A device for identifying impurities in waste paper using X-rays

Abstract

The application discloses a device for recognizing impurities in waste paper by X-rays, and relates to the technical field of waste paper recognition. The device comprises an X-ray light source, a detector, an image analysis module, a state monitoring module and an operation evaluation module. The X-ray light source is arranged above a conveying belt machine and is used for emitting X-rays. When an object to be detected passes, a photoelectric sensor is triggered to control the X-ray light source to start. After the X-rays pass through a collimator, very narrow fan-shaped ray beams are formed and fall on the detector after penetrating the object to be detected on the conveying belt. The detector converts the received X-rays into electric signals, which are amplified by a signal discharge circuit and then quantized. The quantized signals are transmitted to the image analysis module through a serial bus for further processing, so that high-quality images are obtained. The device can recognize impurities such as mixed metals, ceramics, glass and stone blocks, and can measure the proportion and area of the impurities. The device eliminates the safety hazard that the inspector is not separated from the vehicle during the inspection operation, and greatly reduces the manual operation strength.

Description

Technical Field

[0001] This invention relates to the field of waste paper identification technology, specifically a device for identifying impurities in waste paper using X-rays. Background Technology

[0002] Currently, during the inspection of baled waste paper, companies need to manually determine the impurity ratio of the loosened waste paper before calculating deductions according to regulations. This manual inspection is affected by the degree of dispersion of the loose paper in the bales and the differences in individual experience in judging the impurity ratio within the bales. Inconsistencies in individual understanding and application of inspection standards can also lead to discrepancies in judgment results. If significant differences occur in the deductions for impurities from the same customer, the customer may have doubts and misunderstandings about the company's inspection standards, damaging the company's image. From an ethical perspective, there is also the possibility of inspectors engaging in illegal activities such as clandestine dealings with suppliers or related professionals, or accepting bribes, which could harm the company's interests. Based on these shortcomings, this invention proposes a device for identifying impurities in loose waste paper using X-rays. Summary of the Invention

[0003] This invention aims to solve at least one of the technical problems existing in the prior art. To this end, this invention proposes a device for identifying impurities in waste paper using X-rays. Utilizing the penetrating power of X-rays, it enables rapid inspection of raw waste paper, identifying impurities such as metals, ceramics, glass, and stones, and calculating their proportion and area. The equipment is safe and reliable in operation, and can replace current manual on-site inspection, eliminating the safety hazards of personnel not separating from the inspection vehicle, and greatly reducing the intensity of manual operation.

[0004] To achieve the above objectives, an apparatus for identifying impurities in waste paper using X-rays is provided according to an embodiment of the first aspect of the present invention, comprising an X-ray source, a detector, an image analysis module, a status monitoring module, and a computational evaluation module;

[0005] The X-ray source is located above the conveyor belt and is used to emit X-rays. When an item being inspected passes by, a photoelectric sensor is triggered to control the start of the X-ray source. After passing through the collimator, the X-rays form a very narrow fan-shaped beam that penetrates the item being inspected on the conveyor belt and falls onto the detector.

[0006] The detector is used for automatic continuous imaging. Specifically, the detector converts the received X-rays into electrical signals, which are then amplified and quantized by a signal discharge circuit and transmitted to the image analysis module via a serial bus. The image analysis module performs complex calculations and imaging processing on the received electrical signals to obtain a high-quality image and calculates the type and proportion of impurities in the image.

[0007] The status monitoring module is connected to the image analysis module and is used to monitor the status parameters of the image analysis module during the calculation process and evaluate the calculation deviation value. If the calculation deviation value YP is greater than the preset deviation threshold, it indicates that the calculation processing capability of the image analysis module is low and a deviation warning signal is generated to remind the management personnel to replace the industrial computer for calculation processing.

[0008] The status monitoring module is also used to integrate the processing time and deviation value YP of the image analysis module to form a processing record of the image analysis module and store it in the cloud platform with a timestamp; the processing evaluation module is used to evaluate the processing value of the image analysis module based on the processing record with timestamp stored in the cloud platform and transmit the processing value GY of the image analysis module to the display for real-time display.

[0009] Furthermore, the specific monitoring process of the status monitoring module is as follows:

[0010] When the image analysis module is detected to be processing electrical signals, the status parameters of the image analysis module are collected every R2 time interval. The status parameters include the number of access node connections, CPU load rate, bandwidth load rate and real-time network speed.

[0011] The number of access node connections, CPU load rate, bandwidth load rate, and real-time network speed are labeled as W1, W2, W3, and W4 respectively. The operation coefficient YS of the image analysis module is calculated using the formula YS=(W1×b1+W4×b4) / (W2×b2+W3×b3), where b1, b2, b3, and b4 are coefficient factors.

[0012] Establish a curve of the operation coefficient YS changing over time; compare the operation coefficient YS with the preset operation threshold; if YS ≤ the preset operation threshold, extract the corresponding curve segment in the corresponding curve and mark it as the deviation curve segment;

[0013] The processing time of the image analysis module is marked as YT; the number of deviation curve segments is counted as the deviation frequency P1; the difference between the corresponding YS in the deviation curve segment and the preset operation threshold is integrated over time to obtain the deviation reference area M1; the operation deviation value YP is calculated using the formula YP=(P1×a1+M1×a2) / (YT×a3), where a1, a2, and a3 are all coefficient factors.

[0014] Furthermore, the specific evaluation steps of the computation evaluation module are as follows:

[0015] Based on the timestamp, retrieve all the operation and processing records of the image analysis module within the preset time period;

[0016] The total number of operations in the statistical image analysis module is the operation frequency N1; the operation processing time in each operation processing record is marked as NTi, and the operation deviation value is marked as YPi. The operation unit value NPi is calculated using the formula NPi=NTi×a4+YPi×a5, where a4 and a5 are coefficient factors.

[0017] Compare the calculated unit value NPi with the set value; count the percentage of times NPi is greater than the set value as Zb; when NPi is greater than the set value, obtain the difference between NPi and the set value and sum them to obtain the calculation excess value ZY; calculate the excess coefficient NF using the formula NF=Zb×g1+ZY×g2, where g1 and g2 are coefficient factors; use the formula... The computational estimate GY of the image analysis module is calculated, where g3 and g4 are coefficient factors.

[0018] Furthermore, the photoelectric sensor consists of a light-emitting diode and a photoelectric switch, which are symmetrically arranged on both sides of the conveyor belt; the specific working steps are as follows:

[0019] The items being inspected are conveyed by a conveyor belt. When the items being inspected pass the photoelectric sensor, the light emitted by the light-emitting diode is blocked, and the photoelectric switch is in the off state.

[0020] The photoelectric sensor receives the switching signal from the photoelectric switch and transmits the switching signal to the controller, which then triggers the X-ray source to emit X-rays.

[0021] Furthermore, the specific computational processing of the image analysis module is as follows:

[0022] First, a detector is used to acquire image information of the unblemished loose paper, generating standard image data; the standard image data is synthesized by statistical averaging of image information from multiple unblemished loose papers.

[0023] The detector acquires the electrical signal information of the inspected item, performs filtering, sharpening, mathematical morphology transformation, binarization, edge extraction, and contour extraction on the acquired electrical signal information, and obtains the image data of the inspected item, which is then marked as the image data to be tested.

[0024] Based on the SURF algorithm, image matching is performed between the image data to be tested and the standard image data, and the impurity category and proportion in the image are analyzed and calculated.

[0025] The image analysis module is used to transmit the image data to be tested, as well as the types and proportions of impurities in the image calculated by analysis, to the display for real-time display, and to store the relevant data on the server.

[0026] Furthermore, based on the SURF algorithm, image matching is performed between the image data to be tested and the standard image data, specifically including:

[0027] A region is selected as a template in the standard image data, and the SURF algorithm based on color invariants is used to extract the feature points of the image by using the color invariants calculated from the image color information.

[0028] After extracting feature points, feature descriptors are generated for the feature points by combining the grayscale information of the image;

[0029] Euclidean distance is used for similarity measurement. Feature points that match between two image data are extracted. After finding the matching points, the data to be detected and the standard image are registered and compared. The impurity categories and proportions in the image are analyzed and calculated.

[0030] Compared with the prior art, the beneficial effects of the present invention are:

[0031] 1. In this invention, the detector converts the received X-rays into electrical signals, which are then amplified and quantized by the signal discharge circuit and transmitted to the image analysis module via a serial bus. The image analysis module performs complex calculations and imaging processing on the received electrical signals to obtain a high-quality image and calculates the type and proportion of impurities in the image. The device is safe and reliable to operate and can replace the current manual on-site observation and inspection, eliminating the safety hazards of personnel and vehicles not being separated during inspection operations and greatly reducing the intensity of manual operation.

[0032] 2. The status monitoring module monitors the status parameters of the image analysis module during the computation process. First, it calculates the computation coefficient YS by combining the number of access node connections, CPU load rate, bandwidth load rate, and real-time network speed. Then, it evaluates the computation deviation based on the spatiotemporal changes of the computation coefficient YS. If the computation deviation YP exceeds a preset deviation threshold, a deviation warning signal is generated to remind management personnel to replace the industrial computer for computation, thereby improving the accuracy and efficiency of impurity identification. The computation evaluation module evaluates the computational value of the image analysis module based on the computational processing records. This allows management personnel to have a clear understanding of the image analysis module's computational capabilities, enabling them to select the appropriate image analysis module for computation and further improve the accuracy and efficiency of impurity identification. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a system block diagram of the present invention.

[0035] Figure 2 This is a schematic diagram illustrating the working principle of impurity identification using X-rays in this invention. Detailed Implementation

[0036] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] like Figures 1 to 2 As shown, a device for identifying impurities in waste paper using X-rays includes an X-ray source, a photoelectric sensor, a controller, a detector, an image analysis module, a display, a status monitoring module, an alarm module, a cloud platform, and a computation and evaluation module.

[0038] The X-ray source is located above the waste paper conveyor belt and is used to emit X-rays. When loose paper passes by, a photoelectric switch is triggered to control the start of the X-ray source.

[0039] The photoelectric sensor consists of a light-emitting diode (LED) and a photoelectric switch, which are symmetrically arranged on both sides of the conveyor belt. The specific working steps are as follows:

[0040] The items being inspected are conveyed by a conveyor belt. When the items being inspected pass the photoelectric sensor, the light emitted by the light-emitting diode is blocked, and the photoelectric switch is in the off state.

[0041] The photoelectric sensor receives the switching signal from the photoelectric switch and transmits the switching signal to the controller, which then triggers the X-ray source to emit X-rays.

[0042] After passing through the collimator, the X-rays form a very narrow fan-shaped beam that penetrates the inspected items on the conveyor belt and falls onto the detector. The detector is used for automatic continuous imaging, and the imaging area is set up with a semi-enclosed space, leaving an inlet and outlet for waste paper. Specifically, the detector converts the received X-rays into electrical signals, which are amplified and quantized by the signal discharge circuit and transmitted to the image analysis module for further processing via a serial bus. The image analysis module is an industrial computer.

[0043] The image analysis module processes the received electrical signal through complex calculations and imaging to obtain a high-quality image, and calculates the type and proportion of impurities within the image; specifically:

[0044] First, a detector is used to acquire image information of the unblemished loose paper, generating standard image data; the standard image data is synthesized by statistical averaging of image information from multiple unblemished loose papers.

[0045] The detector acquires the electrical signal information of the inspected item, performs filtering, sharpening, mathematical morphology transformation, binarization, edge extraction, and contour extraction on the acquired electrical signal information, and obtains the image data of the inspected item, which is then marked as the image data to be tested.

[0046] Based on the SURF algorithm, image matching is performed between the image data to be tested and the standard image data. The specific steps are as follows:

[0047] A region is selected as a template in the standard image data, and the SURF algorithm based on color invariants is used to extract the feature points of the image by using the color invariants calculated from the image color information.

[0048] After extracting feature points, feature descriptors are generated for the feature points by combining the grayscale information of the image;

[0049] Euclidean distance is used for similarity measurement. Feature points matching between two image data are extracted. After finding the matching points, the registration of the image data to be detected and the standard image are completed and the data are compared. The impurity category and proportion in the image are analyzed and calculated.

[0050] The image analysis module transmits the image data to be tested, along with the calculated types and proportions of impurities within the image, to the display for real-time viewing, and stores the relevant data on the server; the relevant data includes the image data to be tested and the calculated types and proportions of impurities within the image.

[0051] In this embodiment, the identification device consists of an X-ray imaging part and a computing power calculation part. The X-ray imaging part can automatically run and continuously take pictures, and then the computing power calculation part identifies and outputs the results and stores the relevant data on the server. The device operates safely and reliably and can replace the current manual on-site observation and inspection, eliminating the safety hazards of the inspection operation where the personnel and vehicle are not separated, and greatly reducing the intensity of manual operation.

[0052] The status monitoring module is connected to the image analysis module and is used to monitor the status parameters of the image analysis module during the computation process and to evaluate the computational deviation. The specific monitoring process is as follows:

[0053] When the image analysis module is detected to be processing electrical signals, the status parameters of the image analysis module are collected every R2 time interval. The status parameters include the number of access node connections, CPU load rate, bandwidth load rate and real-time network speed.

[0054] The number of access node connections, CPU load rate, bandwidth load rate, and real-time network speed are labeled as W1, W2, W3, and W4 respectively. The operation coefficient YS of the image analysis module is calculated using the formula YS=(W1×b1+W4×b4) / (W2×b2+W3×b3), where b1, b2, b3, and b4 are coefficient factors.

[0055] Establish a curve of the operation coefficient YS changing over time; compare the operation coefficient YS with the preset operation threshold; if YS ≤ the preset operation threshold, extract the corresponding curve segment in the corresponding curve and mark it as the deviation curve segment;

[0056] The processing time of the image analysis module is marked as YT; the number of deviation curve segments is counted as the deviation frequency P1; the difference between the corresponding YS in the deviation curve segment and the preset operation threshold is integrated over time to obtain the deviation reference area M1; the processing time, deviation frequency and deviation reference area are normalized and their values ​​are taken, and the operation deviation value YP is calculated using the formula YP=(P1×a1+M1×a2) / (YT×a3), where a1, a2 and a3 are coefficient factors;

[0057] The calculated deviation value YP is compared with the preset deviation threshold. If YP is greater than the preset deviation threshold, it indicates that the image analysis module has low processing capability and a deviation warning signal is generated.

[0058] The status monitoring module is used to transmit deviation warning signals to the controller; after receiving the deviation warning signal, the controller controls the alarm module to issue an alarm to remind the management personnel to replace the industrial computer for processing, thereby improving the accuracy and efficiency of impurity identification.

[0059] The status monitoring module is also used to integrate the processing time and deviation value YP of the image analysis module to form a processing record of the image analysis module, which is then timestamped and stored on the cloud platform.

[0060] The computation evaluation module is connected to the cloud platform and is used to evaluate the computational value of the image analysis module based on the timestamped computational processing records stored in the cloud platform. This allows managers to have an intuitive understanding of the computational processing of the image analysis module, thereby selecting the appropriate image analysis module for computational processing and further improving the accuracy and efficiency of impurity identification.

[0061] The specific evaluation steps of the computation evaluation module are as follows:

[0062] Based on the timestamp, retrieve all the operation and processing records of the image analysis module within the preset time period;

[0063] The total number of operations in the statistical image analysis module is the operation frequency N1. The operation processing time in each operation processing record is marked as NTi, and the operation deviation value is marked as YPi. The operation unit value NPi is calculated using the formula NPi=NTi×a4+YPi×a5, where a4 and a5 are coefficient factors. The larger the operation unit value NPi is, the greater the computational load of the image analysis module.

[0064] Compare the calculated unit value NPi with the set value; count the percentage of times NPi is greater than the set value as Zb; when NPi is greater than the set value, obtain the difference between NPi and the set value and sum them to obtain the calculation excess value ZY; use the formula NF=Zb×g1+ZY×g2 to calculate the excess coefficient NF, where g1 and g2 are coefficient factors;

[0065] The frequency of operations and the overrun coefficient are normalized and their values ​​are taken, then the formula is used. The computational estimate GY of the image analysis module is calculated, where g3 and g4 are coefficient factors; the computational evaluation module is used to transmit the computational estimate GY of the image analysis module to the display for real-time display.

[0066] The above formulas are all numerical calculations after removing dimensions. The formulas are obtained by software simulation based on a large amount of data and are closest to the real situation. The preset parameters and preset thresholds in the formulas are set by those skilled in the art according to the actual situation or obtained by simulation based on a large amount of data.

[0067] Working principle of the invention:

[0068] A device for identifying impurities in waste paper using X-rays is disclosed. During operation, the inspected item is conveyed via a conveyor belt. When the item passes a photoelectric sensor, a photoelectric switch is triggered to activate the X-ray source. The X-rays, after passing through a collimator, form a very narrow fan-shaped beam that penetrates the item on the conveyor belt and falls onto the detector. The detector converts the received X-rays into an electrical signal, which is amplified and quantized by a signal discharge circuit and transmitted to an image analysis module via a serial bus. The image analysis module performs complex calculations and imaging processing on the received electrical signal to obtain a high-quality image and calculates the type and proportion of impurities within the image. This device is safe and reliable, replacing current manual on-site inspection, eliminating the safety hazards associated with the separation of personnel and the inspection vehicle, and significantly reducing the intensity of manual operation.

[0069] The status monitoring module monitors the status parameters of the image analysis module during the computation process. First, it calculates the computation coefficient YS by combining the number of access node connections, CPU load rate, bandwidth load rate, and real-time network speed. Then, it evaluates the computation deviation based on the spatiotemporal changes of the computation coefficient YS. If the computation deviation YP exceeds a preset deviation threshold, a deviation warning signal is generated to remind management personnel to replace the computation with a new industrial computer, thereby improving the accuracy and efficiency of impurity identification. The computation evaluation module evaluates the computational value of the image analysis module based on the computational processing records. This allows management personnel to have a clear understanding of the image analysis module's computational capabilities, enabling them to select the appropriate image analysis module for computation and further improve the accuracy and efficiency of impurity identification.

[0070] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0071] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A device for identifying impurities in waste paper using X-rays, characterized in that, It includes an X-ray source, detector, image analysis module, status monitoring module, and computational evaluation module; The X-ray source is located above the conveyor belt and is used to emit X-rays. When an item being inspected passes by, a photoelectric sensor is triggered to control the start of the X-ray source. After passing through the collimator, the X-rays form a very narrow fan-shaped beam that penetrates the item being inspected on the conveyor belt and falls onto the detector. The detector is used for automatic continuous imaging. Specifically, the detector converts the received X-rays into electrical signals, which are then amplified and quantized by a signal discharge circuit and transmitted to the image analysis module via a serial bus. The image analysis module performs complex calculations and imaging processing on the received electrical signals to obtain a high-quality image and calculates the type and proportion of impurities in the image. The status monitoring module is connected to the image analysis module and is used to monitor the status parameters of the image analysis module during the computation process and to evaluate the computation deviation value YP. The specific monitoring process is as follows: When the image analysis module is detected to be processing electrical signals, the status parameters of the image analysis module are collected every R2 time interval. The status parameters include the number of access node connections, CPU load rate, bandwidth load rate and real-time network speed. The number of access node connections, CPU load rate, bandwidth load rate, and real-time network speed are labeled as W1, W2, W3, and W4 respectively. The operation coefficient YS of the image analysis module is calculated using the formula YS=(W1×b1+W4×b4) / (W2×b2+W3×b3), where b1, b2, b3, and b4 are coefficient factors. Establish a curve of the operation coefficient YS changing over time; compare the operation coefficient YS with the preset operation threshold; if YS ≤ the preset operation threshold, extract the corresponding curve segment in the corresponding curve and mark it as the deviation curve segment; The processing time of the image analysis module is marked as YT; the number of deviation curve segments is counted as the deviation frequency P1; the difference between the corresponding YS in the deviation curve segment and the preset operation threshold is integrated over time to obtain the deviation reference area M1; The operational deviation value YP is calculated using the formula YP=(P1×a1+M1×a2) / (YT×a3), where a1, a2, and a3 are all coefficient factors. If the deviation value YP is greater than the preset deviation threshold, it indicates that the image analysis module has low processing capability and generates a deviation warning signal to remind the management personnel to replace it with a new industrial computer for processing. The status monitoring module is also used to integrate the processing time and deviation value YP of the image analysis module to form the processing record of the image analysis module and store it in the cloud platform with a timestamp; the operation evaluation module is used to evaluate the operation value GY of the image analysis module according to the time-stamped processing record stored in the cloud platform, and transmit the operation value GY of the image analysis module to the display for real-time display. The specific evaluation steps of the computation evaluation module are as follows: Based on the timestamp, retrieve all the operation and processing records of the image analysis module within the preset time period; The total number of operations in the statistical image analysis module is the operation frequency N1; the operation processing time in each operation processing record is marked as NTi, and the operation deviation value is marked as YPi. The operation unit value NPi is calculated using the formula NPi=NTi×a4+YPi×a5, where a4 and a5 are coefficient factors. Compare the calculated unit value NPi with the set value; count the percentage of times NPi is greater than the set value as Zb; when NPi is greater than the set value, obtain the difference between NPi and the set value and sum them to obtain the calculation excess value ZY; use the formula NF=Zb×g1+ZY×g2 to calculate the excess coefficient NF, where g1 and g2 are coefficient factors; Using formula The computational estimate GY of the image analysis module is calculated, where g3 and g4 are coefficient factors.

2. The device for identifying impurities in waste paper using X-rays according to claim 1, characterized in that, The photoelectric sensor consists of a light-emitting diode (LED) and a photoelectric switch, which are symmetrically arranged on both sides of the conveyor belt. The specific operating steps are as follows: The items being inspected are conveyed by a conveyor belt. When the items being inspected pass the photoelectric sensor, the light emitted by the light-emitting diode is blocked, and the photoelectric switch is in the off state. The photoelectric sensor receives the switching signal from the photoelectric switch and transmits the switching signal to the controller, which then triggers the X-ray source to emit X-rays.

3. The device for identifying impurities in waste paper using X-rays according to claim 1, characterized in that, The specific computational process of the image analysis module is as follows: First, a detector is used to acquire image information of the unblemished loose paper, generating standard image data; the standard image data is synthesized by statistical averaging of image information from multiple unblemished loose papers. The detector acquires the electrical signal information of the inspected item, performs filtering, sharpening, mathematical morphology transformation, binarization, edge extraction, and contour extraction on the acquired electrical signal information, and obtains the image data of the inspected item, which is then marked as the image data to be tested. Based on the SURF algorithm, image matching is performed between the image data to be tested and the standard image data, and the impurity category and proportion in the image are analyzed and calculated. The image analysis module is used to transmit the image data to be tested, as well as the types and proportions of impurities in the image calculated by analysis, to the display for real-time display, and to store the relevant data on the server.

4. The device for identifying impurities in waste paper using X-rays according to claim 3, characterized in that, Based on the SURF algorithm, image matching is performed between the image data to be tested and standard image data, specifically including: A region is selected as a template in the standard image data, and the SURF algorithm based on color invariants is used to extract the feature points of the image by using the color invariants calculated from the image color information. After extracting feature points, feature descriptors are generated for the feature points by combining the grayscale information of the image; Euclidean distance is used for similarity measurement. Feature points that match between two image data are extracted. After finding the matching points, the data to be detected and the standard image are registered and compared. The impurity categories and proportions in the image are analyzed and calculated.

Images