stage 105 and the second roll processing stage 107 to determine position offsets that can be applied to the first-stage abnormal positions such that the abnormal position map in the second roll processing stage 107 is aligned with or matches a portion of the offset version of the abnormal position map from the first roll processing stage 105.

[0029] In various embodiments, the position tracker 102 may be configured to compare each of several different candidate position offsets with a first-stage anomaly position and a second-stage anomaly position to determine a representation of the difference associated with the candidate position offset, and to associate the representation of the difference with the candidate position offset. For example, in some embodiments, the position tracker 102 may be configured to apply the candidate position offset to the first-stage anomaly position to determine the offset first-stage anomaly position, determine the difference between the offset first-stage anomaly position and the second-stage anomaly position, and to associate the representation of the difference with the candidate position offset. Alternatively, in some embodiments, the position tracker 102 may be configured to apply the candidate position offset to the second-stage anomaly position to determine several offset second-stage anomaly positions, and to determine the difference between the offset second-stage anomaly positions and the first-stage anomaly position.

[0030] In various embodiments, trying different candidate position offsets and checking the representation of the difference between the resulting offset anomaly position and the second-stage anomaly position can facilitate a reduction in processing power requirements to reach the precisely determined position offset. In some embodiments, the difference may include distance. In various embodiments, comparing discrete position offsets applied to Boolean anomalies with associated anomaly positions (either they are present or absent) can facilitate tracking that is less dependent on specific light conditions, light angles, camera angles, and other factors that may change between the first roll paper processing stage 105 and the second roll paper processing stage 107, instead of other comparisons such as quantitative anomaly signals. In various embodiments, it may be advantageous that the need for effort to make such factors nearly similar between the two processing stages is reduced. In various embodiments, comparing discrete position offsets applied to Boolean anomalies with associated anomaly positions can facilitate variability dependent on material changes, such as the drying of the paper between the first roll paper processing stage 105 and the second roll paper processing stage 107, if the roll paper 108 is paper.

[0031] In some embodiments, the position tracker 102 may be configured to determine the representation of the difference using a cost function. In various embodiments, the position tracker 102 may be configured to identify the determined position offset based at least partially on the representation of the difference. In some embodiments, the position tracker 102 may be configured to identify a candidate position offset among a plurality of candidate position offsets associated with the smallest of the difference representations. In various embodiments, the identified candidate position offset may be the best offset to use as the determined position offset for position tracking. In various embodiments, the determined position offset may be used to facilitate position tracking of the roll in the second roll processing stage 107 for position tracking in the first roll processing stage 105. For example, in some embodiments, the determined position offset can facilitate mapping the position of the roll 108 in the first roll processing stage 105 to its position on the roll 108 in the second roll processing stage 107, and / or vice versa. In some embodiments, the position tracker 102 may be configured to apply or add the determined position offset to the position of the feature on the roll 108 in the first roll processing stage 105 to convert the position of the feature in the first roll processing stage 105 to an updated position corresponding to the position of those same feature in the second roll processing stage 107. In various embodiments, the ability to convert between the first and second winding paper processing stages can facilitate the adaptive processing of the winding paper 108 in the second winding paper processing stage 107, taking into account what is detected in the first winding paper processing stage.

[0032] In some embodiments, the position tracker 102 may be configured to receive a representation of one or more detected first-stage defect locations representing the location of defects detected on the roll paper 108 in the first roll paper processing stage 105, and to receive a representation of a detected second-stage location of the roll paper 108 in the second roll paper processing stage, where the detected second-stage location represents the current position of the roll paper 108 in the second roll paper processing stage 107. In various embodiments, the position tracker 102 may be configured to determine the defect proximity of the detected second-stage location to at least one of the defects based on the determined position offset, the detected second-stage location of the roll paper, and one or more detected first-stage defect locations, and to generate a signal to adjust the processing in the second roll paper processing stage 107 if the determined defect proximity meets a threshold criterion. For example, in some embodiments, the threshold criterion may include a threshold distance criterion, where the defect proximity can meet the threshold criterion if the defect proximity is less than the threshold distance.

[0033] In various embodiments, the defect may be a major defect for which processing may need to be adjusted. In some embodiments, the position tracker 102 may be configured to determine the defect proximity by applying the determined position offset to one or more detected first-stage defect locations to determine one or more predicted second-stage defect locations representing the predicted defect locations for the roll paper in the second roll paper processing stage, and by comparing one or more predicted second-stage defect locations with the detected second-stage locations. For example, the position tracker may compare by determining the difference or distance between the nearest next of the one or more predicted second-stage defect locations and the detected second-stage location. In various embodiments, the position tracker 102 may be configured to generate a signal to slow down the roll paper processing driver 130 shown in Figure 1 when the detected second-stage location of the roll paper 108 is within a threshold distance prior to one of the predicted second-stage defect locations. In some embodiments, the position tracker 102 may be configured to generate a signal to stop the roll paper processing driver 130 at each of the predicted defect locations or before them. In various embodiments, this deceleration and / or stopping can facilitate the correction and / or removal of defects from the roll paper 108 in the second roll paper processing step 107.

[0034] Tracker-Processor Circuit Referring here to Figure 2, schematic diagrams of the position tracker 102 of the system 100 shown in Figure 1 are shown according to various embodiments. Referring to Figure 2, the position tracker 102 includes a processor circuit including a tracker processor 200 and program memory 202, and a storage memory 204 and an input / output (I / O) interface 212, all of which communicate with the tracker processor 200. In various embodiments, the tracker processor 200 and / or similar processors disclosed herein may include one or more processing units such as a central processing unit (CPU), a graphics processing unit (GPU), and / or a field-programmable gate array (FPGA). In some embodiments, any or all of the functions of the position tracker 102 described herein may be implemented using one or more FPGAs.

[0035] The I / O interface 212 includes interfaces 220 and 222 for communicating with the first anomaly detector 104 and the second anomaly detector 106 shown in Figure 1, respectively. In various embodiments, the I / O interface 212 may include an interface 224 for communicating with the paper winding driver 130 shown in Figure 1 and an interface 226 for communicating with the second position sensor 126. In some embodiments, the I / O interface 212 and / or similar interfaces disclosed herein may include interfaces for facilitating network communication over a network such as a local area network or the Internet, and / or one or more interfaces for enabling user input via one or more user interface devices such as a pointer and / or a keyboard. In some embodiments, any or all of the interfaces and / or similar interfaces disclosed herein can facilitate wireless and / or wired communication. In some embodiments, each of the interfaces and / or similar interfaces disclosed herein may include one or more interfaces, and / or some or all of the interfaces may be implemented as a combined interface or a single interface.

[0036] In some embodiments described herein as devices that receive or transmit information, it will be understood that a device receives or generates a signal representing information via its interface and transmits that signal to another device via its interface. Processor-executable program code for instructing the tracker processor 200 to perform various functions is stored in program memory 202. Referring to Figure 2, program memory 202 includes a block 206 of code for instructing the position tracker 102 to facilitate the roll-up paper position tracking function. In this specification, it may be stated that a particular encoded entity, such as an application or module, performs a particular function. In this specification, when an application, module, or encoded entity is described to act, for example, as part of a function or method, it will be understood that at least one processor (e.g., the tracker processor 200) is instructed to act by programmable code or processor-executable code or instructions that define or form part of the application.

[0037] The memory 204 includes a plurality of storage locations, including location 244 for storing first-stage anomaly location data, location 256 for storing second-stage anomaly location data, location 264 for storing candidate location offset data, location 268 for storing offset first-stage anomaly location data, location 270 for storing cross-direction weight data, location 272 for storing determined difference data, location 274 for storing determined location offset data, location 276 for storing detected defect location data, location 278 for storing detected second-stage location data, location 280 for storing predicted defect location data, location 282 for storing defect proximity data, and location 284 for storing defect threshold data. In various embodiments, the storage locations may be stored in a database within the memory 204.

[0038] In various embodiments, block 206 of code disclosed herein and / or any similar block of code may be integrated into a single block of code, or a portion of block 206 of code may include one or more blocks of code stored in one or more separate locations in program memory 202. In various embodiments, locations 244, 256, 264, and 268-284 and / or any or all of any similar locations disclosed herein may be integrated and / or each may include one or more separate locations in storage memory 204.

[0039] Each of the program memory 202 and storage memory 204 and / or similar memories disclosed herein may be implemented as one or more storage devices, including random access memory (RAM), hard disk drives (HDDs), solid-state drives (SSDs), network drives, flash memory, memory sticks or cards, any other form of non-temporary computer-readable memory or storage medium, and / or combinations thereof. In some embodiments, the program memory 202, storage memory 204, any similar memory, and / or any part thereof disclosed herein may be contained in a device separate from the location tracker 102, for example, and may communicate with the location tracker 102 via an I / O interface 212. In some embodiments, the functions of the tracker processor 200 and / or location tracker 102 and / or similar processor or device described herein may be implemented using multiple processors and / or multiple devices.

[0040] First anomaly detector-processor circuit Referring here to Figure 3, schematic diagrams of the first anomaly detector 104 of the system 100 shown in Figure 1 are shown according to various embodiments. Referring to Figure 3, the first anomaly detector 104 includes a processor circuit including a detector processor 120 and program memory 1202, and a storage memory 1204 and an input / output (I / O) interface 1212, all of which communicate with the detector processor 1200. In various embodiments, the detector processor 1200 and / or similar processors disclosed herein may include one or more processing units such as a central processing unit (CPU), a graphics processing unit (GPU), and / or a field-programmable gate array (FPGA). In some embodiments, any or all of the functions of the first anomaly detector 104 described herein may be implemented using one or more FPGAs.

[0041] The I / O interface 1212 includes interfaces 1220 and 1222 for communicating with one or more first cameras 120 and first position sensors 122 shown in Figure 1, and interface 1224 for communicating with a position tracker 102 shown in Figure 1. In some embodiments, any or all of the interfaces disclosed herein and / or similar interfaces can facilitate wireless and / or wired communication. In some embodiments, each of the interfaces disclosed herein and / or similar interfaces may include one or more interfaces, and / or some or all of the interfaces may be implemented as a coupled interface or a single interface.

[0042] Processor executable program code for instructing the detector processor 1200 to perform various functions is stored in program memory 1202. Referring to Figure 3, program memory 1202 includes a block of code 1206 for instructing the first anomaly detector 104 to facilitate anomaly detection for the roll position tracking function, and a block of code 1208 for instructing the first anomaly detector 104 to detect defects.

[0043] The memory 1204 includes a plurality of storage locations, including a location 1240 for storing first-stage image data, a location 1242 for storing first-stage anomaly identification sensitivity data, a location 1244 for storing first-stage anomaly location data, a location 1246 for storing first-stage anomaly density data, a location 1248 for storing desired first-stage anomaly density data, a location 1250 for storing first-stage anomaly density difference data, a location 1280 for storing calibration image data, a location 1282 for storing calibration anomaly identification sensitivity data, a location 1284 for storing calibration anomaly location data, a location 1286 for storing calibration anomaly density data, a location 1288 for storing desired calibration anomaly density data, and a location 1290 for storing calibration anomaly density difference data. In various embodiments, the storage locations may be stored in a database within the memory 1204.

[0044] In various embodiments, the code blocks 1206 and / or code block 1208 and / or any similar code blocks disclosed herein may be combined into a single code block, or portions of code block 1206 and / or code block 1208 may include one or more codes stored in one or more separate locations within program memory 1202. In various embodiments, any or all of the locations 1240-1250 and 1280-1290 and / or any similar locations disclosed herein may be combined and / or each may include one or more separate locations within storage memory 1204.

[0045] Each of the program memory 1202 and storage memory 1204 and / or similar memories disclosed herein may be implemented as one or more storage devices, including random access memory (RAM), hard disk drives (HDDs), solid-state drives (SSDs), network drives, flash memory, memory sticks or cards, any other form of non-temporary computer-readable memory or storage medium, and / or combinations thereof. In some embodiments, the program memory 1202, storage memory 1204, any similar memory, and / or any portion thereof disclosed herein may be contained in a device separate from the first anomaly detector 104, for example, and may communicate with the first anomaly detector 104 via an I / O interface 1212. In some embodiments, the functions of the detector processor 1200 and / or the first anomaly detector 104 and / or similar processors or devices described herein may be implemented using multiple processors and / or multiple devices.

[0046] Second anomaly detector-processor circuit Referring now to Figure 4, schematic diagrams of the second anomaly detector 106 of the system 100 shown in Figure 1 are shown according to various embodiments. In various embodiments, the second anomaly detector 106 may include elements substantially similar to those included in the first anomaly detector 104 shown in Figure 3. Referring to Figure 4, the second anomaly detector 106 includes a processor circuit including a detector processor 140 and program memory 1402, and a storage memory 1404 and an input / output (I / O) interface 1412, all of which communicate with the detector processor 1400. In various embodiments, the detector processor 1400 and / or similar processors disclosed herein may include one or more processing units, such as a central processing unit (CPU), a graphics processing unit (GPU), and / or a field-programmable gate array (FPGA). In some embodiments, any or all of the functions of the second anomaly detector 106 described herein may be implemented using one or more FPGAs.

[0047] The I / O interface 1412 includes interfaces 1420 and 1422 for communicating with one or more second cameras 124 and second position sensors 126 shown in Figure 1, and interface 1424 for communicating with the position tracker 102 shown in Figure 1. In some embodiments, any or all of the interfaces disclosed herein and / or similar interfaces can facilitate wireless and / or wired communication. In some embodiments, each of the interfaces disclosed herein and / or similar interfaces may include one or more interfaces, and / or some or all of the interfaces may be implemented as a coupled interface or a single interface.

[0048] Processor executable program code for instructing the detector processor 1400 to perform various functions is stored in program memory 1402. Referring to Figure 4, program memory 1402 includes a block 1406 of code for instructing the second anomaly detector 106 to facilitate anomaly detection for the roll paper position tracking function. The memory 1404 includes a plurality of storage locations, including a location 1452 for storing second-stage image data, a location 1454 for storing second-stage anomaly identification sensitivity data, a location 1456 for storing second-stage anomaly location data, a location 1458 for storing second-stage anomaly density data, a location 1460 for storing desired second-stage anomaly density data, a location 1462 for storing second-stage anomaly density difference data, a location 1480 for storing calibration image data, a location 1482 for storing calibration anomaly identification sensitivity data, a location 1484 for storing calibration anomaly location data, a location 1486 for storing calibration anomaly density data, a location 1488 for storing desired calibration anomaly density data, and a location 1490 for storing calibration anomaly density difference data. In various embodiments, the storage locations may be stored in a database within the memory 1404.

[0049] In various embodiments, block 1406 of the code disclosed herein and / or any similar block of code may be integrated into a single block of code, or a portion of block 1406 of code may include one or more blocks of code stored in one or more separate locations in program memory 1402. In various embodiments, any or all of locations 1452-1462 and 1480-1490 and / or any similar locations disclosed herein may be integrated and / or each may include one or more separate locations in storage memory 1404.

[0050] Each of the program memory 1402 and storage memory 1404 and / or similar memories disclosed herein may be implemented as one or more storage devices, including random access memory (RAM), hard disk drives (HDDs), solid-state drives (SSDs), network drives, flash memory, memory sticks or cards, any other form of non-temporary computer-readable memory or storage medium, and / or combinations thereof. In some embodiments, the program memory 1402, storage memory 1404, any similar memory, and / or any portion thereof disclosed herein may be contained in a device separate from, for example, the second anomaly detector 106 and be able to communicate with the second anomaly detector 106 via an I / O interface 1412. In some embodiments, the functions of the detector processor 1400 and / or the second anomaly detector 106 and / or similar processor or device described herein may be implemented using multiple processors and / or multiple devices.

[0051] operation As described above, in various embodiments, the position tracker 102 shown in Figures 1 and 2 may be configured to facilitate roll position tracking. Referring to Figure 5, a flowchart 400 is shown overall illustrating blocks of code for instructing the tracker processor 200 shown in Figure 2 to facilitate roll position tracking in various embodiments. In various embodiments, the blocks of code included in the flowchart 400 may be encoded within blocks of code 206 in the program memory 202 shown in Figure 2.

[0052] Referring to Figure 5, the flowchart 400 begins with block 402, which instructs the processor 200 to receive a representation of a first-stage anomaly location, representing the location of an anomaly detected on the roll paper during a first roll paper processing stage. In some embodiments, block 402 may instruct the tracker processor 200 to receive a representation of a first-stage anomaly location from a first anomaly detector 104. In some embodiments, the first anomaly detector 104 may be configured to receive one or more sets of first-stage images and machine orientation position information associated with each of the images, and to analyze the images to determine the first-stage anomaly location. In some embodiments, the first anomaly detector 104 may be configured to transmit the representation of the first-stage anomaly location to the position tracker 102.

[0053] Referring to Figure 6, a flowchart 500 is shown illustrating blocks of code for instructing the detector processor 1200 shown in Figure 3 to facilitate anomaly detection for roll paper position tracking in various embodiments. In various embodiments, the blocks of code included in the flowchart 500 may be encoded within a block of code 1206 in the program memory 1202 shown in Figure 3.

[0054] Referring to Figure 6, the flowchart 500 begins with block 502, which instructs the detector processor 1200 to receive a set of first-stage images of the rolled paper 108 in the first rolled paper processing stage 105. In some embodiments, block 502 may instruct the detector processor 1200 to receive a set of first-stage images from one or more first cameras 120 of the system 100 shown in Figure 1 via interface 1220 of the I / O interface 1212 shown in Figure 3. In various embodiments, one or more first cameras 120 may be configured to generate first-stage images of the rolled paper 108 in the first rolled paper processing stage 105, including a first set of first-stage images representing a first machine-direction length of the rolled paper 108 during processing in the first rolled paper processing stage 105, the first set of first-stage images including the first-stage image shown in Figure 7.

[0055] For example, in some embodiments, such as when using a matrix camera, the first-stage image may include several redundant or overlapping image regions to avoid mechanical gaps when imaging the roll of paper 108 and / or to facilitate other functions, such as being able to distinguish whether any anomaly is a defect that moves with the roll of paper or, for example, a piece of paper floating in the air.

[0056] In various embodiments, block 502 can receive a first set of first-stage images, including a first-stage image 460, from one or more first cameras 120, and instruct the detector processor 1200 to store the first set of first-stage images at location 1240 in the memory memory 1204 shown in Figure 3. In various embodiments, block 502 can receive an associated mechanical orientation position from a first position sensor 122 shown in Figure 1 for each image included in the first set of first-stage images, and instruct the detector processor 1200 to store the associated mechanical orientation position in relation to each image in the first set of first-stage images stored at location 1240 in the memory memory 1204.

[0057] In some embodiments, the first set of first-stage images received in block 502 may represent a portion of the roll of paper 108, such as the machine-direction length of the roll of paper 108. For example, in some embodiments, the first set of first-stage images may represent the first 1000 m of the roll of paper 108. In various embodiments, considering the roll of paper 108, a set of images and, i.e., a portion at one time, can facilitate adjustment of the anomaly detection sensitivity for each portion along the length of the roll of paper 108. In various embodiments, this can facilitate a more consistent observation of the anomaly density along the mechanical length of the roll of paper 108, despite the fact that the properties of the roll of paper may change along the length of the roll of paper and / or the imaging conditions (e.g., a camera housing that accumulates dirt) change over time. In various embodiments, this can facilitate better roll tracking and / or a more consistent and / or accurate determination of the determined position offset record, as described below.

[0058] Referring to Figure 6, block 504 instructs the detector processor 1200 to determine at least one of the first-stage anomaly locations, at least in part, based on the application of the first-stage anomaly detection sensitivity to the set of first-stage images. In various embodiments, block 504 may instruct the detector processor 1200 to determine a first set of first-stage anomaly locations, based on the application of the first first-stage anomaly detection sensitivity to a first set of first-stage images. In various embodiments, the first first-stage anomaly detection sensitivity may include an anomaly detection threshold, which may be stored in location 1242 of the memory memory 1204 shown in Figure 3. In some embodiments, the first first-stage anomaly detection threshold may be initialized to a value that can be pre-set and selected by the user. For example, in some embodiments, the first first-stage anomaly detection threshold may be set to a value between 1 and 100. In some embodiments, the first first-stage anomaly detection threshold stored in location 1242 may be, for example, 25.

[0059] In various embodiments, block 504 can instruct the detector processor 1200 to determine an anomaly index value for each pixel or pixel location in each of a first set of first-stage images, the anomaly index value representing whether an anomaly may exist at the pixel location. In some embodiments, block 504 can instruct the detector processor 1200 to identify pixel groups associated with an anomaly index value greater than a first-stage anomaly identification threshold stored in location 1242 of the memory memory 1204. In various embodiments, block 504 can instruct the detector processor 1200 to determine the identified pixel groups to represent anomalies. In various embodiments, block 504 can instruct the detector processor 1200 to determine the location for each anomaly using a combination of the location of the identified pixel group and the machine orientation location of the image in which the pixel group was identified.

[0060] In some embodiments, block 504 can instruct the detector processor 1200 to maintain a reference image as a moving average of the field of view. In some embodiments, block 504 can instruct the detector processor 1200 to use exponential smoothing or Fair Exponential Smoothing with Small Alpha (FESSA) (as described, for example, Reunanen, Juha. (2015). Fair Exponential Smoothing with Small Alpha. 10.13140 / RG.2.1.2181.1923). In various embodiments, block 504 can instruct the detector processor 1200 to calculate the difference between the image and the reference image for each image.

[0061] In some embodiments, block 504 can instruct the detector processor 1200 to calculate the difference as new pixel value / reference pixel value - 1 = (new pixel value - reference pixel value) / reference pixel value. In some embodiments, before performing this calculation, block 504 can instruct the detector processor 1200 to set the reference pixel value to be strictly positive, for example by setting reference pixel value = max(reference pixel value; ε), where ε is a small number, e.g., 0.1. In various embodiments, the absolute value of the determined difference may be used as an anomaly index value. In various embodiments, robustness to vignetting can be improved, for example, by dividing by the reference pixel value. In various embodiments, block 504 can instruct the detector processor 1200 to compare an anomaly index value with a first-stage anomaly identification threshold. Pixels whose anomaly index value exceeds the first-stage anomaly identification threshold may be considered anomaly.

[0062] In various embodiments, block 504 can instruct the detector processor 1200 to group nearby pixels associated with anomaly index values ​​exceeding a first stage anomaly identification threshold. In various embodiments, block 504 can instruct the detector processor 1200 to calculate the center and / or bounding box and / or boundary of each group of anomaly pixels, with each group serving as a representation of an anomaly. In various embodiments, each center can serve as a first stage anomaly location in a first set of determined first stage anomaly locations. In various embodiments, block 504 can instruct the detector processor 1200 to determine a center as the center of the bounding box and / or as the centroid of pixels that are part of a group. In various embodiments, block 504 can instruct the detector processor 1200 to weight individual pixels in centroid calculation by the extent to which each individual pixel exceeds a first stage anomaly identification threshold level.

[0063] In some embodiments, block 504 can instruct the detector processor 1200 to classify each group of anomalous pixels using a rule. In various embodiments, block 504 can instruct the detector processor 1200 to classify based at least partially on the size, shape, and pixel value distribution of the anomalous groups of pixels. In some embodiments, such class information may be used in a cost function that can be used to determine the position offset and estimate the position in the second roll-up processing step 107.

[0064] In various embodiments, each of the determined anomaly locations may include, for example, a machine-direction position and an intersecting-direction position in meters. In various embodiments, alternative or additional anomaly identification processes may be used. In various embodiments, block 504 can store a first set of determined first-stage anomaly locations in the first-stage anomaly location record 540, as shown in Figure 8, and can instruct the detector processor 1200 to store the first-stage anomaly location record 540 at location 1244 in the storage memory 1204, as shown in Figure 3. Referring to Figure 8, the first-stage anomaly location record 540 includes anomaly location fields for storing the machine-direction and cross-direction locations for each anomaly. For example, referring to Figure 8, although not all anomaly location fields are shown in Figure 8, the first-stage anomaly location record 540 may include first-stage anomaly location fields 542 and 544 for storing the machine-direction and cross-direction locations of a first anomaly, respectively; second-stage anomaly location fields 546 and 548 for storing the machine-direction and cross-direction locations of a second anomaly, respectively; and 933 anomaly location fields 550 and 552 for storing the machine-direction and cross-direction locations of a 933 anomaly, respectively.

[0065] In some embodiments, flowchart 500 and / or block 504 may include a block of code to instruct the detector processor 1200 to transmit a representation of the first-stage anomaly location to the location tracker 102. For example, in some embodiments, flowchart 500 may include a block to instruct the detector processor 1200 to transmit a signal representing the first-stage anomaly location record 540 shown in Figure 8 to the location tracker 102 via interface 1224 of the I / O interface 1212 shown in Figure 3.

[0066] Referring to Figure 6, in some embodiments, the flowchart 500 may include blocks 506 and 508 that can instruct the detector processor 1200 to determine a second first-stage anomaly recognition sensitivity for application to a second set of first-stage images. In various embodiments, the second first-stage anomaly recognition sensitivity may differ from the first first-stage anomaly recognition sensitivity. Referring to Figure 6, block 506 instructs the detector processor 1200 to determine at least one anomaly density associated with the first-stage anomaly recognition sensitivity. In various embodiments, block 506 can instruct the detector processor 1200 to determine an anomaly density associated with the first first-stage anomaly recognition sensitivity. In some embodiments, the anomaly density may be anomaly count density. In various embodiments, the anomaly count density may be determined as anomalies per length of the roll of paper 108 or as anomalies per area of ​​the roll of paper. In various embodiments, block 506 can instruct the detector processor 1200 to determine the anomaly density by counting how many anomalies are detected within the sampling machine direction proximity of the most recent image included in the first set of first-stage images. For example, in some embodiments, the sampling machine-direction proximity may be 20,000 meters. Thus, in various embodiments, the sampling machine-direction proximity may be greater than the machine-direction length of the portion of the roll of paper 108 represented by the first set of first-stage images.

[0067] In various embodiments, when the roll of paper 108 is imaged at a proximity of less than the sampling machine direction, block 506 can instruct the detector processor 1200 to determine the anomaly density based on all the first-stage images of the received roll of paper 108. For example, in some embodiments, the first set of first-stage images may be the only first-stage images received by the detector processor 1200, and therefore block 506 can instruct the detector processor 1200 to determine the anomaly density based on the first set of first-stage images.

[0068] In various embodiments, block 506 can instruct the detector processor 1200 to determine the number of anomalies represented by a first set of first-stage anomaly locations. For example, in some embodiments, block 506 can instruct the detector processor 1200 to read the first-stage anomaly location record 540 from location 1244 in the storage memory 1204 and determine the number of locations represented by the first-stage anomaly location record 540. In various embodiments where the sampling machine direction proximity is greater than the machine direction length represented by the first set of first-stage images, and position 1244 stores further first-stage anomaly position records, block 506 can instruct the detector processor 1200 to add to the number of positions represented by the first-stage anomaly position records 540 any further positions represented by the anomaly position records, so that the position lies within the sampling machine direction proximity of the most recent (largest machine direction position) image of the first set of first-stage images.

[0069] In various embodiments, block 506 can instruct the detector processor 1200 to determine the anomaly density by dividing by a number determined by the machine direction length represented by the position-considered first-stage image, where the machine direction length may be, for example, 1000m, in some embodiments, such as where only a first set of first-stage images has been analyzed. In some embodiments, the anomaly density may be determined by dividing by a number determined by the corresponding roll area, such as the width of the imaged roll 108 integrated over the relevant machine direction length. In various embodiments, block 506 can instruct the detector processor 1200 to store the determined anomaly density in location 1246 of the memory 1204. For example, in some embodiments, block 506 can instruct the detector processor 1200 to store a first anomaly density of 0.933 anomalies per meter of roll paper 108 in location 1246 of the memory 1204.

[0070] Referring to Figure 6, block 508 instructs the detector processor 1200 to determine a new first-stage anomaly identification sensitivity based at least partially on the determined anomaly density and first-stage anomaly identification sensitivity. In various embodiments, block 508 may instruct the detector processor 1200 to read the first first-stage anomaly identification sensitivity and anomaly density from locations 1242 and 1246 of the memory 1204 and determine a second first-stage anomaly identification sensitivity based at least partially on the first first-stage anomaly identification sensitivity and first anomaly density. In various embodiments, block 508 may instruct the detector processor 1200 to store the second first-stage anomaly identification sensitivity in location 1242 of the memory 1204.

[0071] In various embodiments, block 508 can instruct the detector processor 1200 to adjust the anomaly recognition sensitivity so that the second first-stage anomaly recognition sensitivity is more sensitive (and therefore, for example, a lower threshold) when fewer anomalies than desired are represented by the first anomaly density, or to adjust the anomaly recognition sensitivity so that the second first-stage anomaly recognition sensitivity is less sensitive (and therefore, for example, a higher threshold) when more anomalies than desired are represented by the first anomaly density.

[0072] Referring to Figure 9, a flowchart 580 is shown depicting blocks of code that may be included in block 508 of flowchart 500 shown in Figure 6, according to various embodiments. Flowchart 580 begins with block 582, which instructs the detector processor 1200 to determine the difference between the anomaly density and a desired first-stage anomaly density. In some embodiments, the desired first-stage anomaly density may be pre-set and stored in location 1248 of memory 1204. In some embodiments, the desired first-stage anomaly density may be set based on a desired anomaly density relative to the machine direction length, which can facilitate roll position tracking using the position tracker 102. For example, in some embodiments, the desired first-stage anomaly density may be 1,000 anomalies per meter of machine direction length. In various embodiments, the desired first-stage anomaly density may be pre-set by the user and / or may be adjusted so that, for example, enough anomalies are detected to start tracking early, but at the same time, so that too many anomalies are detected that could cause, for example, processing capacity problems.

[0073] Referring to Figure 9, in various embodiments, block 582 can instruct the detector processor 1200 to determine a difference by subtracting a desired first-stage anomaly density stored at location 1248 of the memory from the anomaly density stored at location 1246 of the memory 1204, and block 582 can instruct the detector processor 1200 to store the difference at location 1250 of the memory 1204. Referring back to Figure 9, block 584 of flowchart 580 can instruct the detector processor 1200 to determine a new first-stage anomaly sensitivity, at least in part, based on the difference determined in block 582. In some embodiments, block 584 can instruct the detector processor 1200 to adjust the sensitivity to be more sensitive when fewer anomalies than desired are detected, and to be less sensitive when more anomalies than desired are detected.

[0074] In various embodiments, block 584 of flowchart 580 can instruct the detector processor 1200 to determine a second first-stage anomaly sensitivity based at least in part on the difference determined in block 582. In various embodiments, the first and second first-stage anomaly discrimination sensitivities may include first and second first-stage anomaly thresholds, and block 584 may use the following formula: TIFF0007880968000001.tif1561 The detector processor 1200 can be instructed to determine a second first-stage anomaly threshold using, where T 12 This is the second first-stage anomaly threshold, T 11 is the first stage anomaly threshold, a1 is the magnification, and D1 is the relative difference determined by dividing the difference determined in block 582 and stored at position 1250 of memory 1204 shown in Figure 3 by the desired first stage anomaly density from position 1248 of memory 1204. In some embodiments, a1 may be set to a small number such as 0.0001, for example, when the first set of first stage images represents 1000 m in the machine direction length of the roll of paper 108. In various embodiments, block 584 can instruct the detector processor 1200 to store a value determined using the above formula as a second first-stage anomaly identification sensitivity in location 1242 of the memory memory 1204 shown in Figure 3.

[0075] Referring to Figure 6, in various embodiments, after block 508 is completed, the detector processor 1200 may return to block 502, where a further set of first-stage images may be received. For example, in various embodiments, a second set of first-stage images may be received in the next execution of block 502. In various embodiments, block 504 may be executed by applying a second first-stage anomaly detection sensitivity stored in memory memory 1204 at location 1242 to the second set of first-stage images. In various embodiments, a second first-stage anomaly location record having a format substantially similar to the first-stage anomaly location record 540 shown in Figure 8 may be generated and stored in memory memory 1204 at location 1244. In various embodiments, blocks 506 and 508 may be executed to determine a third first-stage anomaly detection sensitivity.

[0076] In various embodiments, blocks 502-508 may be performed for each set of first-stage images representing each portion of the roll of paper 108. In various embodiments, after the execution of flowchart 500 is complete, one or more first-stage abnormal position records representing abnormal positions may be stored at position 1244 in memory 1204. Referring to Figure 1, in various embodiments, the flowchart 500 may be executed during the processing of the roll paper 108 in the first roll paper processing stage 105. Referring back to Figure 5, in various embodiments, block 402 can receive a representation of the first stage anomaly location record stored at location 1244 of the storage memory 1204 from the first anomaly detector 104 via interface 220 of the I / O interface 212, and instruct the tracker processor 200 to store the first stage anomaly location record at location 244 of the storage memory 204. In various embodiments, after processing in the first winding paper processing stage is completed, the winding paper 108 may move to a second winding paper processing stage 107, where it may be processed further. In various embodiments, block 404 of the flowchart shown in Figure 5 may be executed after or during the processing of the winding paper in the second winding paper processing stage 107.

[0077] Referring to Figure 5, in various embodiments, the flowchart 400 proceeds to block 404, which instructs the tracker processor 200 to receive a second-stage anomaly location representation, which represents the location of an anomaly detected on the roll paper 108 in the second roll paper processing step 107 shown in Figure 1. In various embodiments, block 404 may instruct the tracker processor 200 to receive a second-stage anomaly location representation from the second anomaly detector 106 via interface 222 of the I / O interface 212 shown in Figure 2. In various embodiments, the second anomaly detector 106 may be configured substantially similarly to the first anomaly detector 104. Referring to Figure 10, a flowchart 590 is shown that depicts a block of code for instructing the detector processor 1400 shown in Figure 4 to facilitate anomaly detection for roll position tracking in various embodiments. In various embodiments, the block of code included in flowchart 590 may be encoded within a block of code 1406 in the program memory 1402 shown in Figure 4. In various embodiments, flowchart 590 may include a block of code substantially similar to that included in flowchart 500 shown in Figure 6.

[0078] In some embodiments, block 404 can instruct the tracker processor 200 to execute block 406 when the minimum length of the roll of paper 108 is represented by the second-stage abnormal position received in block 404. For example, the minimum length may be about 50 m. In some embodiments, block 404 can instruct the tracker processor 200 to execute block 406 when the minimum number of second-stage abnormal positions is received. For example, in some embodiments, the minimum number of second-stage abnormal positions may be 50.

[0079] Referring to Figure 10, flowchart 590 begins with block 592, which instructs the detector processor 1400 to receive a set of second-stage images of the rolled paper 108 in the second rolled paper processing stage 107. In various embodiments, block 592 may instruct the detector processor 1400 to receive a first set of second-stage images of the rolled paper 108 in the second rolled paper processing stage 107 from one or more second cameras 124 shown in Figure 1, the first set of second-stage images including the second image 600 shown in Figure 11. In various embodiments, block 592 may instruct the detector processor 1400 to receive a machine direction position associated with each of the second-stage images from a second position sensor 126 shown in Figure 1.

[0080] In various embodiments, block 592 can instruct the detector processor 1400 to store a first set of second-stage images at location 1452 in memory memory 1404. In various embodiments, the first set of second-stage images may include recently imaged threshold distance images of the roll of paper 108. For example, in some embodiments, block 592 can instruct the detector processor 1400 to receive the most recent 1000m image of the roll of paper 108 and store it as the first set of second-stage images at location 1452 in memory memory 1404.

[0081] Referring to Figure 10, block 594 instructs the detector processor 1200 to determine at least one of the second-stage anomaly locations based at least in part on the application of a second-stage anomaly detection sensitivity to a set of second-stage images. In various embodiments, block 594 may instruct the detector processor 1400 to determine a first set of second-stage anomaly locations based on the application of a first second-stage anomaly detection sensitivity to a first set of second-stage images. In some embodiments, the first second-stage anomaly detection sensitivity may be stored at location 1454, which may be, for example, 25.

[0082] In various embodiments, blocks 592 and 594 may be executed simultaneously so that, for each second-stage image received in block 592, the image is analyzed for second-stage anomaly locations without waiting for the entire set of second-stage images to be received. In various embodiments, separating the images into sets facilitates the application of different second-stage anomaly recognition sensitivities to the roll of paper 108, with each sensitivity used for a different set of images.

[0083] In various embodiments, the position tracker 102 may be able to request and / or receive a representation of a second-stage anomaly location at any time, and / or a block of code may be executed by the detector processor 1400 to transmit determined second-stage anomaly locations to the position tracker 102 sequentially as they are determined. For example, in some embodiments, block 594 may include a block of code to instruct the detector processor 1400 to send a signal representing a second-stage anomaly location record to the position tracker 102 via interface 1424 of the I / O interface 1412 shown in Figure 4, when a second-stage anomaly location record is determined or updated with a new anomaly location. Thus, in various embodiments, the position tracker 102 may receive second-stage anomaly locations sequentially before a complete set of second-stage images is received in block 592 of the flowchart 590.

[0084] In various embodiments, block 594 can instruct the detector processor 1400 to store a first set of determined second-stage anomaly locations in the second-stage anomaly location record 620, as shown in Figure 12, and to store the second-stage anomaly location record 620 at location 1456 in the storage memory 1404, as shown in Figure 4. Referring to Figure 12, the second-stage anomaly location record 620 includes anomaly location fields for storing the machine direction and cross direction locations for each anomaly. For example, referring to Figure 12, the second-stage anomaly location record 620 may include first second-stage anomaly location fields 622 and 624 for storing the machine direction and cross direction locations of a first anomaly, respectively; second second-stage anomaly location fields 626 and 628 for storing the machine direction and cross direction locations of a second anomaly, respectively; and first-to-first anomaly location fields 632 and 634 for storing the machine direction and cross direction locations of a first-to-first anomaly, respectively.

[0085] Referring to Figure 11, in some embodiments, the flowchart 590 may include blocks 596 and 598 that can instruct the detector processor 1400 to determine a second second-stage anomaly recognition sensitivity for application to a second set of second-stage images. In various embodiments, the second second-stage anomaly recognition sensitivity may differ from the first second-stage anomaly recognition sensitivity. In various embodiments, block 596 can instruct the detector processor 1400 to determine an anomaly density associated with the second-stage anomaly recognition sensitivity. In various embodiments, block 596 can instruct the detector processor 1400 to determine an anomaly density associated with the first second-stage anomaly recognition sensitivity. In some embodiments, the anomaly density may be an anomaly count density. In various embodiments, block 596 can instruct the detector processor 1400 to count how many anomalies are detected within the sampling machine direction proximity of the most recent image included in the first set of second-stage images. For example, in some embodiments, the sampling machine direction proximity may be 10,000 m. In some embodiments, the sampling machine-direction proximity in the second reel processing stage 107 may be smaller than that in the first reel processing stage 105. In various embodiments, this can facilitate improved tracking in the second reel processing stage 107, where reels received from the first reel processing stage 105 may be processed in a different order and / or in reverse order, so that changes that may appear relatively gradual and / or smooth in the first reel processing stage 105, such as a change in paper machine grade, may appear abruptly in the second reel processing stage 107.

[0086] In various embodiments, block 596 can instruct the detector processor 1400 to determine the number of anomalies represented by a first set of second-stage anomaly locations. For example, in some embodiments, block 596 can instruct the detector processor 1400 to read the second-stage anomaly location record 620 from location 1456 in the storage memory 1404 and determine the number of locations represented by the second-stage anomaly location record 620. In various embodiments where the sampling machine direction proximity is greater than the machine direction length represented by the first set of second-stage anomaly positions and position 1456 stores further second-stage anomaly position records, block 596 can instruct the detector processor 1400 to add any further positions represented by the anomaly position records to the number of positions represented by the second-stage anomaly position records 620, so that the positions are within the sampling machine direction proximity of the most recent (largest increasing machine direction position) image of the first set of second-stage images.

[0087] In various embodiments, block 596 can instruct the detector processor 1400 to determine the anomaly density by dividing a number determined by the mechanical direction length represented by the position-considered second-stage image, which in some embodiments may be a location where only a first set of second-stage images has been analyzed, e.g., 1000m. In various embodiments, block 596 can instruct the detector processor 1400 to store the determined anomaly density in location 1458 of the memory 1404. For example, in some embodiments, block 596 can instruct the detector processor 1400 to store an anomaly density of 1.002 anomalies per meter in location 1458 of the memory 1404.

[0088] Referring to Figure 10, in various embodiments, block 598 can instruct the detector processor 1400 to determine a second second-stage anomaly identification sensitivity, at least in part, based on the determined anomaly density and the first second-stage anomaly identification sensitivity. In various embodiments, block 598 may contain code substantially similar to block 508 in the flowchart 500 shown in Figure 6 and / or blocks 582 and 584 in the flowchart 580 shown in Figure 9. In various embodiments, block 598 can instruct the detector processor 1400 to determine the difference between the anomaly density stored in location 1458 of the memory 1404 and a desired second-stage anomaly density stored in location 1460 of the memory 1404. In various embodiments, block 598 can instruct the detector processor 1400 to store the determined difference in location 1462 of the memory 1404.

[0089] In various embodiments, false anomalies may be expected. In some embodiments, false anomalies may appear as dark spots in the camera image, rather than being a characteristic of the roll paper 108. For example, in various embodiments, there may be paper fragments floating in the air between the roll paper 108 and one or more first cameras 120. In some embodiments, the expected rate of false anomalies may be higher in the first roll paper processing stage 105 than in the second roll paper processing stage 107. Therefore, in various embodiments, the desired second-stage anomaly density may be smaller than the desired first-stage anomaly density in order to attempt to detect roughly the same amount of true anomalies (i.e., actual anomalies in the roll paper 108). For example, in some embodiments, the desired second-stage anomaly density may be less than 90% of the desired first-stage anomaly density. In various embodiments, the desired second-stage anomaly density being less than 90% of the desired first-stage anomaly density can facilitate matching and / or tracking despite the low expected rate of false anomalies in the second roll paper processing stage 107. In some embodiments, for example, the desired second-stage anomaly density may be about 0.800 anomalies per meter, and the desired first-stage anomaly density may be about 1.000 anomalies per meter.

[0090] In various embodiments, block 598 is given by the following formula: TIFF0007880968000002.tif1580 The detector processor 1400 can be instructed to determine a second second-stage anomaly threshold using, where T 22 This is the second stage anomaly threshold, T 21 a2 is the first second-stage anomaly threshold, a2 is the multiplier, and D2 is the relative difference determined by dividing the difference stored at position 1462 of the memory 1404 shown in Figure 4 by a desired anomaly density from position 1460 of the memory 1404. In some embodiments, a2 may be set to a small number such as 0.0001, for example, when the first set of second-stage images represents 1000 m in the machine direction length of the roll of paper 108.

[0091] In various embodiments, block 598 can instruct the detector processor 1400 to store a value determined using the above formula as the second second-stage anomaly identification sensitivity at location 1454 in the memory 1404 shown in Figure 4. Referring to Figure 10, in various embodiments, after block 598 is completed, the detector processor 1400 may return to block 592, a second set of second-stage images may be received, blocks 594-598 may be executed on the second set of second-stage images, and the second second-stage anomaly identification sensitivity may be applied to the second set of first-stage images. In various embodiments, further second-stage anomaly location records may be generated and stored at location 1456 in the memory 1204. In various embodiments, blocks 592-598 may be executed repeatedly or continuously during the processing of the paper roll in the second paper roll processing stage 107.

[0092] In various embodiments, block 404 can instruct the tracker processor 200 to receive a representation of a second-stage anomaly location record stored at location 1456 of the memory memory 1404 via interface 222 of the I / O interface 212 shown in Figure 2 from the second anomaly detector 106. In various embodiments, block 404 can instruct the tracker processor 200 to store one or more representations of second-stage anomaly location records representing anomalies at location 256 of the memory memory 204 shown in Figure 2. In some embodiments, block 404 can instruct the tracker processor 200 to update the stored second-stage anomaly location records so that only the most recently detected anomalies are considered. For example, in some embodiments, block 404 can instruct the tracker processor 200 to remove anomaly location records that are not among the most recent 2000 anomaly location records.

[0093] In some embodiments, block 404 can instruct the tracker processor 200 to receive a representation of the second stage anomaly location record 620 shown in Figure 12 from the second anomaly detector 106. In some embodiments, block 404 can instruct the tracker processor 200 to generate an updated or revised second stage anomaly location record 640, as shown in Figure 13, and store it in location 256 of the memory 204 shown in Figure 2. Referring to Figure 13, in various embodiments, block 404 can instruct the tracker processor 200 to include a first increasing machine direction position field 642 for storing increasing machine directions obtained from the position field 622 of the second stage anomaly location record 620 shown in Figure 12, and a first decreasing machine direction position field 644 for storing decreasing machine direction positions which may be based on increasing machine direction positions. In some embodiments, block 404 can instruct the tracker processor 200 to determine a decreasing machine direction position for each increasing machine direction position included in the second stage abnormal position record 620 shown in Figure 12, and to include the determined decreasing machine direction in the second stage abnormal position record 640 shown in Figure 13.

[0094] In various embodiments, block 404 can instruct the tracker processor 200 to determine each decreasing machine direction position by subtracting an associated increasing machine direction position from the total length of the roll of paper 108. In some embodiments, the total length of the roll of paper 108 may be provided in advance and stored in memory 204. In some embodiments, the total length of the roll of paper 108 may be detected or determined by a first position sensor 122 and / or a first anomaly detector 104 shown in Figure 1, and block 404 can receive a representation of the detected total length of the roll of paper 108 and instruct the tracker processor 200 to store the detected total length of the roll of paper 108 in memory 204. For example, in some embodiments, the stored total length of the roll of paper 108 may be 105,336.47 m, and block 404 can instruct the tracker processor 200 to determine each decreasing machine direction position by subtracting an associated increasing machine direction position from 105,336.47 m.

[0095] In various embodiments, the second-stage abnormal position record 640 may include each crossing direction position field, including a crossing direction position field 646 for storing the first crossing direction position obtained from the position field 624 of the second-stage abnormal position record 620 shown in Figure 12. In various embodiments, the positions stored in the decreasing machine direction position field and associated cross direction position field of the second stage abnormal position record 640 shown in Figure 13 can function as second stage abnormal positions representing the location of an abnormality detected on the roll paper 108 in the second roll paper processing stage 107.

[0096] Referring back to Figure 5, in various embodiments, block 404 may be executed continuously while the roll paper 108 is processed in the second roll paper processing step 107 shown in Figure 1. However, in various embodiments, while block 404 is being executed or afterward, the tracker processor 200 may be instructed to execute block 406. In various embodiments, block 406 instructs the tracker processor 200 to consider candidate position offsets as target candidate position offsets. In some embodiments, block 406 may instruct the tracker processor 200 to determine candidate position offsets to be used by utilizing a grid search technique. In some embodiments, block 406 may instruct the tracker processor 200 to establish a first grid of machine direction offsets and cross direction offsets. For example, in some embodiments, machine direction offsets of -100 to 1000 meters may be considered. The grid step size for the machine direction may be, for example, 0.5 m. For the cross direction, offsets of, for example, -0.5 m to +0.5 m may be considered. The grid step size in the direction of interaction may be, for example, 0.1 m. In various embodiments, block 406 can instruct the tracker processor 200 to set a target candidate position offset by selecting a first candidate position offset from a possible offset pair from the first grid.

[0097] In various embodiments, block 406 can instruct the tracker processor 200 to store the target candidate position offset in the target candidate position offset record at location 264 of the memory 204 shown in Figure 2. In various embodiments, block 406 can instruct the tracker processor 200 to store the target candidate position offset record 660 shown in Figure 14 at location 264 of the memory 204. Referring to Figure 14, the target candidate position offset record 660 includes a machine direction offset field 662 and a cross direction offset field 664 for storing the machine direction and cross direction candidate position offsets, respectively.

[0098] Next, block 407 instructs the tracker processor 200 to compare the target candidate position offset with the first and second stage anomaly positions to determine a difference representation associated with the candidate position offset. In some embodiments, the difference representation may represent how well the target candidate position offset can map the position or location of the roll of paper 108 in the first roll processing stage 105 shown in Figure 1 to the same position or location of the roll of paper 108 in the second roll processing stage 107. In some embodiments, the difference representation may represent how well the target candidate position offset can convert the first stage anomaly position to the second stage anomaly position. For example, in some embodiments, block 407 may include code to instruct the tracker processor 200 to apply or add the candidate position offset to the first stage anomaly position to determine the offset first stage anomaly position and to determine the difference between the offset first stage anomaly position and the second stage anomaly position, and thus block 407 may include blocks 408 and 410 of the code shown in Figure 15.

[0099] Referring to Figure 15, block 408 instructs the tracker processor 200 to apply the target candidate position offset to the first stage anomaly position to determine a plurality of offset first stage anomaly positions. In some embodiments, block 408 can instruct the tracker processor 200 to read one or more first stage anomaly position records from position 244 of the memory 204 (including, for example, the first stage anomaly position record 540 shown in Figure 8), and the target candidate position offset record 660 from position 264 of the memory 204 (shown in Figure 14). Block 408 can instruct the tracker processor 200 to add an offset from the machine direction offset field 662 to each of the machine direction position fields included in the first stage anomaly position record. Block 408 can instruct the tracker processor 200 to add an offset from the cross direction offset field 664 to each of the cross direction position fields included in the first stage anomaly position record. Block 408 can instruct the tracker processor 200 to store the result of the first-stage anomaly location record, offset to position 268 (shown at 700 in Figure 16) in the memory 204 shown in Figure 2.

[0100] Referring to Figure 15, block 410 instructs the tracker processor 200 to determine the difference between the offset first-stage anomaly position and the second-stage anomaly position. In some embodiments, block 410 may instruct the tracker processor 200 to read one or more second-stage anomaly position records (including the second-stage anomaly position record 640 shown in Figure 13) from location 256 of the storage memory 204 shown in Figure 2, and to read the offset first-stage anomaly position record (shown at 700 in Figure 16) from location 268 of the storage memory 204 shown in Figure 2.

[0101] In various embodiments, block 410 can instruct the tracker processor 200 to match each position within the second-stage anomaly position record 640 (using the decreasing machine direction position) with a position within the offset first-stage anomaly position record 700 and determine the respective differences between each of the matched positions. In some embodiments, the positions may match by determining the position within the closest offset first-stage anomaly position record 700 for each decreasing machine direction position within the second-stage anomaly position record 640 and using that as the matched position. In various embodiments, identifying the closest offset first-stage anomaly for each second-stage anomaly can be performed efficiently, for example, using a quadtree or FLANN (see Muja, Marius and Lowe, David G. (2019). Fast Approximate Nearest Neighbors with Automatic Algorithm Configuration. In International Conference on Computer Vision Theory and Applications (VISAPP’09)). In some embodiments, block 410 can instruct the tracker processor 200 to determine a weighted squared distance for each pair of matched positions, where the weighted squared distance is as follows TIFF0007880968000003.tif1569 defined as such, where Δ MD is the difference in machine direction values for the matched positions, W CD is the cross-direction weight, and Δ CD is the difference in cross-direction values for the matched positions. In some embodiments, W CD may be set to, for example, 10. In various embodiments, the value used as W CD may be pre-set and stored at the position 270 of the memory, and block 410 can obtain W CDThe tracker processor 200 can be instructed to retrieve the data. In some embodiments, since the crossing direction positions may be measured relatively accurately, it is advantageous to weight the differences in the crossing direction values ​​relatively heavily, and therefore, by giving them relatively heavy weights, improved tracking can be facilitated, for example.

[0102] In various embodiments, block 410 can instruct the tracker processor 200 to store the weighted squared distance in the candidate position offset difference record 720 shown in Figure 17 at position 272 of the memory 204. Referring to Figure 17, in various embodiments, the candidate position offset difference record may include difference fields (e.g., difference fields 722, 724, 726, and 728) for storing the weighted squared distance determined for each of the matched positions. In some embodiments, block 410 can instruct the tracker processor 200 to sort the candidate position offset difference record 720 in ascending order of the values ​​stored in the difference fields.

[0103] Referring back to Figure 5, in various embodiments, block 412 can instruct the tracker processor 200 to associate the difference representation with the candidate position offset. In various embodiments, block 412 can instruct the tracker processor 200 to determine the difference representation by aggregating or summing the smallest differences stored in the candidate position offset difference record 720. In some embodiments, block 412 can instruct the tracker processor 200 to sum, for example, a representative subset of the determined differences. In some embodiments, the representative subset may be a percentage of the difference that is the smallest value. For example, in some embodiments, the percentage may be 10%. In some embodiments, the number of differences included in the representative subset may be the maximum and minimum numbers of the percentage, which may be predetermined and set to values ​​such as 20.

[0104] In various embodiments, the sum can function as a representation of the difference. Block 412 can instruct the tracker processor 200 to update the target candidate position offset record (shown at 660 in Figure 14) in position 264 of the storage memory 204 shown in Figure 2 to include a representative difference field 670 that stores the determined representation of the difference, as shown in Figure 18. In various embodiments, after block 412 is executed, the tracker processor 200 may be instructed to return to block 406 of the flowchart 400 shown in Figure 5, to consider the new candidate position offset as the target candidate position offset, and to repeat blocks 406, 407, and 412 with the new candidate position offset. In various embodiments, block 406 may instruct the tracker processor 200 to identify the new candidate position offset from the first grid described above.

[0105] In various embodiments, after performing blocks 407 and 412 for a new candidate position offset, a further candidate position offset record 740, shown in Figure 19, may be stored at position 264 in the memory 204 shown in Figure 2. In various embodiments, blocks 406, 407, and 412 may be repeated for further candidate position offsets. In some embodiments, once all grid positions in the first grid have been considered, the tracker processor 200 may be instructed to proceed to block 414. In various embodiments, this may result in many candidate position offset records having a format substantially similar to the candidate position offset record 660 shown in Figure 18, which is stored at position 264 of the memory 204 shown in Figure 2. For example, in some embodiments, there may be 22,011 candidate position offset records stored at position 264 of the memory 204.

[0106] Referring to Figure 5, in various embodiments, blocks 406, 407, and 412 are executed and two or more candidate position offset records are stored in position 264 of the memory 204, after which the tracker processor 200 can proceed to block 414. Block 414 instructs the tracker processor 200 to identify the position offset determined from the candidate position offsets, at least in part on the difference representation. In some embodiments, block 414 can instruct the tracker processor 200 to identify a determined position offset by identifying a candidate position offset among a plurality of candidate position offsets associated with the smallest or most minimal representation of the difference. In various embodiments, block 414 can instruct the tracker processor 200 to identify a candidate position offset record having the minimum value stored in a different field from the candidate position offset records recorded at position 264 of the memory 204 shown in Figure 2. In various embodiments, block 414 can instruct the tracker processor 200 to store the identified candidate position offset record as a determined position offset record 780, as shown in Figure 20, and to store the determined position offset record 780 at position 274 of the memory 204 shown in Figure 2.

[0107] Referring to Figure 20, in various embodiments, the determined position offset record 780 may include a machine direction offset field 782, a cross direction offset field 784, and a representative difference field 786. In various embodiments, the machine direction offset stored in the machine direction offset field 782 and the cross direction offset stored in the cross direction offset field 784 may be used to identify the location of an anomaly in the second winding paper processing stage 107 during processing.

[0108] In some embodiments, after block 414 of the flowchart 400 shown in Figure 5 is completed, the tracker processor 200 may be instructed to return to block 406 to consider further candidate position offsets near the determined position offset. In various embodiments, blocks 406, 407, and 412 may be repeated with a second, higher-density grid near the determined position offset, and the newly determined position offset may be found when block 414 is executed again. In some embodiments, an even higher-density grid with a smaller grid step size may be used, and the more accurately determined position offset record 790 may be stored at position 274 in the storage memory 204, as shown in Figure 21.

[0109] In various embodiments, after block 414 is executed, a further representation of the second-stage abnormal position may be received by the position tracker 102, and blocks 404-414 may be repeatedly executed to update the determined position offset record as the roll paper 108 is processed in the second roll paper processing stage 107 shown in Figure 1.

[0110] Referring to Figure 22, various embodiments show a flowchart 820 that depicts blocks of code that may be executed after or during the execution of flowchart 400 shown in Figure 5. In various embodiments, the blocks of code included in flowchart 820 may be included in blocks of code 206 of program memory 202 shown in Figure 2. In some embodiments, flowchart 820 may be executed sequentially as the roll of paper 108 is processed in the second roll of paper processing stage 107 shown in Figure 1.

[0111] Referring to Figure 22, flowchart 820 begins with block 822, which instructs the tracker processor 200 to receive a representation of detected first-stage defect locations, representing the locations of defects detected on the roll paper 108 in the first roll paper processing stage 105. In various embodiments, the representation of detected first-stage defect locations may be in the form of images from one or more first cameras 120, or in the form of a representation of defect location records received from a defect detector.

[0112] In some embodiments, the first anomaly detector 104 can function as a defect detector and can determine the location of the first-stage defect by analyzing the first-stage image stored at location 1240 before transmitting the first-stage defect location to the tracker processor 200. For example, in some embodiments, the first anomaly detector 104 may include a block of code 1208 shown in Figure 3 for instructing the detector processor 1200 to detect a defect. In various embodiments, the block of code 1208 may be substantially similar to the elements included in the flowchart 500 shown in Figure 6, but may include codes with other detection parameters that lead to a higher threshold and / or lower detection sensitivity, so that only defects that are relevant from a product quality standpoint and / or defects that may require adjustment of processing in the second roll paper processing stage 107, or major defects, are identified as defects.

[0113] In various embodiments, block 1208 of the code may instruct the detector processor 1200 to receive the first stage images, including the first stage images 460 shown in Figure 7 and the mechanical orientation position associated with each image, and to determine the location of the defect detected from the images and mechanical orientation position information. In some embodiments, the first stage images may be stored at location 1240 in the storage memory 1204.

[0114] In various embodiments, block 1208 can instruct the detector processor 1200 to apply a defect detection process to the first-stage image to identify defects. For example, in some embodiments, block 1208 can instruct the detector processor 1200 to determine a defect index value for each pixel in a first set of first-stage images and to identify groups of pixels associated with a defect index value greater than a defect threshold. In various embodiments, the defect index value may be determined similarly to an anomaly index value, and the defect threshold may be greater than the threshold used to identify anomalies in block 504 of the flowchart 500 shown in Figure 6, so that fewer defects are identified compared to anomalies. In various embodiments, alternative processes and / or additional criteria may be used to find defects, such as a defect detection process that requires a minimum defect size for a defect to be identified as a defect. In some embodiments, block 1208 can instruct the detector processor 1200 to generate at least one detected defect location record and to transmit a signal representing the detected defect location record to the location tracker 102 via the interface 1224 of the first anomaly detector 104 shown in Figure 3.

[0115] In various embodiments, block 822 can receive a representation of at least one detected first-stage defect location record from the first anomaly detector 104 via interface 220 of the I / O interface 212 shown in Figure 2, and can instruct the tracker processor 200 to store the detected first-stage defect location record at location 276 of the storage memory 204 shown in Figure 2. Referring to Figure 23, an exemplary detected first-stage defect location record 860 that may be stored at location 276 of the storage memory 204 shown in Figure 2 is shown.

[0116] Referring back to Figure 22, block 824 instructs the tracker processor 200 to receive a representation of the detected second stage position of the roll in the second roll processing stage, where the detected second stage position represents the current position of the roll 108 in the second roll processing stage 107 shown in Figure 1. In various embodiments, the detected second stage position can represent the current or near current position of the roll 108 being processed in the second roll processing stage 107. In various embodiments, block 824 may instruct the tracker processor 200 to receive a representation of the detected second stage position from the second position sensor 126 shown in Figure 1 via the interface 226 shown in Figure 2, which may include a value representing the detected second stage position in meters, for example. In various embodiments, block 824 may instruct the tracker processor 200 to convert an increasing machine direction position from the received detected second stage position to a decreasing machine direction position, as described herein with respect to block 404 of the flowchart 400 shown in Figure 5, before storing the detected second stage position in the memory 204 at position 278 shown in Figure 2.

[0117] Referring to Figure 22, block 826 instructs the tracker processor 200 to determine the defect proximity of the detected second-stage location to at least one of the defects, based on the determined position offset, the detected second-stage location of the roll, and one or more detected first-stage defect locations. In some embodiments, the defect proximity may represent how close the location of the roll 108 currently being processed in the second roll processing stage 107 is to a predicted defect in the roll 108.

[0118] In some embodiments, block 826 can instruct the tracker processor 200 to apply the determined position offset to one or more detected first-stage defect locations to determine one or more predicted second-stage defect locations that represent the predicted defect locations for the roll paper 108 in the second roll paper processing step 107. Block 826 can then instruct the tracker processor 200 to compare the one or more predicted second-stage defect locations with the detected second-stage locations.

[0119] In various embodiments, block 826 can instruct the tracker processor 200 to read the determined position offset record 790 shown in Figure 21 from position 274 and the detected first-stage defect position record 860 shown in Figure 23 from position 276 in the storage memory 204 shown in Figure 2, add the value from the machine direction offset field of the determined position offset record 790 to each machine direction position field in the detected first-stage defect position record 860, and in some embodiments, add the value from the cross direction offset field of the determined position offset record 790 to each cross direction position field in the detected first-stage defect position record 860. In various embodiments, block 826 can instruct the tracker processor 200 to store the result as a predicted second-stage defect position record 900 shown in Figure 24 in position 280 of the storage memory 204 shown in Figure 2.

[0120] In various embodiments, block 826 can instruct the tracker processor 200 to identify one of the predicted second-stage defect locations represented by the predicted defect location record 900 shown in Figure 24 as the defect location closest to the detected second-stage location stored at location 278 in the memory 204 shown in Figure 2. In various embodiments, block 826 can instruct the tracker processor 200 to identify the closest defect location as the nearest future defect location by requiring that the identified location has a lower machine direction position than the detected second-stage location. Block 826 can instruct the tracker processor 200 to determine the defect proximity as the difference or distance between the detected second-stage location and the identified closest defect location. In some embodiments, block 826 can instruct the tracker processor 200 to determine the defect proximity based solely on the machine direction position, so block 826 can instruct the tracker processor 200 to determine the defect proximity as the machine direction difference or machine direction distance between the detected second-stage location and the identified closest defect location. In various embodiments, block 826 can instruct the tracker processor 200 to store the determined defect proximity at location 282 in the memory 204 shown in Figure 2.

[0121] Referring back to Figure 22, in various embodiments, block 828 instructs the tracker processor 200 to generate a signal so that the processing in the second winding stage 107 shown in Figure 1 is adjusted if the determined defect proximity meets a threshold criterion. For example, in some embodiments, block 828 can instruct the tracker processor 200 to determine whether the defect proximity meets a threshold criterion by determining whether the defect proximity is greater than a threshold distance, which in some embodiments may be provided or generated and stored in location 284 of the memory 204. In various embodiments, block 828 can instruct the tracker processor 200 to generate a signal so that the processing in the second winding stage 107 is adjusted at each of the predicted defect locations. In various embodiments, block 828 can instruct the tracker processor 200 to send a control signal to the winding driver 130 via interface 224 to slow down the winding driver if the defect proximity is less than a threshold distance. In various embodiments, this can facilitate the deceleration of the roll 108 when the current position of the second roll processing stage enters within a threshold distance before any of the positions identified in the predicted defect location record 900 shown in Figure 24. In some embodiments, the tracker processor 200 can continuously and / or repeatedly execute the codes from blocks 824, 826, and 828 to check the current position of the second roll processing stage, update the predicted defect location record 900 shown in Figure 24 (for example, if the determined position offset is updated), update the defect proximity, and check whether the defect proximity is less than the threshold distance. In some embodiments, the codes from blocks 824-828 may be executed continuously during processing in the second roll processing stage 107.In some embodiments, the threshold distance may be 2500m, and block 828 can instruct the tracker processor 200 to start decelerating the roll paper processing driver 130 when, for example, the current second roll paper processing stage is within 2500m of the next location identified in the predicted defect location record.

[0122] In various embodiments, the threshold distance or proximity may depend on the speed at which the roll of paper 108 is moving. For example, in some embodiments, block 828 can receive the current speed of the roll of paper 108 from the second position sensor 126 and instruct the tracker processor 200 to determine the threshold distance based on the current speed. For example, from the equation of motion, the threshold distance is v 2 It may also be determined as / 2a, where a is the desired reduction rate and v is the current speed at which the roll of paper 108 is moving. In the case of a slitter winding machine, a typical value is, for example, a = 0.5 m / s 2 And v can be 30 m / s. However, in practical applications, the deceleration is zero to 0.5 m / s. 2 It cannot be changed abruptly to or from a certain point; instead, it may need to rise and fall smoothly. Therefore, in some embodiments, two additional terms b+c may be added to the threshold distance. In various embodiments, b and c may depend on a desired (constant) acceleration a and a desired jerk j, and in addition, b may also depend on the initial velocity v and initial acceleration. For example, assuming the initial acceleration is zero, b=v*4s, and c=2m, when deceleration begins from a constant speed of 1800m / min, the threshold distance is b+v 2 / 2a+c=120m+900m+2m=1022m, where 120m corresponds to an increase in deceleration, 900m corresponds to a constant deceleration period, and 2m corresponds to a decrease in deceleration. Therefore, in various embodiments, the faster the roll of paper 108 is moving at the present moment, the more the roll of paper processing driver 130 may need to start decelerating the roll of paper in order to stop the roll of paper 108 in a timely manner. In some embodiments, it may be necessary to calculate a threshold distance when the initial velocity is not constant. In various such embodiments, the threshold distance may depend not only on the current velocity but also on the current acceleration or deceleration of the roll of paper 108. Therefore, in various embodiments, the tracker processor 200 may be configured to estimate the current acceleration or deceleration from, for example, the received current velocity value. In various embodiments, the higher the current acceleration, the more the roll of paper processing driver 130 may need to start decelerating the roll of paper earlier in order to stop the roll of paper 108 in a timely manner. Similarly, if the roll of paper has already decelerated, in various embodiments the threshold distance may be shorter compared to a situation where the initial velocity is constant.

[0123] In various embodiments, a tracker processor 200 that slows down the roll paper processing driver 130 can facilitate the correction and / or removal of defects from the roll paper 108. In some embodiments, the roll paper processing driver 130 shown in Figure 1 can typically operate at speeds of about 2000 m / min or more, and the tracker processor 200 may be configured to slow down the driver to about 5 m / min, allowing for the correction or removal of defects without, for example, completely stopping the roll paper 108. In various such embodiments, block 828 can instruct the tracker processor 200 to add a certain value to a threshold distance such that the target is to reach a low slow speed before a predicted defect location, such as 3 meters before the predicted defect location. In various embodiments, this can facilitate both high-speed processing and the ability to remove or repair defects in the roll paper 108 in a second roll paper processing stage 107.

[0124] In various embodiments, the tracker processor 200 may be configured to generate a signal to the driver 130 to completely stop the movement of the roll of paper 108 whenever a defect is predicted to be located in a position that may be advantageous for an operator or machine to correct or remove the defect. In some embodiments, the tracker processor 200 may be configured to either completely stop the driver 130 or simply slow it down, depending on the class associated with the predicted defect.

[0125] In some embodiments, when a defect is removed from the roll of paper 108, this may involve removing or altering a portion of the roll of paper 108, and therefore the previously considered second-stage image and associated determined position offset may no longer be applicable or accurate. Thus, in various embodiments, the flowchart 820 shown in Figure 22 may include a block of code instructing the tracker processor 200 to update the second-stage anomaly locations stored at position 256 for consideration when determining the position offset after the current roll of paper position indicates that a detected defect location has passed. For example, the block may instruct the tracker processor 200 to update the second-stage anomaly locations stored at position 256 to include only the second-stage anomaly locations representing the position after the detected defect location has passed. In some embodiments, for example, the block may instruct the tracker processor 200 to remove all second-stage anomaly location records from position 256 in the storage memory 204 whenever a predicted defect location is passed from the predicted defect location record 900.

[0126] calibration In various embodiments, the first-stage anomaly identification sensitivity and / or the second-stage anomaly identification sensitivity may be determined and / or updated using alternative and / or additional processes. In some embodiments, the anomaly identification sensitivity may be set by a standalone calibration process rather than during the analysis of previously identified anomalies. In some embodiments, the calibration process may include identifying anomaly locations that should not be used later to determine the positional offset to be determined. In some embodiments, for example, the first stage anomaly detection sensitivity may be initially determined or set using a standalone calibration process which may include analysis of the first set of first stage images itself and / or images of adjacent portions of the roll of paper 108.

[0127] Referring now to Figure 25, a flowchart 940 is shown that depicts blocks of code that may be executed by the detector processor 1200 before the execution of flowchart 500 shown in Figure 6, according to various embodiments. For example, in some embodiments, the blocks included in flowchart 940 may be executed before the execution of block 502 of flowchart 500 shown in Figure 6 in order to determine a first stage anomaly identification sensitivity before flowchart 500 shown in Figure 6 is executed.

[0128] Referring to Figure 24, the flowchart begins at block 942, which instructs the detector processor 1200 to receive a calibration set of first stage images of the roll of paper 108 in the first roll processing step 105 shown in Figure 1. In some embodiments, the first set of first stage images may include at least one of the calibration sets of first stage images. In some embodiments, the first set of first stage images may include a calibration set of first stage images. In some embodiments, the calibration set of first stage images and the first set of first stage images may be the same set of images.

[0129] In various embodiments, using a first set of first-stage images as a calibration set of first-stage images to determine a calibration-based first-stage anomaly recognition sensitivity that functions as a first first-stage anomaly recognition sensitivity can facilitate good calibration so that a desired anomaly density of the first set of first-stage images is achievable. In some embodiments, the calibration set of first-stage images may include images representing at least a portion of the roll paper 108 adjacent to the portion represented by the first set of first-stage images. In some embodiments, block 942 may contain code substantially similar to block 502 in the flowchart 500 shown in Figure 6, and can instruct the detector processor 1200 to receive a calibration set of first-stage images and the mechanical orientation position associated with each image. In various embodiments, block 942 can instruct the detector processor 1200 to store the calibration set of first-stage images and the associated positions at position 1280 in the memory memory 1204 shown in Figure 3.

[0130] Referring to Figure 25, block 944 instructs the detector processor 1200 to determine the first-stage calibration anomaly density, at least in part, based on the application of the first-stage calibration anomaly detection sensitivity to the calibration set of first-stage images. In various embodiments, block 944 may contain code substantially similar to the code contained in blocks 504 and 506 of the flowchart 500 shown in Figure 6. In various embodiments, block 944 can instruct the detector processor 1200 to determine the first-stage calibration set of anomaly locations and to determine the number of anomalies represented by the first-stage calibration set of anomaly locations, at least in part on the application of calibration first-stage anomaly identification sensitivity to the calibration set of first-stage images.

[0131] In some embodiments, the first-stage calibration anomaly identification sensitivity may include a first-stage calibration anomaly identification threshold that is pre-initialized and may be stored at location 1282 in the memory 1204 shown in Figure 3. In some embodiments, for example, the first-stage calibration anomaly identification sensitivity may include an anomaly identification threshold set to 25, and block 944 can instruct the detector processor 1200 to identify any group of pixels having an anomaly index value greater than 25 as representing an anomaly. In various embodiments, block 944 can generate a first-stage calibration anomaly location record 980, as shown in Figure 26, and can instruct the detector processor 1200 to store the first-stage calibration anomaly location record 980 at location 1284 in the memory 1204 shown in Figure 3. In various embodiments, block 944 can instruct the detector processor 1200 to determine the number of anomalies by counting how many locations are stored in the first-stage calibration anomaly location record 980. In various embodiments, block 944 can instruct the detector processor 1200 to determine the first-stage calibration anomaly density by dividing the number of anomalies by the mechanically oriented length of the portion of the roll paper 108 represented by the calibration set of first-stage images. In various embodiments, block 944 can instruct the detector processor 1200 to store the determined first-stage calibration anomaly density in location 1286 of the memory 1204.

[0132] Referring to Figure 25, block 946 instructs the detector processor 1200 to determine a calibration-based first-stage anomaly recognition sensitivity based at least in part on the first-stage calibration anomaly recognition sensitivity and the first-stage calibration anomaly density. In some embodiments, the calibration-based first-stage anomaly recognition sensitivity may be used later as the first first-stage anomaly recognition sensitivity described above with respect to the flowchart 500 shown in Figure 6. In various embodiments, block 946 may contain code substantially similar to block 508 in the flowchart 500 shown in Figure 6 and blocks 582 and 584 in the flowchart 580 shown in Figure 9. In various embodiments, block 946 may determine the difference between the first-stage calibration density and a desired first-stage calibration density and instruct the detector processor 1200 to determine the first-stage calibration anomaly recognition sensitivity based at least in part on the determined difference. In various embodiments, the desired first-stage calibration density may be pre-stored in location 1288 of the memory 1204. In various embodiments, block 946 can instruct the detector processor 1200 to store the determined difference at location 1290 in memory 1204.

[0133] In some embodiments, the calibration-based first-stage anomaly detection sensitivity may include a calibration-based first-stage anomaly threshold. In various embodiments, block 946 is given by the following formula: TIFF0007880968000004.tif1571 The detector processor 1200 can be instructed to determine the calibration-based first-stage anomaly threshold using, where T 11 This is the calibration-based first-stage anomaly threshold, T 1C This is the first-stage calibration anomaly threshold, and a C This is the magnification, D C This is the relative difference determined by dividing the difference stored at position 1290 of the memory 1204 shown in Figure 3 by a desired abnormal density from position 1288 of the memory 1204. In some embodiments, a CFor example, when the calibration set for the first stage image represents a machine-direction length of 1000m of roll paper 108, it may be set to a small number such as 0.0001.

[0134] In various embodiments, block 946 can instruct the detector processor 1200 to store a value determined using the above formula as the first stage anomaly identification sensitivity in location 1242 of the memory 1204 shown in Figure 3. In various embodiments, after block 946 is completed, the detector processor 1200 can execute the flowchart 500 shown in Figure 6. In various embodiments, the second anomaly detector 106 shown in Figure 4 may be configured to execute code substantially similar to the code included in flowchart 940 before executing flowchart 590 shown in Figure 10, thereby determining the initial settings for the first second-stage anomaly identification sensitivity, which are stored in memory memory 1404 at position 1454.

[0135] Various embodiments In some embodiments, the system 100 shown in Figure 1, or a system substantially similar to system 100, may be implemented according to various alternative or additional embodiments. Referring to Figure 27, for example, schematic diagrams of a system 2100 for facilitating roll position tracking according to various embodiments are provided. In various embodiments, system 2100 may include several elements similar to those included in system 100 shown in Figure 1. System 2100 includes a position tracker 2102 communicating with a first anomaly detector 2104 and a second anomaly detector 2106. In some embodiments, the first anomaly detector 2104 and the second anomaly detector 2106 may be configured to detect or sense anomalies in a roll or a portion of a roll 2108 in a first roll processing step 2105 and a second roll processing step 2107, respectively.

[0136] In some embodiments, a first anomaly detector 2104 can communicate with one or more first cameras 2120 and a first position sensor 2122 configured to detect the position or location of a machine-oriented roll of paper associated with each of the images acquired by the one or more first cameras 2120. In some embodiments, a second anomaly detector 2106 can communicate with one or more second cameras 2124 and a second position sensor 2126 configured to detect the position or location of a machine-oriented roll of paper associated with each of the images acquired by the one or more second cameras 2124. In various embodiments, a position tracker 2102 can communicate with a second position sensor 2126 so that the position tracker 2102 may be configured to track the current roll position in a second roll processing stage 2107. In some embodiments, the system 2100 may include a roll processing driver 2130 communicating with the position tracker 2102.

[0137] In various embodiments, the position tracker 2102, the first anomaly detector 2104, the second anomaly detector 2106, the first roll paper processing stage 2105 and the second roll paper processing stage 2107, the roll paper 2108, the cameras 2120 and 2124, as well as the first position sensors 2122 and the second position sensors 2126, and the roll paper processing driver 2130 may include some elements and / or functions substantially similar to some elements included in the position tracker 102, the first anomaly detector 104, the second anomaly detector 106, the first roll paper processing stage 105 and the second roll paper processing stage 107, the roll paper 108, the cameras 120 and 124, as well as the first position sensors 122 and the second position sensors 126, and the roll paper processing driver 130 shown in Figure 1 and described herein.

[0138] Referring to Figures 28, 29, and 30, schematic diagrams of the position tracker 2102, first anomaly detector 2104, and second anomaly detector 2106 of the system 2100 shown in Figure 27 are shown according to various embodiments. In various embodiments, the position tracker 2102, first anomaly detector 2104, and second anomaly detector 2106 shown in Figures 28, 29, and 30 may include elements substantially similar to those included in the position tracker 102, first anomaly detector 104, and second anomaly detector 106 shown in Figures 2, 3, and 4 and described herein.

[0139] Referring to Figure 28, the position tracker 2102 includes a processor circuit including a tracker processor 2200 and program memory 2202, and a storage memory 2204 and an input / output (I / O) interface 2212, all of which communicate with the tracker processor 2200. The I / O interface 212 includes interfaces 2220 and 2222 for communicating with the first anomaly detector 2104 and the second anomaly detector 2106, respectively, as shown in Figure 27. In various embodiments, the I / O interface 2212 may also include interface 2224 for communicating with the paper winding driver 2130, as shown in Figure 27, and interface 2226 for communicating with the second position sensor 2126.

[0140] Processor executable program code for instructing the tracker processor 2200 to perform various functions is stored in program memory 2202. Referring to Figure 2, program memory 2202 includes a block 2206 of code for instructing the position tracker 2102 to facilitate the roll paper position tracking function. The memory 2204 includes a plurality of storage locations, including location 2244 for storing first-stage abnormal location data, location 2256 for storing second-stage abnormal location data, location 2264 for storing candidate location offset data, location 2268 for storing offset first-stage abnormal location data, location 2270 for storing cross-direction weight data, location 2272 for storing determined difference data, location 2274 for storing determined location offset data, location 2276 for storing detected defect location data, location 2278 for storing detected second-stage location data, location 2280 for storing offset and detected second-stage location data, location 2282 for storing defect proximity data, and location 2284 for storing defect threshold data. In various embodiments, the storage locations may be stored in a database within the memory 2204.

[0141] Referring now to Figure 29, schematic diagrams of the first anomaly detector 2104 of the system 2100 shown in Figure 27 are shown according to various embodiments. Referring to Figure 29, the first anomaly detector 2104 includes a processor circuit including a detector processor 3200 and program memory 3202, and a storage memory 3204 and an input / output (I / O) interface 3212, all of which communicate with the detector processor 3200. The I / O interface 3212 includes interfaces 3220 and 3222 for communicating with one or more first cameras 2120 and first position sensors 2122 shown in Figure 27, and interface 1224 for communicating with the position tracker 2102 shown in Figure 27.

[0142] Processor executable program code for instructing the detector processor 3200 to perform various functions is stored in program memory 3202. Referring to Figure 3, program memory 3202 includes a block of code 3206 for instructing the first anomaly detector 2104 to facilitate anomaly detection for the roll position tracking function, and a block of code 3208 for instructing the first anomaly detector 2104 to detect defects. The memory 3204 includes a plurality of storage locations, including location 3240 for storing first-stage image data, location 3242 for storing first-stage anomaly identification sensitivity data, location 3243 for storing first-stage anomaly pixel data, location 3244 for storing first-stage anomaly location data, location 3246 for storing first-stage anomaly density data, location 3248 for storing desired first-stage anomaly density data, location 3250 for storing first-stage anomaly density difference data, location 3280 for storing calibration image data, location 3282 for storing calibration anomaly identification sensitivity data, location 3284 for storing calibration anomaly location data, location 3286 for storing calibration anomaly density data, location 3288 for storing desired calibration anomaly density data, and location 3290 for storing calibration anomaly density difference data. In various embodiments, the storage locations may be stored in a database within the memory 3204.

[0143] Referring now to Figure 30, schematic diagrams of the second anomaly detector 2106 of the system 2100 shown in Figure 27 are shown according to various embodiments. In various embodiments, the second anomaly detector 2106 may include elements substantially similar to those included in the first anomaly detector 2104 shown in Figure 29. Referring to Figure 30, the second anomaly detector 2106 includes a processor circuit including a detector processor 3400 and program memory 3402, and a storage memory 3404 and an input / output (I / O) interface 3412, all of which communicate with the detector processor 3400. The I / O interface 3412 includes interfaces 3420 and 3422 for communicating with one or more second cameras 2124 and second position sensors 2126 shown in Figure 27, and interface 3424 for communicating with the position tracker 2102 shown in Figure 27.

[0144] Processor executable program code for instructing the detector processor 3400 to perform various functions is stored in program memory 3402. Referring to Figure 30, program memory 3402 includes a block 3406 of code for instructing the second anomaly detector 2106 to facilitate anomaly detection for the roll paper position tracking function. The memory 3404 includes a plurality of storage locations, including location 3452 for storing second-stage image data, location 3454 for storing second-stage anomaly identification sensitivity data, location 3455 for storing second-stage anomaly pixel data, location 3456 for storing second-stage anomaly location data, location 3458 for storing second-stage anomaly density data, location 3460 for storing desired second-stage anomaly density data, location 3462 for storing second-stage anomaly density difference data, location 3480 for storing calibration image data, location 3482 for storing calibration anomaly identification sensitivity data, location 3484 for storing calibration anomaly location data, location 3486 for storing calibration anomaly density data, location 3488 for storing desired calibration anomaly density data, and location 3490 for storing calibration anomaly density difference data. In various embodiments, the storage locations may be stored in a database within the memory 3404.

[0145] Referring to Figure 31, a flowchart is shown in Figure 2400 that depicts blocks of code for instructing the tracker processor 2200 of the position tracker 2102 shown in Figure 28 to facilitate roll position tracking in various embodiments. In various embodiments, the blocks of code included in the flowchart 2400 may be encoded within a block of code 2206 in the program memory 2202 of the position tracker 2102 shown in Figure 28. Referring to Figure 31, the flowchart 2400 begins with block 2402, which instructs the processor 2200 to receive a representation of a first-stage anomaly location, which represents the location of an anomaly detected on the roll paper during the first roll paper processing stage. In some embodiments, block 2402 may be substantially similar to block 402 of the flowchart 400 shown in Figure 5.

[0146] Referring to Figure 32, a flowchart 2500 is shown illustrating blocks of code for instructing the detector processor 3200 of the first anomaly detector 2104 shown in Figure 29 to facilitate anomaly detection for roll position tracking in various embodiments. In various embodiments, the blocks of code included in the flowchart 2500 may be encoded within a block of code 3206 in the program memory 3202 of the first anomaly detector 2104 shown in Figure 29. Referring to Figure 32, the flowchart 2500 begins with block 2502, which instructs the detector processor 3200 to receive a set of first-stage images of the roll paper 2108 in the first roll paper processing step 2105. In various embodiments, block 2502 may be substantially similar to block 502 of the flowchart 500 shown in Figure 6.

[0147] In various embodiments, block 2502 can receive a first set of first-stage images from one or more first cameras 2120 of the system 2100 shown in Figure 27 via interface 3220 of the I / O interface 3212 shown in Figure 29, and instruct the detector processor 3200 to store the first set of first-stage images in the memory memory 3204 at location 3240 shown in Figure 29. In various embodiments, block 2502 can receive an associated mechanical orientation position from a first position sensor 2122 shown in Figure 27 for each image included in the first set of first-stage images, and instruct the detector processor 3200 to store the associated mechanical orientation position in relation to each image in the first set of first-stage images stored in the memory memory 3204 at location 3240.

[0148] Referring to Figure 32, block 2504 instructs the detector processor 3200 to determine at least one of the first-stage anomaly locations, at least in part, by applying the first-stage anomaly recognition sensitivity to the set of first-stage images. In various embodiments, block 2504 may instruct the detector processor 1200 to determine a first set of first-stage anomaly locations based on the application of the first-stage anomaly recognition sensitivity to a first set of first-stage images. In various embodiments, the first stage anomaly detection sensitivity may include multiple anomaly detection thresholds, each associated with a different pixel location. In some embodiments, using different anomaly detection thresholds, each associated with a different pixel location, can reduce false identification of anomalies and / or facilitate more sensitive detection of true but relatively weak anomalies, and / or result in a more uniform distribution of detected anomalies across the camera sensor and / or cross directions, which can further facilitate better matching and / or tracking. In some embodiments, the anomaly detection thresholds may be stored at location 3242 of the memory memory 3204 of the first anomaly detector 2104 shown in Figure 29. For example, in some embodiments, the first stage image may include 1600x800 pixels, and the first stage anomaly detection sensitivity may include 1,280,000 anomaly detection thresholds, each associated with a different pixel.

[0149] In some embodiments, the first stage anomaly detection sensitivity may include, for example, a first stage anomaly threshold record 3600 shown in Figure 33, which may be stored at location 3242 in the storage memory 3204 of the first anomaly detector 2104 shown in Figure 29. Referring to Figure 33, the first stage anomaly detection sensitivity includes threshold fields 3602-3612 that store the respective thresholds associated with each respective pixel location. In some embodiments, the anomaly detection thresholds may be initialized to values ​​that can be pre-set and selected by the user. For example, in some embodiments, the thresholds may be set to values ​​between 1 and 100. In some embodiments, the thresholds stored at position 1242 may be initially set to, for example, 25 each.

[0150] In various embodiments, block 2504 can instruct the detector processor 3200 to determine a per-pixel anomaly index value, substantially the same as shown in Figure 6 and described with reference to block 504 of flowchart 500 described herein. In various embodiments, block 2504 can instruct the detector processor 3200 to compare the per-pixel anomaly index value with a first-stage anomaly identification threshold from a first-stage anomaly threshold record 3600 associated with that pixel. Pixels whose anomaly index value exceeds the associated first-stage anomaly identification threshold may be identified as anomaly or considered anomaly. In some embodiments, block 2504 can instruct the detector processor 3200 to consider any pixel within the threshold anomaly distance of a pixel having an anomaly index value exceeding the associated first-stage anomaly identification threshold as anomaly. In some embodiments, where certain parts of an image may occasionally represent a higher anomaly density (for example, due to true anomalies in a product and / or false anomalies caused by lighting conditions), this can help, for example, the detector processor 3200 to better fit the first-stage anomaly threshold record 3600, and as a result, the first-stage anomaly density record 3640 may be more uniform overall across different pixels, or otherwise better suited to the purpose, without excessive delay. For example, in some embodiments, the threshold anomaly distance may be the width of a pixel, and all eight pixels adjacent to a pixel having an anomaly index value exceeding an associated first-stage anomaly identification threshold may be identified as anomaly.

[0151] In various embodiments, block 2504 can instruct the detector processor to store a first-stage abnormal pixel identifier at location 3243 of the storage memory 3203, which stores identification information for each abnormal pixel in the first-stage image stored at location 3240. For example, in some embodiments, each of the first-stage abnormal pixel identifiers may include a binary image, and block 2504 can instruct the detector processor to store a binary image for each of the first-stage images stored at location 3240, the binary image storing non-zero or zero values ​​in relation to each pixel location, where a non-zero value such as 1 or 255 identifies the pixel associated with that pixel location as an abnormal pixel, and a zero value such as 0 identifies the pixel having that pixel location as a non-abnormal pixel.

[0152] In some embodiments, the anomaly detection sensitivity and / or threshold per camera pixel may be automatically adjusted so that the sensitivity of the second anomaly detector 106 can be automatically reduced whenever the line speed of the second paper winding stage 107 is low. In some embodiments, the sensitivity may be reduced in proportion to how far below a pre-configured threshold speed the line speed falls (or remain uncorrected if the line speed does not fall below a pre-configured threshold). In various embodiments, such as paper winding machines in the paper industry, this sensitivity suppression can help prevent false anomaly detection, for example, when unwinding is first initiated. In some embodiments, the sensitivity may similarly decrease for similar reasons when the tension of the paper winding or similar product is low.

[0153] In various embodiments, block 2504 of the flowchart 2500 shown in Figure 32 can instruct the detector processor 3200 to determine anomaly locations based on identified anomaly pixels and their associated locations, substantially similar to what is described herein with respect to block 504 of the flowchart 500 shown in Figure 6. Block 2504 can instruct the detector processor 3200 to store a first set of determined first-stage anomaly locations in the first-stage anomaly location record at location 3244 in the memory memory 3204 shown in Figure 29.

[0154] In some embodiments, flowchart 2500 and / or block 2504 may include a block of code to instruct the detector processor 3200 to transmit a representation of the first-stage anomaly location to the position tracker 2102 shown in Figure 28. For example, in some embodiments, flowchart 2500 may include a block to instruct the detector processor 3200 to transmit a signal representing the first-stage anomaly location record to the position tracker 2102 via interface 3224 of the I / O interface 3212 shown in Figure 29.

[0155] Referring to Figure 32, in some embodiments, the flowchart 2500 may include blocks 2506 and 2508, which can instruct the detector processor 3200 to determine a second first-stage anomaly detection sensitivity for application to a second set of first-stage images. In various embodiments, the second first-stage anomaly detection sensitivity may differ from the first first-stage anomaly detection sensitivity. Referring to Figure 32, block 2506 instructs the detector processor 3200 to determine at least one anomaly density associated with the first-stage anomaly identification sensitivity. In various embodiments, block 2506 can instruct the detector processor 3200 to proceed substantially as described herein for block 506 in the flowchart 500 shown in Figure 6, except that it determines a plurality of anomaly densities, each associated with a different pixel location. In various embodiments, the anomaly density may be an anomaly pixel density. In various embodiments, the anomaly density may be determined per pixel location, for example, as the number of identified anomaly pixels per length of the roll of paper 2108. In various embodiments, block 2506 can instruct the detector processor 1200 to determine each anomaly density by counting how many anomalies have been detected at a pixel location within the sampling machine direction proximity of the most recent image included in a first set of first-stage images. Block 2506 can instruct the detector processor 3200 to determine the anomaly density based on the first-stage anomaly pixel identifier stored in the memory 3204 at location 3243.

[0156] In various embodiments, block 2506 can instruct the detector processor 3200 to store the anomaly density in the first-stage anomaly density record 3640 shown in Figure 34, and block 2506 can instruct the detector processor 3200 to store the first-stage anomaly density record 3640 at location 3246 of the storage memory 3204. Referring to Figure 32, block 2508 instructs the detector processor 3200 to determine a new first-stage anomaly recognition sensitivity, at least partially based on the determined at least one anomaly density and first-stage anomaly recognition sensitivity. In various embodiments, block 2508 may contain code substantially similar to the code contained in block 508 of flowchart 500 shown in Figure 6, except that block 2508 can instruct the detector processor 3200 to determine a plurality of new first-stage anomaly thresholds, each associated with a respective pixel location.

[0157] In various embodiments, block 2508 can instruct the detector processor 3200 to read the first stage anomaly threshold record 3600 shown in Figure 33 and the first stage anomaly density record 3640 shown in Figure 34 from locations 3242 and 3246 of the memory 3204 shown in Figure 29, and to determine the second first stage anomaly threshold record 3680 shown in Figure 35, based at least in part on the first stage anomaly threshold record 3600 and the first stage anomaly density record 3640. In various embodiments, block 2508 can instruct the detector processor 3200 to store the second first stage anomaly threshold record 3680 as the second first stage anomaly discrimination sensitivity at location 3242 of the memory 3204.

[0158] In various embodiments, block 2508 may contain code substantially similar to the block in flowchart 580 shown in Figure 9 for determining each first-stage anomaly threshold included in the second first-stage anomaly threshold record 3680 shown in Figure 35. In various embodiments, block 2508 can instruct the detector processor 3200 to determine a difference for each pixel position by subtracting a desired first-stage anomaly density stored at location 3248 of memory memory 3204, shown in Figure 29, from an anomaly density field value of the first-stage anomaly density record 3640 shown in Figure 34 and stored at location 3248 of memory memory 3204; block 2508 can instruct the detector processor 3200 to store the difference at location 3250 of memory memory 3204.

[0159] In some embodiments, a desired first-stage anomaly density may be pre-set and stored in memory memory 1204 at location 3248. In some embodiments, the desired first-stage anomaly density may be set based on a desired anomaly density relative to the machine direction length, which facilitates roll position tracking using the position tracker 2102. For example, in some embodiments, the desired first-stage anomaly density may be 0.00001 anomaly pixels per meter of machine direction length. In various embodiments, the desired first-stage anomaly density may be pre-set by the user and / or adjusted so that enough anomalies are detected to initiate tracking early, for example, but at the same time so that too many anomalies are detected that are false and / or otherwise undetectable in a second roll processing stage 107, for example.

[0160] In some embodiments, block 2508 is given by the following formula: TIFF0007880968000005.tif1577 The detector processor 3200 can be instructed to determine each of the new first-stage anomaly thresholds that can function as a second first-stage anomaly threshold, where T 12 This is the second first-stage anomaly threshold, T 11 is the first stage anomaly threshold, a1 is the magnification, and D1 is the relative difference determined by dividing the difference determined and stored in the memory memory 3204 at position 3250 shown in Figure 29 by the desired first stage anomaly density from the memory memory 1204 at position 3248. In some embodiments, a1 may be set to a small number such as 0.0001, for example, when the first set of first stage images represents 1000 m in the mechanical direction length of the roll of paper 2108. In various embodiments, block 2508 can instruct the detector processor 3200 to store each of the values ​​determined using the above formula in the first-stage anomaly threshold field of the second first-stage anomaly threshold record 3680 shown in Figure 35, and to store the second first-stage anomaly threshold record 3680 at location 3242 in the storage memory 3204 shown in Figure 29.

[0161] Referring to Figure 32, in various embodiments, after block 2508 is completed, the detector processor 3200 may return to block 2502, where a further set of first-stage images may be received. In various embodiments, block 2504 may include a second first-stage anomaly threshold record 3680, shown in Figure 35, and may be performed by applying a second first-stage anomaly discrimination sensitivity, stored in memory memory 3204 at position 3242, to a second set of first-stage images. In various embodiments, blocks 2502 to 2508 of the flowchart 2500 shown in Figure 32 may be executed for each set of first-stage images representing each portion of the roll of paper 2108. In various embodiments, after the execution of flowchart 2500 is complete, one or more first-stage abnormal position records representing abnormal positions may be stored at position 3244 in memory 3204.

[0162] Referring to Figure 27, in various embodiments, the flowchart 2500 shown in Figure 32 may be executed during the processing of the roll paper 2108 in the first roll paper processing stage 2105. Referring back to Figure 31, in various embodiments, block 2402 can receive a representation of the first stage anomaly location record stored at location 3244 of the storage memory 3204 from the first anomaly detector 2104 via interface 2220 of the I / O interface 2212 shown in Figure 28, and instruct the tracker processor 2200 to store the first stage anomaly location record at location 2244 of the storage memory 2204.

[0163] In various embodiments, after processing in the first roll paper processing stage is completed, the roll paper 2108 may move to a second roll paper processing stage 2107, where it may be processed further. In various embodiments, after or during the processing of the roll paper in the second roll paper processing stage 2107, block 2404 of the flowchart shown in Figure 31 may be executed. Referring to Figure 31, in various embodiments, the flowchart 2400 proceeds to block 2404, which instructs the tracker processor 2200 to receive a second-stage anomaly location representation, which represents the location of anomalies detected on the roll paper 2108 in the second roll paper processing step 2107 shown in Figure 27. In various embodiments, block 2404 may instruct the tracker processor 2200 to receive a second-stage anomaly location representation from the second anomaly detector 2106 via interface 2222 of the I / O interface 2212 shown in Figure 28.

[0164] In various embodiments, the second anomaly detector 2106 may be configured substantially similarly to the first anomaly detector 2104. Referring to Figure 36, a flowchart 2590 is shown that depicts blocks of code 2592, 2594, 2596, and 2598 for instructing the detector processor 3400 shown in Figure 30 to facilitate anomaly detection for roll position tracking in various embodiments. In various embodiments, the blocks of code included in flowchart 2590 may be encoded within block 3406 of code in the program memory 3402 shown in Figure 30. In various embodiments, flowchart 2590 may include blocks of code substantially similar to those included in flowchart 590 shown in Figure 10, except that it is adapted to apply pixel-specific thresholds substantially similar to those described herein with respect to flowchart 2500 shown in Figure 32.

[0165] In some embodiments, the expected rate of false anomalies may be higher in the first winding paper processing step 2105 than in the second winding paper processing step 2107. Therefore, in various embodiments, the desired second-stage anomaly density may be smaller than the desired first-stage anomaly density. For example, in some embodiments, the desired second-stage anomaly density may be less than 90% of the desired first-stage anomaly density. For example, in some embodiments, the desired second-stage anomaly density may be about 0.000008 anomalies per meter, and the desired first-stage anomaly density may be about 0.00001 anomalies per meter.

[0166] In some embodiments, flowchart 2590 may include block 2610 of the code shown in Figure 37 for instructing the detector processor 3400 shown in Figure 30 to determine the anomaly severity for each of a plurality of second-stage anomaly locations. In some embodiments, block 2610 may be included in block 2594 of the code in flowchart 2590 shown in Figure 36, or may be executed concurrently with block 2594 of the code. In various embodiments, the anomaly severity may include an anomaly intensity value, and block 2610 can instruct the detector processor 3400 to determine the intensity value as the sum of the anomaly index values ​​of the pixels that are part of the anomaly. For example, in some embodiments, the intensity value may be determined as the sum of the anomaly index values ​​of the pixels that are part of the anomaly. Alternatively, in some embodiments, the intensity value may be determined as the maximum anomaly index value across the pixels that are part of the anomaly.

[0167] In various embodiments, block 2610 can instruct the detector processor 3400 to generate a second-stage anomaly location severity record 2620 shown in Figure 38 and store it in location 3456 of the memory memory 3404 shown in Figure 30. In various embodiments, the second-stage anomaly location severity record 2620 includes first second-stage anomaly location fields 2622 and 2624 for storing increasing mechanical and cross-directional locations of the first anomaly, respectively, and a first anomaly severity field 2626 for storing the anomaly severity associated with the first anomaly and the second-stage anomaly location fields 2622 and 2624. In various embodiments, the second-stage anomaly location severity record 2620 includes further second-stage anomaly location fields and associated anomaly severity fields.

[0168] Referring back to Figure 31, block 2404 instructs the tracker processor 2200 to receive a representation of the second-stage anomaly location, which represents the location of the anomaly detected on the roll paper 2108 in the second roll paper processing stage 2107 shown in Figure 27. In some embodiments, block 2404 can instruct the tracker processor 2200 to receive the respective anomaly severity associated with each of the second-stage anomaly locations. In some embodiments, block 2404 can receive a representation of the second-stage anomaly location severity record 2620 shown in Figure 38 from the second anomaly detector 2106 via the interface 2222 shown in Figure 28, and instruct the tracker processor 2200 to store the second-stage anomaly location severity record 2620 in the storage memory 2204 shown in Figure 28 at location 2256. In some embodiments, block 2404 can instruct the tracker processor 200 to generate an updated second-stage abnormal position severity record 2640 shown in Figure 39, which includes a decreasing machine direction field based on the second-stage abnormal position severity record 2620 shown in Figure 38, and store it at position 2256, as generally described herein with respect to block 404 of the flowchart 400 shown in Figure 5 and the second-stage abnormal position record 640 shown in Figure 13.

[0169] In some embodiments, a second-stage anomaly location initially received in block 2404 of the flowchart 2400 shown in Figure 31 and stored in the second-stage anomaly location severity record 2640 shown in Figure 39 can function as a candidate second-stage anomaly location, and the flowchart 2400 may include a block of code to instruct the tracker processor 2200 to ignore at least one of the received candidate second-stage anomaly locations. For example, in various embodiments, block 2404 of the flowchart 2400 shown in Figure 31 may include blocks 2662 and 2664 of code shown in Figure 40. In various embodiments, ignoring at least one of the received candidate second-stage anomaly locations can facilitate improvements such as faster and / or more accurate offset determination.

[0170] Referring to Figure 40, in various embodiments, block 2662 can instruct the tracker processor 2200 to receive representations of multiple candidate second-stage anomaly locations. In some embodiments, block 2662 can instruct the tracker processor 2200 to receive a representation of the second-stage anomaly location severity record 2620 shown in Figure 38 from the second anomaly detector 2106 via the interface 2222 shown in Figure 28, generate the second-stage anomaly location severity record 2640 shown in Figure 39, and store it in the memory memory 2204 at location 2256 shown in Figure 28. Continuing to refer to Figure 40, block 2664 instructs the tracker processor 2200 to determine multiple second-stage anomaly locations as a subset of multiple candidate second-stage anomaly locations. In various embodiments, block 2664 may instruct the tracker processor 2200 to rank the multiple candidate second-stage anomaly locations and select a subset as one or more highest-ranked candidate second-stage anomaly locations. In some embodiments, ranking the candidate second-stage anomaly locations and then selecting a subset based on the ranking can facilitate a faster and / or more accurate offset determination later on.

[0171] In some embodiments, block 2664 can instruct the tracker processor 2200 to rank each of a plurality of candidate second-stage anomaly locations based at least partially on the proximity of the second-stage anomaly location to the current position of the roll in the second roll processing stage. In some embodiments, this can facilitate a faster and / or more accurate offset determination, as more recently detected anomaly locations may be more important to consider when determining how much the roll 2108 shown in Figure 27 has been offset in the second roll processing stage 2107 compared to the first roll processing stage 2105. In some embodiments, block 2664 can instruct the tracker processor 2200 to rank each of a plurality of candidate second-stage anomaly locations based at least in part on the severity of the anomaly associated with the candidate second-stage anomaly location. In some embodiments, this can facilitate faster and / or more accurate offset determination, as more important or larger anomalies may be detected more accurately and therefore more important to consider when determining the offset.

[0172] In some embodiments, block 2664 can instruct the tracker processor 2200 to determine an anomaly severity score for each of the candidate second-stage anomaly locations. For example, in some embodiments, block 2664 may instruct the following TIFF0007880968000006.tif1557 The tracker processor 2200 can be instructed to determine the anomaly severity score as follows, where I a This is the abnormal importance score, S ais the anomaly severity associated with the anomaly, and Da is the increasing machine direction position of the second-stage anomaly position (determined, for example, from the second-stage anomaly position severity record, or by subtracting decreasing machine direction positions from the determined total length of the roll paper 108). In various embodiments, a higher increasing machine direction position may correspond to anomalies that are closer to the position of the roll paper 2108 in the second roll paper processing stage.

[0173] In various embodiments, block 2664 can instruct the tracker processor 2200 to rank candidate second-stage anomaly locations according to their anomaly severity scores, and block 2664 can instruct the tracker processor 2200 to generate a new second-stage anomaly location severity record that is substantially similar to the second-stage anomaly location severity record 2640 but has a field containing a subset of candidate second-stage anomaly locations. For example, in some embodiments, block 2664 can instruct the tracker processor 200 to include only a threshold number of candidate second-stage anomaly locations based on their anomaly severity scores. For example, in some embodiments, the top threshold number may be the top 500 anomaly locations ranked based on their anomaly severity scores. In some embodiments, the top threshold number may be 2000. In various embodiments, block 2664 can instruct the tracker processor 2200 to store a determined subset of second-stage anomaly locations in an updated second-stage anomaly location severity record at location 2256 of the memory 2204 shown in Figure 28.

[0174] Referring to Figure 31, in various embodiments, block 2404 may be executed continuously while the roll paper 2108 is processed in the second roll paper processing stage 2107 shown in Figure 27. In some embodiments, blocks 2662 and 2664 may be executed repeatedly, for example, after each representation of the second stage abnormal position has been received. In various embodiments, while block 2404 is being executed or afterward, the tracker processor 2200 may be instructed to test candidate position offsets to determine a position offset. Thus, in various embodiments, while block 2404 is being executed or afterward, the tracker processor 2200 may be instructed to proceed to block 2406. In various embodiments, block 2406 instructs the tracker processor 200 to consider a candidate position offset as the target candidate position offset. In some embodiments, block 2406 may contain several elements that are substantially similar to those contained in block 406 of the flowchart 400 shown in Figure 2.

[0175] In some embodiments, the roll paper 2108 shown in Figure 27 may stretch, shrink, and / or otherwise deform between the first roll paper processing step 2105 and the second roll paper processing step 2107. Alternatively or additionally, the first position sensor 2122 and the second position sensor 2126 may not be properly calibrated relative to each other, and / or other factors may cause perceived stretching, shrinking, and / or deformation between the first roll paper processing step 2105 and the second roll paper processing step 2107. Therefore, in some embodiments, the target candidate position offset may include at least one candidate magnification to offset any true and / or perceived stretching, shrinking, and / or deformation between the first roll paper processing step 2105 and the second roll paper processing step 2107.

[0176] Referring to Figure 42, a target candidate position offset record 2700 that may be stored at position 2264 according to various embodiments is shown. In some embodiments, the target candidate position offset record 2700 includes a machine direction offset field 2702 and a cross direction offset field 2704 for storing the machine direction and cross direction candidate position offsets, respectively. In some embodiments, the target candidate position offset record 2700 includes a machine direction magnification field 2706 for storing a magnification used to compensate for perceived stretching, shrinking, and / or deformation of the roll paper 2108 in the machine direction. In some embodiments, the target candidate position offset record 2700 includes a cross direction magnification field 2708 for storing a magnification used to compensate for perceived stretching, shrinking, and / or deformation of the roll paper 2108 in the cross direction. In various embodiments, the magnification may be used to scale or adjust the machine direction offset and / or cross direction offset depending on the position of the second stage abnormal position to which it is added, before it is added to the second stage abnormal position.

[0177] Next, block 2407 instructs the tracker processor 2200 to compare the target candidate position offset with the first and second stage anomaly positions to determine a representation of the difference associated with the candidate position offset. In some embodiments, block 2407 may include code to instruct the tracker processor 200 to apply the candidate position offset to the second stage anomaly position to determine the offset second stage anomaly position and to determine the difference between the offset second stage anomaly position and the first stage anomaly position, and as a result, block 2407 may include blocks 2408 and 2410 of the code shown in Figure 41. In various embodiments, blocks 2408 and 2410 shown in Figure 41 may be substantially similar to blocks 408 and 410 of the flowchart 400 shown in Figure 15, with some differences, including, for example, that blocks 2408 and 2410 can compensate for scaling and can instruct the tracker processor 2200 to apply the candidate position offset to the second stage anomaly position instead of the first stage anomaly position when checking the candidate offset.

[0178] In various embodiments, it may be advantageous to match anomaly locations by determining the nearest (offset) location in the first-stage anomaly location record for each location in the second-stage anomaly location record, rather than by other avoidance methods, namely determining a corresponding second-stage anomaly for each first-stage anomaly. In some embodiments, this may be because the probability of not finding a corresponding second-stage anomaly for a first-stage anomaly is higher than the reverse (i.e., not finding a corresponding first-stage anomaly for a second-stage anomaly). For example, the dry end of the paper machine in the first winding paper processing stage 2105 may have more paper fragments floating in the air than are generally present in the winder in the second winding paper processing stage 2107. In some embodiments, it may be advantageous to match anomaly locations by using a quadtree or FLANN-based search or similar method to determine the nearest (offset) location in the first-stage anomaly location record for each location in the second-stage anomaly location record, rather than by other avoidance methods.

[0179] Referring to Figure 41, in various embodiments, block 2408 instructs the tracker processor 2200 to apply target candidate position offsets to second-stage anomaly positions to determine a plurality of offset second-stage anomaly positions. In various embodiments, block 2408 can instruct the tracker processor 2200 to determine at least one candidate offset distance adjusted based on the position of the second-stage anomaly for each of a plurality of second-stage anomaly positions, and to add at least one candidate offset distance to the second-stage anomaly position. For example, in some embodiments, block 2408 may instruct the following TIFF0007880968000007.tif1584 As shown above, the tracker processor 2200 can be instructed to determine the machine direction offset distance for each of the second stage abnormal positions, where MD d This is the machine direction offset distance, MD OThis is the machine direction offset from the machine direction offset field 2702 shown in Figure 42, and S f This is the machine direction magnification from the field 2706 shown in Figure 42, MD S This is a decreasing mechanical direction position that functions as a second-stage abnormal position.

[0180] In various embodiments, block 2408 can instruct the tracker processor 2200 to determine the crossing directional offset distance in substantially the same manner. In various embodiments, block 2408 can instruct the tracker processor 2200 to determine the respective machine-direction offset distance and cross-direction offset distance and add them to each of the second-stage abnormal position records stored in the memory memory 2204 at position 2256 shown in Figure 28. In various embodiments, block 2408 can instruct the tracker processor 2200 to store the resulting offset second-stage abnormal position record 2740, as shown in Figure 43, at position 2268 of the memory memory 2204.

[0181] Referring to Figure 41, block 2410 instructs the tracker processor 2200 to determine the difference between the offset second-stage anomaly position and the first-stage anomaly position. In various embodiments, block 2410 may contain code similar to the code contained in block 410 of flowchart 400 shown in Figure 15. In some embodiments, block 2410 can instruct the tracker processor 2200 to determine the respective offset difference for each of the offset second-stage anomaly locations. For example, in some embodiments, block 2410 can instruct the tracker processor 2200 to determine the difference between each offset second-stage anomaly location in the offset second-stage anomaly location record 2740 with respect to a matching location from the first-stage anomaly location record stored at location 2244 in the memory memory 2204 shown in Figure 28. In various embodiments, block 2410 can instruct the tracker processor 2200 to store the results in the candidate position offset difference record 2760 shown in Figure 44 at position 2272 in the memory 2204 shown in Figure 28.

[0182] Referring to Figure 31, in various embodiments, block 2412 can instruct the tracker processor 2200 to associate the difference representation with the candidate position offset. In some embodiments, block 2412 can instruct the tracker processor 2200 to determine the difference representation using a cost function. In some embodiments, block 2412 can instruct the tracker processor 2200 to determine each weighted offset difference based on the offset difference and the anomaly weight associated with the offset difference. In some embodiments, block 2412 can instruct the tracker processor 2200 to determine each anomaly weight based on the severity of the anomaly for which the associated offset difference was determined. In some embodiments, this can facilitate a more accurate and / or consistent identification of the determined positional offsets by highlighting more important or critical anomaly matches.

[0183] In some embodiments, block 2412 can instruct the tracker processor 2200 to use the anomaly severity associated with the second-stage anomaly location for which the offset difference has been determined as the anomaly weight. In various embodiments, block 2412 can instruct the tracker processor 2200 to generate the weighted candidate location offset difference record 2780 shown in Figure 45 by multiplying each offset difference from the candidate location offset difference record 2760 by the anomaly severity associated with the second-stage anomaly location for which the offset difference has been determined. In various embodiments, block 2412 can instruct the tracker processor 2200 to store the weighted candidate position offset difference record 2780 at position 2272 of the memory 2204.

[0184] In various embodiments, block 2412 can instruct the tracker processor 2200 to identify a representative subset of weighted candidate position offset differences from the weighted candidate position offset difference record 2780 shown in Figure 45, and to determine a representation of the differences based on the representative subset. In some embodiments, this can facilitate the ignoring of differences arising from second-stage anomalies that do not have a corresponding match in the first-stage anomaly, which can result in a better representation of the differences for ranking candidate position offsets.

[0185] In various embodiments, block 2412 can instruct the tracker processor 2200 to identify a representative subset of weighted candidate position offset differences based on unweighted differences from which weighted differences are derived. For example, in some embodiments, block 2412 can instruct the tracker processor 2200 to identify a representative subset using the candidate position offset difference record 2760. In some embodiments, block 2412 can instruct the tracker processor 2200 to identify a representative subset as a weighted difference derived from one of the smallest unweighted differences stored in the candidate position offset difference record 2760 shown in Figure 44. In some embodiments, the representative subset may be a percentage of the determined weighted differences. For example, in some embodiments, the percentage may be 10%. In some embodiments, the number of differences included in the representative subset may be the maximum and minimum numbers of the percentage, which may be predetermined and set to values ​​such as 20.

[0186] In some embodiments, block 2412 can instruct the tracker processor 2200 to scale a subset of the total weighted candidate position offset differences. In various embodiments, this can facilitate the comparison of the determined total subsets for different candidate position offsets. In various embodiments, this can further facilitate the comparison of different candidate position offsets. For example, in some embodiments, block 2412 can instruct the tracker processor 2200 to divide the sum of a representative subset of weighted candidate position offset differences by the total or summed weight of the second-stage anomalies included in the representative subset. In various embodiments, this may reduce the preferred solution of ignoring the most significant second-stage anomalies, which may be the most useful landmarks in the roll paper 2108. In some embodiments, block 2412 can instruct the tracker processor 2200 to multiply the result or quotient by the total weight of all considered second-stage anomalies. In various embodiments, the resulting value may act as a representation of the difference.

[0187] In various embodiments, block 2412 can instruct the tracker processor 2200 to update the target candidate position offset record 2700 in position 2264 of the storage memory 2204 shown in Figure 28, to include a representative difference field 2710 that stores a determined representation of the difference, as shown in Figure 46. In various embodiments, after block 2412 is executed, the tracker processor 2200 may be instructed to return to block 2406 of the flowchart 2400 shown in Figure 31 to consider the new candidate position offset as the target candidate position offset and to repeat blocks 2406, 2407, and 2412 with the new candidate position offset. In various embodiments, block 2406 may instruct the tracker processor 2200 to identify the new candidate position offset from a multidimensional grid containing different values ​​of MD offset, CD offset, MD multiplier, and CD multiplier. In some embodiments, the MD offset and CD offset may be as described herein with respect to the flowchart 400 shown in Figure 5. In some embodiments, MD multipliers of 0.996 to 1.002 may be considered. The grid step size for the MD multiplier may be, for example, 0.002. Similarly, CD multipliers of 0.98 to 1.02 may be considered. The grid step size for the CD multiplier may be, for example, 0.02.

[0188] In some embodiments, the magnification considered may not be centered around 1.000, as average stretching or shrinking may be expected between the first roll processing step 2105 and the second roll processing step 2107 shown in Figure 27. Alternatively or additionally, particularly in the case of MD magnification, it may be expected that the first position sensor 2122 and the second position sensor 2126 may not be properly calibrated relative to each other, but instead, one sensor may be expected to measure relatively longer values ​​on average for the same physical distance than the other.

[0189] In various embodiments, blocks 2406, 2407, and 2412 of the flowchart 2400 shown in Figure 31 may be repeated for further candidate position offsets. In some embodiments, once all grid positions in the first grid have been considered, the tracker processor 2200 may be instructed to proceed to block 2414. In some embodiments, after the first grid has been considered, blocks 2406, 2407, and 2412 are, TIFF0007880968000008.tif1529 This may be repeated for at least one further grid that can be offset from the first grid by different offsets that may be in the range of g d is the grid step size within the corresponding search dimension d. In various embodiments, this can effectively facilitate continuous search operations using different offset grids, which can help asymptotically cover the entire search space.

[0190] In various embodiments, the first grid may not be considered at all, but instead blocks 2406, 2407, and 2412 within each search dimension d TIFF0007880968000009.tif1529 This may only be performed for at least one further grid that can be offset from the first grid by different offsets that may be within a certain range.

[0191] In some embodiments, the offset within each dimension d between the first grid and each further grid is generated separately for each dimension d of each grid to be applied. TIFF0007880968000010.tif1529 It can be a random value. In various embodiments, the offset may be generated deterministically. For example, in some embodiments, the search range may be generated in a manner similar to how a quadtree may be constructed. TIFF0007880968000011.tif2031 In some cases, a continuous offset may be selected from the recursive subdivision.

[0192] In various embodiments where different offset grids can be used for each search operation, the best (=lowest cost) solution (e.g., a determined position offset record with the lowest value in the representative difference field) may be maintained between different searches, and as a result, if a better solution with a lower value in the representative difference field is not found as a result of a new search, the previously found best solution may continue to be used. In various embodiments, repeated execution of blocks 2406, 2407, and 2412 can result in many candidate position offset records having a format substantially similar to the candidate position offset record 2700 shown in Figure 46, which is stored at position 2264 of the memory 2204 shown in Figure 28.

[0193] Referring to Figure 31, in various embodiments, blocks 2406, 2407, and 2412 are executed and two or more candidate position offset records are stored in position 2264 of the storage memory 2204, after which the tracker processor 2200 can proceed to block 2414. Block 2414 instructs the tracker processor 2200 to identify the position offset determined from the candidate position offsets, at least partially based on the difference representation. In some embodiments, block 2414 may be substantially similar to block 414 in the flowchart 400 shown in Figure 5, and can instruct the tracker processor 2200 to identify a position offset determined by identifying a candidate position offset among a plurality of candidate position offsets associated with the smallest or most minute of the difference representations.

[0194] In various embodiments, block 2414 can instruct the tracker processor 2200 to store a candidate position offset record, identified as the determined position offset record 2820 shown in Figure 47, at position 2274 in the storage memory 2204 shown in Figure 28. In some embodiments, after block 2414 of the flowchart 2400 shown in Figure 31 is completed, the tracker processor 2200 may be instructed to return to block 2406 and consider further candidate position offsets near the determined position offset as described herein with respect to the flowchart 400 shown in Figure 5.

[0195] Referring to Figure 48, in various embodiments, a flowchart 3820 is shown that depicts a block of code that may be executed after or during the execution of flowchart 2400 shown in Figure 31. In various embodiments, the block of code included in flowchart 3820 may include several elements that are substantially the same as those included in flowchart 820 shown in Figure 22. In various embodiments, the block of code shown in Figure 48 may include a block of code 2206 in program memory 2202 shown in Figure 28. In some embodiments, the blocks may be executed sequentially as the roll of paper 2108 is processed in the second roll of paper processing stage 2107 shown in Figure 1.

[0196] Referring to Figure 48, flowchart 3820 begins with block 3822, which instructs the tracker processor 2200 to receive a representation of detected first-stage defect locations, representing the location of defects detected on the roll paper 2108 in the first roll paper processing step 2105. In various embodiments, block 3822 may be substantially the same as block 822 of flowchart 820 shown in Figure 22, which can receive a representation of at least one detected first-stage defect location record from the first anomaly detector 2104 via interface 2220 of the I / O interface 2212 shown in Figure 28, and instruct the tracker processor 2200 to store the detected first-stage defect location record at location 2276 of the storage memory 2204 shown in Figure 28. In some embodiments, the first anomaly detector 2104 can function as a defect detector and, for example, analyze a first-stage image stored at position 3240 to determine the defect location before transmitting a representation of the defect location to the tracker processor 2200 via the interface 3224 of the first anomaly detector 2104 shown in Figure 29.

[0197] Referring back to Figure 48, block 3824 instructs the tracker processor 2200 to receive a representation of the detected second-stage position of the roll in the second roll processing stage 2107, where the detected second-stage position represents the current position of the roll 2108 in the second roll processing stage 2107, as shown in Figure 27. In various embodiments, block 3824 may be substantially similar to block 824, which is described herein and shown in Figure 22. In various embodiments, block 3824 may instruct the tracker processor 200 to receive a representation of the detected second-stage position from the second position sensor 2126 shown in Figure 27 via the interface 2226 shown in Figure 28. In various embodiments, block 3824 may instruct the tracker processor 2200 to convert increasing machine direction positions from the received detected second-stage position to decreasing machine direction positions before storing the detected second-stage position in the memory 2204 at position 2278 shown in Figure 2.

[0198] Referring to Figure 48, block 3826 instructs the tracker processor 2200 to determine the defect proximity of the detected second-stage position to at least one of the defects, based on the determined position offset, the detected second-stage position of the roll, and one or more detected first-stage defect positions. In some embodiments, the defect proximity may represent how close the position of the roll 2108 currently being processed in the second roll processing stage 2107 is to a predicted defect in the roll 2108.

[0199] In some embodiments, block 3826 instructs the tracker processor 200 to apply the determined position offset to the detected second-stage position to determine the offset detected position, and to determine the defect proximity by comparing one or more detected first-stage defect positions to the offset detected position. In various embodiments, block 3826 may instruct the tracker processor 2200 to read the determined position offset record 2820 shown in Figure 47 from position 2274 in the memory 2204 shown in Figure 28, and to determine the machine direction offset distance as described herein with respect to block 2408 of the flowchart 2400 shown in Figure 41. In various embodiments, block 3828 may instruct the tracker processor 2200 to add or apply the determined machine direction offset distance to the detected second-stage position from position 2278 in the memory 2204. In various embodiments, block 3826 can instruct the tracker processor 2200 to store the result as an offset detection position at position 2280 of the memory 2204 shown in Figure 28.

[0200] In various embodiments, block 3826 can instruct the tracker processor 2200 to identify one of the first-stage defect locations stored at location 2276 of the memory 2204 shown in Figure 28 as the defect location closest to an offset detection location stored at location 2280 of the memory 2204 shown in Figure 2. In various embodiments, block 3826 can instruct the tracker processor 2200 to identify the closest defect location as the nearest future defect location by requiring that the identified location has a lower machine direction position than the offset detection location. Block 3826 can instruct the tracker processor 2200 to determine the defect proximity as the difference or distance between the offset detection location and the identified nearest first-stage defect location. In some embodiments, block 3826 can instruct the tracker processor 2200 to determine the defect proximity as the machine direction difference or machine direction distance between the offset detection location and the identified nearest first-stage defect location. In various embodiments, block 3826 can instruct the tracker processor 2200 to store the determined defect proximity at location 2282 in the memory 2204 shown in Figure 28.

[0201] Referring back to Figure 48, in various embodiments, block 3828 instructs the tracker processor 2200 to generate a signal so that the processing in the second roll handling stage 2107 shown in Figure 27 is adjusted if the determined defect proximity meets a threshold criterion. In various embodiments, block 3828 may be substantially similar to block 828, described herein and shown in Figure 22. In some embodiments, the threshold distance stored in location 2284 of the memory 2204 may be 2500m, and block 3828 may instruct the tracker processor 2200 to generate a signal via interface 2224 shown in Figure 28 to cause the roll handling driver 2130 shown in Figure 27 to begin deceleration when, for example, the defect proximity is less than 2500m and the current second roll handling stage is within 2500m of the next defect location identified in the first stage defect location record stored in location 2276.

[0202] In various embodiments, the threshold distance may depend on various factors, including, for example, the speed at which the roll of paper 2108 is moving, as described herein with respect to block 828 of the flowchart 800 shown in Figure 22. In various embodiments, the first anomaly detector 2104 and the second anomaly detector 2106 are configured to perform a calibration substantially similar to the calibration described with respect to flowchart 940 shown in Figure 25, but may be modified to use any or all of the elements / principles described herein with respect to flowcharts 2500 and / or 2590 shown in Figures 32 and 36. For example, in some embodiments, flowchart 940 shown in Figure 25 may include code for applying a pixel-specific anomaly identification threshold, each associated with each pixel location, as described herein with respect to flowchart 2500.

[0203] In various embodiments, a system functioning substantially similarly to system 100 shown in Figure 1 and / or system 2100 shown in Figure 27 may include a single device combining the functions of a position tracker, a first anomaly detector, and / or a second anomaly detector. In various embodiments, a system functioning substantially similarly to system 100 shown in Figure 1 and / or system 2100 shown in Figure 27 may include a number of separate devices configured to perform different portions of the functions described herein performed by one of the position tracker, the first anomaly detector, and / or the second anomaly detector.

[0204] In various embodiments, systems substantially similar to system 100 shown in Figure 1 and / or system 2100 shown in Figure 27 may include one or more defect detectors configured to perform defect detection and / or defect location determination in a first roll paper processing stage, separate from the first anomaly detector. For example, in some embodiments, the defect detector may be configured to detect the defect location and transmit a representation of the defect location record to a location tracker, which is configured to perform blocks similar to those included in the flowchart 820 shown in Figure 22. In some embodiments, it may be beneficial to combine functions required for both anomaly detection and defect detection, such as image preprocessing.

[0205] In various embodiments, the difference between the matched positions determined in block 410 of Figure 15 and / or block 2410 of Figure 41, and / or their representations determined in block 412 or block 2412, may be determined using alternative differences and / or representations of the differences, for example, by using an alternative cost function. In various embodiments, machine-direction offsets and cross-direction offsets that are not expected to be normal may incur relatively significant penalties. For example, in some embodiments, it may be possible, though unlikely, that a very large amount of paper is removed from the surface of the reel before rewinding begins (possibly by another winding machine). In some embodiments, it may be beneficial to make such a situation detectable, but only if there is relatively more evidence (than is needed to identify smaller offsets).

[0206] In various embodiments, the flowchart 400 shown in Figure 5 and / or the flowchart 2400 shown in Figure 31 may include a block of code to instruct the tracker processor 200 or 2200 to set limits so that the determined position offset may be considered invalid if the determined position offset is outside the limits (e.g., one of the discovered machine direction offset or cross direction offset or magnification is not physically realistic). In various embodiments, the block of code may instruct the tracker processor 200 or 2200 to generate a warning that is sent to the user and / or the roll handling driver to indicate that synchronization may be lost. In various embodiments, the roll handling driver may be configured not to adjust the processing in such a situation, for example. In some embodiments, accordingly, the roll handling driver may stop the automatic processing and instead prompt and / or require the user to manually engage and control a second roll handling stage, for example, until a valid determined position offset is found again.

[0207] In some embodiments, after completing block 414 of flowchart 400 shown in Figure 5 and / or block 2414 of flowchart 2400 shown in Figure 31, the tracker processor 200 or 2200 may be instructed to return to block 406 or 2406 to consider further candidate position offsets near the determined position offset. In various embodiments, blocks 406, 407, and 412, and / or blocks 2406, 2407, and 2412 may be repeated with further, higher-density grids near the determined position offset, and a new, determined position offset may be found when blocks 414 and / or 2414 are run again. In some embodiments, further, higher-density grids may be used.

[0208] In some embodiments, block 2664 shown in Figure 40 can instruct the tracker processor 2200 to divide the entire rewind length seen so far into M equally spaced (machine-direction) regions or lengths, and then select an equal number of anomalies from each region or length for a subset. In some embodiments, when the desired subset size is not divisible by M, the number of anomalies from each region may not be equal, but instead, there may be one more anomaly from a region where the next anomaly considered for inclusion is relatively severe. In some embodiments, if there are no anomalies in a region, a corresponding number of anomalies may be selected from other regions. For example, if a 2-kilometer roll of paper 2108 has been rewound so far and the desired number of regions is 20, the 2 kilometers may be evenly divided into 100-meter sections. For example, if it is desirable to select 500 anomalies, 25 of the most severe anomalies from each 100-meter section may be selected for the subset.

[0209] In some embodiments, it may be beneficial to similarly divide the observed full width into K equally spaced (horizontal) regions and then proceed in substantially the same manner as outlined above for M equally spaced (vertical) regions. This can be useful if there is a possibility that an unusually long "streak" is occurring at a particular cross-directional position. In some embodiments, the entire region may be divided into MK regions, and as a result, the effects (on the matching of the anomaly position map) caused by both anomalous bursts and streaks can be mitigated. In some embodiments, after block 414 of flowchart 400 shown in FIG. 5 and / or block 2414 of flowchart 2400 shown in FIG. 31 is completed, the tracker processor 200 or 2200 may be instructed to calculate the gradient of a cost function or a difference function with respect to the parameters at the determined position offset. In various embodiments, the tracker processor 200 or 2200 may be instructed to perform a gradient descent step, i.e., to find a newly determined position offset as the vector sum of the determined position offset minus the gradient multiplied by a small scalar constant such as, for example, 0.01. In various embodiments, the tracker processor may then be instructed to calculate the gradient of the cost function at the newly determined position offset and perform a new gradient descent step. In various embodiments, the gradient descent step may be repeated, for example, for a pre-calibrated number of steps and / or until some convergence criterion is met. In some embodiments, the gradient descent step may be repeated, for example, for 100 steps.

[0210] In various embodiments, instead of or in addition to using the gradient descent method described above to find the newly determined position offset, the tracker processor 200 or 2200 may be configured to perform the optimization using alternative and / or more complex methods. For example, in various embodiments, the optimization method can use second-order information regarding the cost function. In some embodiments, the tracker processor 200 or 2200 may be configured to establish the newly determined position offset, for example, by using the Broyden-Fletcher-Goldfarb-Shanno optimization algorithm to find the minimum value of the cost function.

[0211] In various embodiments, for example, various combinations of the above methods of selecting candidate position offsets may be used, such as when a first process is used to obtain a first determined position offset and a further process is applied near the first determined position offset to approach a better determined position offset. In various embodiments disclosed herein, the first and second anomaly detectors can receive the machine direction position or position information associated with each of the first-stage image and the second-stage image, and the received machine direction position may be used to determine the anomaly position for each image, but in some embodiments, the machine direction position or position information for each anomaly may be derived from the image itself. In some embodiments, when the position of the anomaly can be derived from an image of the web, the dedicated position sensor from the system 100 or 2100 shown in FIGS. 1 and 27 may be omitted.

[0212] In some embodiments, the first-stage anomaly detection sensitivity and / or the second-stage anomaly detection sensitivity may be maintained unchanged. Accordingly, in various embodiments, blocks 506 and 508 of the flowchart 500 shown in Figure 6 may be omitted, and / or blocks 596 and 598 of the flowchart 590 shown in Figure 10 may be omitted. In various embodiments, blocks 2506 and 2508 of the flowchart 2500 shown in Figure 32 may be omitted, and / or blocks 2596 and 2598 of the flowchart 2590 shown in Figure 36 may be omitted.

[0213] In various embodiments, the first-stage anomaly position and the second-stage anomaly position may be interchangeably switched and / or processed. For example, in some embodiments, block 408 of the flowchart 400 shown in Figure 5 may instruct the tracker processor 200 to determine an offset second-stage anomaly position by applying a candidate position offset to the second-stage anomaly position, and block 410 may instruct the tracker processor 200 to determine the difference between the offset second-stage anomaly position and the first-stage anomaly position. In various embodiments, similar switching may occur in the system 2100 shown in Figure 27. In various embodiments, blocks 404 and 406, 407, 412, and 414 of the flowchart 400 shown in Figure 5, or blocks 2404 and 2406, 2407, 2412, and 2414 of the flowchart 2400 shown in Figure 31, may be executed sequentially as the roll of paper is processed and further second-stage abnormal positions are received, resulting in the sequential addition and / or updating of second-stage abnormal position records stored at positions 256 or 2256 in memory 204 or 2204, and the sequential updating of result data including determined position offset records.

[0214] In various embodiments, position trackers 102 or 2102, or trackers substantially similar to position trackers 102 or 2102, may be included in systems for facilitating roll tracking at locations other than paper mills where paper products are manufactured. For example, in various embodiments, position trackers 102 or 2102 can facilitate tracking of liquid packaging paper production between a paperboard machine, one or more winding machines, and / or one or more unwinding machines. In various embodiments, position trackers 102 or 2102 can facilitate tracking of tissue paper products between a tissue paper machine and one or more conversion lines. In some embodiments, position trackers 102 or 2102 can facilitate tracking of steel coils or other metal coils between different processing stages. In various embodiments, the rolls 108 and / or 2108 may include paper rolls, cardboard rolls, tissue paper rolls, laminated rolls, fiberglass mats, metal strips, and / or other products, which may contain minor defects and / or other landmarks that may be identifiable as anomalies in order to facilitate the matching and / or tracing described herein.

[0215] In various embodiments, the machine direction offset stored in the machine direction offset field of the determined position offset record may be used without considering the cross direction to determine the predicted defect location in a second winding paper processing stage during processing. In various embodiments, this can reduce computational requirements and still provide satisfactory results. In various such embodiments, the candidate position offset may include a machine direction magnification but not a cross direction magnification. In some embodiments, the system 100 shown in Figure 1 and / or the system 2100 shown in Figure 27 may include additional position sensors for detecting the position of the roll of paper in or near the second roll processing stage. In various embodiments, block 826 of the flowchart 820 shown in Figure 22 may instruct the tracker processor 200 to receive the current position of the roll of paper 108 from the additional position sensors.

[0216] In some embodiments, block 504 of flowchart 500 shown in Figure 6 or block 2504 of flowchart 2500 shown in Figure 32 can instruct the detector processor 1200 or 3200 to calculate the difference as a new pixel value-reference pixel value. In various embodiments, the absolute value of the determined difference may be used as an anomaly index value. In various embodiments, instead of anomaly density based on the number of anomalies, block 506 of the flowchart 500 shown in Figure 6 can instruct the detector processor 1200 to determine an anomaly pixel density associated with a first-stage anomaly recognition sensitivity. In various embodiments, block 506 of the flowchart 500 shown in Figure 6 can instruct the detector processor 1200 to determine at least one anomaly density by determining the number of anomalies represented by a first set of multiple first-stage anomaly locations. In various embodiments, block 506 can instruct the detector processor 1200 to determine anomaly pixel density as anomaly pixels per length of the roll of paper 108 or as anomaly pixels per area of ​​the roll of paper. In various such embodiments, block 508 can instruct the detector processor 1200 to adjust the anomaly recognition sensitivity so that the anomaly recognition threshold is more sensitive when fewer anomaly pixels than desired are represented by the anomaly pixel density, or so that the anomaly recognition sensitivity is less sensitive (and therefore, for example, a higher threshold) when more anomaly pixels than desired are represented by the anomaly pixel density. For example, in some embodiments, the desired anomalous pixel density may be 0.00001 pixels / meter * 1600 * 800 = 12.8 pixels / meter. In various embodiments, block 596 can instruct the detector processor 1400 to determine the anomalous pixel density associated with the first second-stage anomalous recognition sensitivity in substantially the same manner as described above for the first first-stage anomalous recognition sensitivity.

[0217] In various embodiments, it may be possible to efficiently calculate, retrospectively, a new sensitivity or threshold that should give the desired anomaly density. In various embodiments, the anomaly identification sensitivity can be updated accordingly, for example, using exponential smoothing, i.e., FESSA. In various embodiments, the principles and / or features or elements of systems 100 and 2100 shown in Figures 1 and 27 are described herein for illustrative purposes only and may be interchangeable between systems. In various embodiments, any or all of the principles and / or features or elements of system 2100 may be incorporated into system 100, and vice versa.

[0218] In various embodiments, flowchart 940 shown in Figure 25 or code substantially similar to flowchart 940 shown in Figure 25 may be modified to incorporate the functions described herein with respect to flowchart 500 shown in Figure 6, flowchart 590 shown in Figure 10, flowchart 2500 shown in Figure 32, and / or flowchart 2590 shown in Figure 36. In various embodiments, flowchart 940 shown in Figure 25 or code substantially similar to flowchart 940 shown in Figure 25 may be modified in any or all of the methods described herein with respect to flowchart 500 shown in Figure 6, flowchart 590 shown in Figure 10, flowchart 2500 shown in Figure 32, and / or flowchart 2590 shown in Figure 36.

[0219] In some embodiments, one or all of the cameras may include one or more line-scan cameras. In various such embodiments, the corresponding image may include, for example, 2048 x 1 pixels. While certain embodiments of this disclosure are described and illustrated, such embodiments are merely illustrative and should not be considered to limit this disclosure to the extent that it should be interpreted in accordance with the appended claims.

Claims

1. A method for facilitating roll paper position tracking, performed by a processor, In the first winding paper processing stage, multiple representations of first-stage abnormal positions representing the location of an abnormality detected on the winding paper are received, In the second winding paper processing stage, a plurality of representations of second-stage abnormal positions representing the location of an abnormality detected on the winding paper are received, For each of the multiple different candidate position offsets, The candidate position offset is compared with the plurality of first-stage abnormal positions and the plurality of second-stage abnormal positions to determine the representation of the difference associated with the candidate position offset. The representation of the difference is associated with the candidate position offset, Identifying a position offset determined from the candidate position offset based at least partially on the representation of the difference. Methods that include...

2. Comparing the candidate position offset with the plurality of first-stage abnormal positions and the plurality of second-stage abnormal positions, The candidate position offset is applied to the plurality of first-stage abnormal positions to determine the plurality of offset first-stage abnormal positions, Determining the difference between the plurality of offset first-stage abnormal positions and the plurality of second-stage abnormal positions. The method according to claim 1, including the method described in claim 1.

3. The method according to claim 2, wherein determining the difference includes determining the respective offset difference for each of the plurality of offset first-stage abnormal positions.

4. The method according to claim 2, wherein applying the candidate position offset includes determining, for each of the plurality of first-stage abnormal positions, at least one candidate offset distance adjusted based on the position of the first-stage abnormal position, and adding the at least one candidate offset distance to the first-stage abnormal position.

5. Comparing the candidate position offset with the plurality of first-stage abnormal positions and the plurality of second-stage abnormal positions, The candidate position offset is applied to the plurality of second-stage abnormal positions to determine the plurality of offset second-stage abnormal positions, Determining the difference between the plurality of offset second-stage abnormal positions and the plurality of first-stage abnormal positions. The method according to claim 1, including the method described in claim 1.

6. The method according to claim 5, wherein determining the difference includes determining the respective offset difference for each of the plurality of offset second-stage abnormal positions.

7. The method according to claim 5, wherein applying the candidate position offset includes determining, for each of the plurality of second-stage abnormal positions, at least one candidate offset distance adjusted based on the position of the second-stage abnormal position, and adding the at least one candidate offset distance to the second-stage abnormal position.

8. The method according to claim 3, wherein comparing the candidate position offset with the plurality of first-stage abnormal positions and the plurality of second-stage abnormal positions includes determining a weighted offset difference for each of the offset differences based on the offset difference and the abnormal weight associated with the offset difference.

9. The method according to claim 8, comprising determining, for each of the offset differences, the associated anomaly weight based on the severity of the associated anomaly in which the offset difference was determined.

10. The method according to claim 1, wherein receiving the representations of the plurality of second-stage abnormal positions includes receiving representations of a plurality of candidate second-stage abnormal positions and determining the plurality of second-stage abnormal positions as a subset of the plurality of candidate second-stage abnormal positions.

11. The method according to claim 10, wherein determining the plurality of second-stage abnormal locations includes ranking the plurality of candidate second-stage abnormal locations and selecting the subset as one or more highest-ranked candidate second-stage abnormal locations.

12. The method according to claim 11, wherein ranking the plurality of candidate second-stage abnormal positions includes ranking each of the plurality of candidate second-stage abnormal positions based at least in part on the proximity of the second-stage abnormal position to the position of the rolled paper in the second rolled paper processing stage.

13. The method according to claim 11, wherein ranking the plurality of candidate second-stage anomaly locations includes ranking each of the plurality of candidate second-stage anomaly locations on at least partly the severity of the anomaly associated with the second-stage anomaly location.

14. The method according to claim 1, wherein the second winding paper processing step is downstream of the first winding paper processing step.

15. The method according to claim 1, wherein identifying the determined position offset includes identifying a candidate position offset among a plurality of candidate position offsets associated with the smallest of the representations of the difference.

16. Receiving the representation of the plurality of first-stage abnormal positions, Receiving one or more sets of first-stage images of the rolled paper in the first rolled paper processing step, Determining the plurality of first-stage anomaly locations based at least partially on the application of at least one first-stage anomaly recognition sensitivity to one or more sets of first-stage images. The method according to claim 1, including the method described in claim 1.

17. The one or more sets of first-stage images include a first set of first-stage images and a second set of first-stage images. The at least one first-stage anomaly identification sensitivity includes a first first-stage anomaly identification sensitivity and a second first-stage anomaly identification sensitivity, wherein the second first-stage anomaly identification sensitivity differs from the first first-stage anomaly identification sensitivity. Determining the multiple first-stage abnormal positions is Determining a first set of first-stage anomaly locations based at least in part on the application of the first first-stage anomaly recognition sensitivity to the first set of first-stage images, Determining a second set of first-stage anomaly locations based at least in part on the application of the second first-stage anomaly recognition sensitivity to the second set of first-stage images. The method according to claim 16, including the method described in claim 16.

18. The method according to claim 17, wherein each of the first and second first-stage anomaly recognition sensitivities includes a plurality of anomaly recognition thresholds, each associated with a respective pixel position.

19. The method according to claim 17, comprising determining at least one anomaly density associated with the first first-stage anomaly recognition sensitivity, and determining the second first-stage anomaly recognition sensitivity based at least in part on the at least one anomaly density associated with the first first-stage anomaly recognition sensitivity and the first first-stage anomaly recognition sensitivity.

20. The method according to claim 19, wherein determining the at least one anomaly density includes determining the number of anomalies represented by the first set of the plurality of first-stage anomaly locations.

21. The method according to claim 19, wherein determining the at least one anomaly density includes determining the number of at least one anomaly pixels included in the anomaly represented by the first set of the plurality of first-stage anomaly locations.

22. The method according to claim 19, comprising determining at least one difference between the at least one anomaly density associated with the first first-stage anomaly recognition sensitivity and a desired first-stage anomaly density, wherein determining the second first-stage anomaly recognition sensitivity comprises determining the second first-stage anomaly recognition sensitivity based at least in part on the determined at least one difference.

23. Receiving the representation of the plurality of second-stage abnormal positions, Receiving one or more sets of second-stage images of the rolled paper in the second rolled paper processing step, Determining the plurality of second-stage anomaly locations based at least partially on the application of at least one second-stage anomaly recognition sensitivity to one or more sets of second-stage images. Includes, The one or more sets of second-stage images include a first set of second-stage images and a second set of second-stage images. The at least one second-stage anomaly identification sensitivity includes a first second-stage anomaly identification sensitivity and a second second-stage anomaly identification sensitivity, Determining the multiple second-stage abnormal positions is Determining a first set of second-stage anomaly locations based at least in part on the application of the first second-stage anomaly recognition sensitivity to the first set of second-stage images, Determining a second set of second-stage anomaly locations based at least in part on the application of the second second-stage anomaly recognition sensitivity to the second set of second-stage images. Includes, The method includes determining at least one anomaly density associated with the first second-stage anomaly recognition sensitivity, and determining the second second-stage anomaly recognition sensitivity based at least partially on the at least one anomaly density associated with the first second-stage anomaly recognition sensitivity and the first second-stage anomaly recognition sensitivity. The method includes determining the difference between the at least one anomaly density associated with the first second-stage anomaly identification sensitivity and a desired second-stage anomaly density, wherein the desired second-stage anomaly density is less than 90% of the desired first-stage anomaly density. Determining the second second-stage anomaly recognition sensitivity includes determining the second second-stage anomaly recognition sensitivity based at least in part on the determined difference between the at least one anomaly density associated with the first second-stage anomaly recognition sensitivity and the desired second-stage anomaly density. The method according to claim 22.

24. The method according to claim 23, wherein determining the at least one anomaly density associated with the first second-stage anomaly identification sensitivity includes determining the number of anomaly pixels included in the anomaly represented by the first set of the plurality of second-stage anomaly locations.

25. The method according to claim 23, wherein determining the at least one anomaly density associated with the first second-stage anomaly identification sensitivity includes determining the number of anomalies represented by the first set of the plurality of second-stage anomaly locations.

26. The method according to claim 23, wherein each of the first and second second-stage anomaly recognition sensitivities includes a plurality of anomaly recognition thresholds, each associated with a respective pixel position.

27. Receiving the representation of the plurality of second-stage abnormal positions, Receiving one or more sets of second-stage images of the rolled paper in the second rolled paper processing step, Determining the plurality of second-stage anomaly locations based at least partially on the application of at least one second-stage anomaly recognition sensitivity to one or more sets of second-stage images. The method according to claim 1, including the method described in claim 1.

28. The one or more sets of second-stage images include a first set of second-stage images and a second set of second-stage images. The at least one second-stage anomaly identification sensitivity includes a first second-stage anomaly identification sensitivity and a second second-stage anomaly identification sensitivity, wherein the second second-stage anomaly identification sensitivity differs from the first second-stage anomaly identification sensitivity. Determining the multiple second-stage abnormal positions is Determining a first set of second-stage anomaly locations based at least in part on the application of the first second-stage anomaly recognition sensitivity to the first set of second-stage images, Determining a second set of second-stage anomaly locations based at least in part on the application of the second second-stage anomaly recognition sensitivity to the second set of second-stage images. The method according to claim 27, including the method described in claim 27.

29. The method according to claim 28, comprising determining at least one anomaly density associated with the first second-stage anomaly recognition sensitivity, and determining the second second-stage anomaly recognition sensitivity based at least in part on the at least one anomaly density associated with the first second-stage anomaly recognition sensitivity and the first second-stage anomaly recognition sensitivity.

30. The method according to claim 29, wherein determining the at least one anomaly density includes determining the number of anomalies represented by the first set of the plurality of second-stage anomaly locations.

31. The method according to claim 29, wherein determining the at least one anomaly density includes determining the number of at least one anomaly pixels included in the anomaly represented by the first set of the plurality of second-stage anomaly locations.

32. The method according to claim 29, comprising determining at least one difference between the at least one anomaly density associated with the first second-stage anomaly identification sensitivity and a desired second-stage anomaly density, wherein determining the second second-stage anomaly identification sensitivity comprises determining the second second-stage anomaly identification sensitivity based at least in part on the determined at least one difference.

33. The method according to claim 28, wherein each of the first and second first-stage anomaly recognition sensitivities includes a plurality of anomaly recognition thresholds, each associated with a respective pixel position.

34. Receiving a calibration set of the first stage image of the rolled paper in the first rolled paper processing step, Determining at least one first-stage calibration anomaly density based at least partially on the application of the first-stage calibration anomaly detection sensitivity to the calibration set of first-stage images, Determining the calibration-based first-stage anomaly recognition sensitivity based at least partially on the first-stage calibration anomaly recognition sensitivity and the at least one first-stage calibration anomaly density. The method according to claim 1, including the method described in claim 1.

35. Determining the at least one first-stage calibration anomaly density is The first-stage calibration set of abnormal locations is determined at least partially based on the application of the first-stage calibration abnormality detection sensitivity to the calibration set of the first-stage images, Determine the number of abnormalities represented by the first-stage calibration set of abnormal locations. The method according to claim 34, including the method described in claim 34.

36. The method according to claim 34, wherein determining the at least one first-stage calibration anomaly density includes determining the number of anomaly pixels included in the anomaly in the calibration set of the first-stage image.

37. The method according to claim 34, wherein each of the first-stage calibration anomaly detection sensitivity and the calibration-based first-stage anomaly detection sensitivity includes a plurality of anomaly detection thresholds, each associated with a respective pixel position.

38. The method according to claim 34, wherein determining the calibration-based first-stage anomaly recognition sensitivity includes determining at least one difference between the at least one first-stage calibration anomaly density and a desired first-stage calibration anomaly density, and determining the calibration-based first-stage anomaly recognition sensitivity at least in part on the determined at least one difference.

39. The method according to claim 34, wherein the first set of first-stage images includes at least one of the calibration sets of first-stage images.

40. The method according to claim 39, wherein the first set of first-stage images and the calibration set of first-stage images are the same image.

41. In the first winding paper processing step, one or more representations of detected first-stage defect locations representing the location of a defect detected on the winding paper are received. Receiving a representation of the detected second stage position of the rolled paper in the second rolled paper processing stage, wherein the detected second stage position represents the current position of the rolled paper in the second rolled paper processing stage. Based on the determined position offset, the detected second-stage position of the rolled paper, and the one or more detected first-stage defect positions, the degree of defect proximity of the detected second-stage position to at least one of the defects is determined. If the determined defect proximity satisfies a threshold criterion, a signal is generated to adjust the processing in the second roll paper processing stage. The method according to claim 1, including the method described in claim 1.

42. The method according to claim 41, wherein determining the defect proximity includes applying the determined position offset to one or more detected first-stage defect locations to determine one or more predicted second-stage defect locations representing predicted defect locations for the rolled paper in the second rolled paper processing step, and comparing the one or more predicted second-stage defect locations with the detected second-stage locations.

43. The method according to claim 41, wherein determining the defect proximity includes applying the determined position offset to the detected second-stage position to determine an offset detection position, and comparing the one or more detected first-stage defect positions with the offset detection position.

44. A system for facilitating the tracking of the position of a roll of paper, comprising at least one processor configured to perform the method according to any one of claims 1 to 43.

45. A non-temporary computer-readable medium that, when executed by at least one processor, stores code causing the at least one processor to perform the method according to any one of claims 1 to 43.

Images