Method for calibrating code spraying position of paper mounting machine based on visual identification

Abstract

The invention discloses a paper mounting machine code spraying position calibration method based on visual identification, and relates to the technical field of image identification, and the method comprises the steps: periodically collecting a mounting paper code spraying image sequence, and shooting the code-sprayed mounting paper in a preset evaluation period through an industrial camera; a code spraying area connected domain meeting the area condition is extracted, and a minimum enclosing rectangle is fitted; calculating a pixel intersection set and a pixel difference set of an actual code spraying area and a standard code spraying area by constructing a two-dimensional coordinate system to obtain a position deviation rate; when the deviation rate exceeds a threshold value, determining an abnormal state, and comprehensively determining the state of the whole machine according to the number of abnormal images; aiming at the abnormal condition, generating a correction instruction containing rotation, transverse and longitudinal offset parameters, and dynamically adjusting the code spraying position of the next period; according to the method, real-time monitoring and automatic calibration of the code spraying position are realized, and the code spraying precision and the production stability are effectively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of image recognition, and in particular relates to a method for calibrating the inkjet position of a paper laminating machine based on visual recognition. Background Art

[0002] With the continuous progress of society and sustained economic development, the demand for automation and intelligent technologies in industrial production is becoming increasingly urgent. In the application process of paper laminating equipment, inkjet coding technology plays a key role. It can provide unique identification and information records for products, facilitating product tracking, management, and quality control.

[0003] The existing technology has many shortcomings and disadvantages in the calibration of the coding position of paper laminating machines. First, the calibration accuracy is limited, making it difficult to accurately determine whether the coding position is accurate, which leads to deviations in the coding position and affects product quality and aesthetics. Second, there is a lack of effective real-time monitoring and evaluation mechanisms, making it impossible to promptly detect deviations from the standard area. As a result, the problem cannot be dealt with in a timely manner, which may result in a large number of defective products.

[0004] In order to solve the above problems, this application proposes a method for calibrating the inkjet coding position of a paper laminating machine based on visual recognition. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the present invention provides a method for calibrating the inkjet position of a paper laminating machine based on visual recognition, which solves the problem of poor accuracy in determining the inkjet position of a paper laminating machine in the existing technology.

[0006] The purpose of the present invention can be achieved through the following technical solutions: A method for calibrating the inkjet printer position of a paper laminating machine based on visual recognition includes the following steps: Step 1: Determine an evaluation period. During the current evaluation period, use a pre-built industrial camera to photograph a number of mounted papers that have completed coding, obtaining a number of mounted paper coding images. The number of mounted paper coding images are recorded as a mounted paper coding image sequence according to the shooting timeline. Step 2: traverse the paper-mounted coding image sequence, extract the coding area in the paper-mounted coding image, perform position deviation analysis on the coding area, and determine the coding position status; Step 3: If the inkjet position status is determined to be abnormal, the inkjet position correction parameters are generated in this evaluation cycle in combination with the pre-built standard inkjet area, and the inkjet position in the next evaluation cycle is calibrated.

[0007] As a further solution of the present invention, in step 1, the specific method of determining the evaluation period is: Determine a number of normal laminating machines and record the number of normal laminating machines The time it takes to complete the coding of the mounting paper is several times, among which, Preset values ​​for operators; Take the average of several durations as the duration of an evaluation cycle, recorded as .

[0008] As a further solution of the present invention, in step 1, the specific method of recording the plurality of mounting inkjet coding images as a mounting inkjet coding image sequence is as follows: The current moment is taken as the starting moment of the evaluation cycle, recorded as ; exist The evaluation cycle is superimposed on the The duration of ; Leverage pre-built industrial cameras to to The mounted paper after coding is photographed, and the total number of the photographed mounted paper coding images is recorded as ; In the order of shooting The paper-mounted inkjet coding images are recorded as: paper-mounted inkjet coding image sequence ; Among them, the mapping relationship between the mounted paper inkjet coding image and the real object is determined by the operator.

[0009] As a further solution of the present invention, in step 2, the specific method of extracting the inkjet coding area in the mounted paper inkjet coding image is: Determine the mounting coding image sequence Any mounted paper inkjet image ,in, is the counting index, ; right Perform grayscale processing to obtain a grayscale image ; Get the segmentation pixel value threshold preset by the operator ; based on Generate grayscale image The associated binary image ; Extract binary image All connected areas with pixel values ​​of 225 are summarized as the connected area set ,in, is the total number of connected regions in C; Build The minimum bounding rectangle of each connected area is the set of minimum bounding rectangles ; Get the area range of the coding area preset by the operator ; The area of ​​the minimum bounding rectangle associated with the connected area selected from C satisfies A subset of connected regions , ; exist Extract the connected region with the largest area, and use the minimum circumscribed rectangle corresponding to the connected region with the largest area as The associated coding area is recorded as ; Repeat the above steps to get The inkjet printing area associated with all the paper-mounted inkjet printing images in The order of is sorted as follows: .

[0010] As a further solution of the present invention, in step 2, a grayscale image is generated. The associated binary image The specific method is: Extracting grayscale images All pixels in the segmentation pixel value threshold are sequentially compared Make a comparison; If the pixel value of any pixel is less than , then the pixel value of this pixel is assigned to 0, otherwise it is assigned to 255.

[0011] As a further solution of the present invention, in step 2, the position deviation analysis of the coding area is performed, and the specific method of determining the coding position state is as follows: Extract the determined mounting paper inkjet coding image and its associated coding area ; by The first pixel in the lower left corner is used as the origin of the two-dimensional coordinate system, the left boundary is used as the vertical axis, and the lower boundary is used as the horizontal axis to construct the two-dimensional coordinate system; Will Marking in a two-dimensional coordinate system; Obtain the standard coding area pre-constructed by the operator and map it in the constructed two-dimensional coordinate system; extract The intersection between the standard coding area and the total number of pixels in the intersection is counted and recorded as ; Get The total number of pixels in ; use get The total number of pixels in the difference between the standard coding area and the ; Get the total number of pixels in the standard inkjet printing area, recorded as ; use get The deviation rate of the coding area position from the standard coding area ; Get the position deviation rate threshold of the coding area preset by the operator ,like ,determination The inkjet printing position status in is abnormal, otherwise it is judged to be normal.

[0012] As a further solution of the present invention, in step 3, if it is determined that The corresponding image of the paper-mounted inkjet coding There are The status of each coding position and If the status of more than one inkjet printer position is normal, it is determined that the working status of the corresponding paper laminating machine during this evaluation period is normal, among which, Values ​​preset for the operator.

[0013] As a further solution of the present invention, in step 3, if it is determined that the number of paper-mounted inkjet images with normal inkjet position status within this evaluation period is less than , it is determined that the working status of the corresponding paper laminating machine in this evaluation cycle is abnormal, and the coding position correction parameters are generated in this evaluation cycle in combination with the pre-built standard coding area to calibrate the coding position in the next evaluation cycle.

[0014] As a further solution of the present invention, in step 3, the specific method of generating the coding position correction parameter is: Compare this assessment to the Coding area Perform fitting to obtain the coding area, which is recorded as ; Extract any constructed two-dimensional coordinate system, clear the coding area, and then And the standard coding area is mapped in it; Construction process The geometric center and parallel to Straight lines at the upper and lower boundaries ; Construction process Geometric center and parallel to Straight lines on the left and right boundaries ; and perpendicular to each other; Construct a straight line passing through the geometric center of the standard coding area and parallel to the upper and lower boundaries of the standard coding area ; Construct a straight line passing through the geometric center of the standard coding area and parallel to the left and right boundaries of the standard coding area ; and perpendicular to each other; Will Rotate to parallel, Rotate to Parallel, extract the rotation angle and record it as: rotation offset ; Will Adjust to the state with the largest intersection with the standard coding area and record During the adjustment process, the minimum distance of translation on the horizontal axis scale is recorded as the horizontal offset , the minimum distance of translation on the vertical axis scale, recorded as the longitudinal offset ; Based on the determined rotation offset , lateral offset , longitudinal offset Combining the mapping relationship between the mounted paper inkjet image and the physical object, the inkjet position correction parameter is obtained, which is expressed as: rotation offset parameter , lateral offset parameters , longitudinal offset parameters .

[0015] Beneficial effects of the present invention: (1) The present invention accurately analyzes the position deviation by constructing a two-dimensional coordinate system to compare the intersection of the coding area and the standard coding area, and can effectively determine whether the coding position status is abnormal; and through this operation, each mounted paper coding image can be processed according to the same standard and process, ensuring that the calibration result is objective and unified, and effectively avoiding interference from human factors; in addition, when the position deviation rate is greater than or equal to the preset threshold, it is quickly determined to be an abnormal state, thereby facilitating subsequent equipment adjustment actions in a timely manner and reducing the scrap rate caused by coding position errors; (2) The present invention effectively balances the detection frequency and system load by setting the evaluation period based on the average of the normal operating hours of multiple devices, making the setting of the evaluation period reasonable and consistent with the actual production rhythm, ensuring the efficiency and regularity of the subsequent calibration work; (3) The present invention realizes adaptive and operable inkjet position calibration through geometric analysis and spatial transformation, and solves the geometric relationship between the fitting area and the standard area in a two-dimensional coordinate system: the rotation offset is obtained by constructing and rotating the central axis, and the lateral offset and longitudinal offset are determined by finding the minimum translation distance. The actual spatial posture difference is quantified by combining the mapping relationship between the paper-mounted inkjet image and the real object, providing clear physical quantity guidance for the adjustment of the paper-mounted machine inkjet process. Such operation effectively reduces the waste and downtime caused by position deviation. More importantly, its predictive maintenance capability is to perform active calibration before the inkjet abnormality accumulates to a serious level through the position deviation rate threshold calibrated by the operator, which greatly ensures the continuity and stability of production. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below with reference to the accompanying drawings.

[0017] Figure 1 It is a schematic flow chart of the method of the present invention; Figure 2 Schematic diagram of the process of the method described in Example 2 of the present invention. DETAILED DESCRIPTION

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0019] Example 1 The paper laminating machine inkjet position calibration method based on visual recognition, such as Figure 1 As shown, the specific steps include: Before implementing this method, an evaluation cycle needs to be determined in advance, and subsequent operations are based on this evaluation cycle; The operator determines a number of laminating machines in normal operation (of the same model to avoid the influence of different models' working efficiency on the determination of the evaluation period), and then determines a value. , and record in real time the number of paper laminating machines that have been determined The time it takes to complete the coding operation on the mounting paper. So far, several time lengths can be obtained. By taking the average of the several time lengths obtained, a unique time interval is obtained, that is, the time point when the first mounting paper is processed is taken as the starting time point of the time interval, and the time point of the first mounting paper is taken as the starting time point of the time interval. The time point when the mounting paper processing is completed is used as the end time point of the time interval.

[0020] Next, extract the duration of this time interval and use it as the duration of an evaluation cycle, recorded as At this point, the duration of the evaluation cycle is determined. It should be noted that the operator determines the value Need to meet the passing value The determined evaluation cycle can represent the time window of the actual production rhythm and avoid randomness.

[0021] Extract the duration of the determined evaluation period and use the current time as the start time of the evaluation period, recorded as , by starting at An evaluation cycle is added on top of The end time of this evaluation cycle is obtained, which is recorded as ; At the beginning Until the end In the process, the operator uses the pre-built industrial camera to shoot the paper of any determined paper mounting machine after completing the coding operation on a number of paper mountings, and obtain a number of paper mounting coding images. The total number of paper mounting coding images is counted and recorded as , according to the shooting timeline, several paper-mounted inkjet images are recorded as a paper-mounted inkjet image sequence, which can be expressed as: (It should be noted that the light source and shooting position are fixed during the shooting process to eliminate shooting errors); It should be explained here that the mapping relationship between the mounted paper inkjet coding image sequence and the real object needs to be determined by the operator based on the actual situation. This step can be achieved by photographing a calibration plate of known size and placing it at the mounting position to obtain a conversion matrix from pixel coordinates to physical coordinates. Without the calibration of the mapping relationship between the mounted paper inkjet coding image sequence and the real object, the subsequent position deviation analysis will lose its physical meaning.

[0022] Based on the paper-mounted inkjet coding image sequence obtained in the above steps , from which the first mounted paper inkjet image Start by coding the entire mounting paper image sequence During the traversal, the following operations are performed on each mounted paper inkjet coding image: Extract the coding areas from all paper-mounting coding image sequences and perform fitting to obtain a unique coding area associated with the paper-mounting machine determined within this evaluation cycle (based on the deviation of all samples (coding areas) within the entire evaluation cycle, comprehensively judge the overall status of the coding position in this cycle). Perform position deviation analysis on the coding area and compare it with the threshold preset by the operator (quantify the degree of deviation between the actual position of the current coding area and the expected standard position) to judge the coding position status. If the coding position status is determined to be abnormal during this process, generate coding position correction parameters based on the pre-constructed standard coding area within this evaluation cycle, and calibrate the coding position of the paper-mounting machine in the next evaluation cycle (this calibration process is continuous, that is, real-time monitoring and real-time calibration); It should be explained here that the analysis of the current evaluation cycle is based on the state without applying the new parameters. Immediate application may affect the evaluation results of the current cycle. Therefore, it is determined that the adjustment should be made within one evaluation cycle. By evaluating the deviations of all samples (coding areas) within a cycle, the overall status of the coding position during that cycle is comprehensively judged. This measure can effectively avoid accidental fluctuations caused by abnormalities in individual mounting papers, allowing the overall solution to focus on systematic deviations. The coding position correction parameters described in the above content are quantified results, that is, they need to be implemented in combination with the determined mounting paper coding image sequence and the mapping relationship between the physical object. In addition, the above content also involves the standard coding area. Similarly, this standard coding area needs to be calibrated in combination with the pixel position in the mounting paper coding image. Because the same model of industrial camera is used (and the position is fixed), the pixel size captured is fixed, and the calibrated area is also fixed relative to the entire mounting paper coding image. Therefore, the standard coding area can be obtained by summarizing the calibrated pixel points.

[0023] Example 2 This embodiment discloses a method for extracting a coding area from a mounted paper coding image and determining the coding position state based on embodiment 1. Figure 2 As shown, the specific steps include: The determined paper-mounting coding image sequence can be obtained through the content described in Example 1 First, you need to print the image sequence from the paper Extract the inkjet coding area, this embodiment uses the paper-mounted inkjet coding image sequence Take any paper-mounted inkjet coding image in as an example, and perform the following example processing, the paper-mounted inkjet coding image sequence The rest of the mounted paper inkjet coding images are processed according to the following method: Mark any determined paper-mounted coding image as: ,in, is a counting index, ranging from 1 to ; At this time, the determined paper-mounted inkjet image For color images, grayscale processing is performed on them, and the color image is converted into a grayscale image to obtain a grayscale image. For the convenience of distinction, it is marked as ; Get the segmentation pixel value threshold preset by the operator based on the pixel value of the inkjet printer determined by the paper laminating machine (It should be explained here that in the actual inkjet coding process, the inkjet code is generally information such as the production date, which needs to be clearly contrasted with the mounting paper. Therefore, the inkjet code is generally black, and the pixel value of black is 0. Then the operator can combine this pixel value 0 to preset a segmentation pixel value threshold. In this way, the coded image can be effectively segmented from the entire mounting paper; however, there may be black pixels in other areas of the mounting paper, so subsequent processing is required); At this time, based on the determined segmentation pixel value threshold , from the grayscale image Extract all the pixels and compare them with the segmentation pixel value threshold in turn Perform a comparison operation. If the pixel value of any pixel among the extracted pixels is greater than or equal to the segmentation pixel value threshold, , then the pixel value of this pixel is assigned to 255; If the pixel value of any pixel among the extracted pixels is less than the segmentation pixel value threshold , then the pixel value of this pixel is assigned to 0; At this point, only the binary image of the inkjet code is saved. , wherein the binary image The binary value in refers to the pixel value 225 and the pixel value 0, that is, the pixel value of the inkjet code is 255, and the pixel values ​​of the other pixels are all 0; Extract all connected areas with pixel values ​​of 225, that is, binarize the image The connected pixels with a pixel value of 255 are aggregated; Based on the characteristics of inkjet printing, the connected areas are filtered and all connected areas are extracted for character recognition (this technology is existing and will not be further studied in this solution), and the connected areas without characters are eliminated; Next, the format of the connected area with characters is identified. The format in this format identification is the format of the inkjet code (for example: date format , specifically preset by the operator), if the format of the characters in the connected area conforms to the format preset by the operator, it is marked as a valid coding area, and the connected area corresponding to the valid coding area is retained; if the format of the characters in the connected area does not conform to the format preset by the operator, the corresponding connected area is removed; Through the above process, the final 1 to several connected regions are combined (usually there is only one) to obtain the connected region set C, which is expressed as: ,in, is the total number of connected regions in the connected region set C; Confirm The minimum bounding rectangles of the connected regions are obtained by the minimum bounding rectangle set R, which is expressed as: ; Next, you need to obtain the area range of the coding area preset by the operator , the area of ​​this coding area It needs to be determined in combination with the standard coding area described in Example 1. Generally, 90% and 110% of the area of ​​the standard coding area are taken as the minimum and maximum values ​​in the coding area range, respectively. The final determination of the specific value still needs to be determined by the operator based on the actual situation. Different coding requirements correspond to different values.

[0024] Then the minimum circumscribed rectangles associated with all connected regions in the connected region set C are respectively compared with the area interval of the coding area. Compare and select the area range that meets the coding area requirements A subset of connected regions ,in, .

[0025] In the determined connected region subset Extract the connected area with the largest area, and use the corresponding minimum circumscribed rectangle as the paper-mounting coding image The associated coding area is recorded as ; What needs to be explained here is that the connected area subset The largest connected region in the image is used as the mounting inkjet image For the associated coding areas, larger connected areas are more likely to contain complete information, while smaller areas may only contain local information fragments or noise (generally, character recognition can determine a valid coding information, and in actual application, the paper cover has only one confirmed coding. The reason for this design of the solution is to provide a backup for abnormal situations, preventing the occurrence of abnormal situations during the coding determination process, which would cause the entire method to crash during actual operation and hinder the normal operation of subsequent processes); So far, the paper-mounted inkjet coding image has been determined. Associated coding area , and so on, for the mounting paper inkjet image sequence All the paper-mounted inkjet coding images are processed to obtain the inkjet coding areas associated with all the paper-mounted inkjet coding images, and the obtained inkjet coding areas are sorted according to the paper-mounted inkjet coding image sequence. The order of is sorted as follows: .

[0026] Next, based on the determined Coding area: , perform position deviation analysis and determine the inkjet position status; Still based on the determined mounting paper inkjet image and its associated coding area Perform sample processing; Printing images on mounting paper The first pixel point in the lower left corner is used as the origin of the two-dimensional coordinate system, the left boundary is used as the vertical axis of the two-dimensional coordinate system, and the lower boundary is used as the horizontal axis of the two-dimensional coordinate system to construct the paper-mounted inkjet printing image. The associated two-dimensional coordinate system; Print the image on the mounting paper Associated coding area Highlight in this two-dimensional coordinate system, that is, the coding area Printing image on mounting paper The pixels in the image are highlighted. Next, obtain the operator's printing image for paper mounting A pre-constructed standard coding area (the position of the standard coding area compared to the mounting coding image has been explained in Example 1) is mapped in the constructed two-dimensional coordinate system; At this point, the coding area in the two-dimensional coordinate system can be obtained And standard inkjet area, extract inkjet area And the intersection between the standard coding area, and count the total number of pixels in the intersection, marked as: ; Get the coding area again The total number of pixels in , marked as: ; Next, use Calculate the coding area And the total number of pixels in the difference between the standard coding area, marked as: ; Then get the total number of pixels in the standard coding area, marked as: ; Continue to use: Calculate the paper-mounted inkjet printing image Coding area The deviation rate of the coding area position from the standard coding area ; Then obtain the position deviation rate threshold of the coding area preset by the operator , the coding area position deviation rate obtained in the above calculation steps is Position deviation rate threshold from the coding area Make a comparison; If the position deviation rate of the coding area is determined Greater than or equal to the coding area position deviation rate threshold , then the inkjet printing position status is determined to be abnormal; If the position deviation rate of the coding area is determined Less than the position deviation rate threshold of the coding area , it is determined that the inkjet printing position status is normal.

[0027] The purpose of this embodiment is to accurately locate the effective coding area in the mounted paper coding image, and then determine whether the coding position status is normal through position deviation analysis. First, a single image in the mounted paper coding image sequence is processed, and a series of operations such as grayscale processing, binarization processing based on segmentation pixel value thresholds, connected area extraction and filtering are performed to finally determine the coding area. Then, all images in the image sequence are processed in this way, and arranged in order to obtain a coding area sequence, and then position deviation analysis is performed. By constructing a two-dimensional coordinate system, the pixel points and related data of the coding area and the standard coding area are compared, and the position deviation rate of the coding area is calculated, and compared with the preset threshold to determine the coding position status.

[0028] Example 3 This embodiment further discloses a method for generating correction parameters and calibrating a coding position whose coding position state is abnormal based on Embodiment 1 and Embodiment 2, which specifically includes the following steps: Based on the content described in Example 2, it can be determined that The inkjet position status of each mounted paper inkjet image is associated with the inkjet position status of each mounted paper inkjet image. If the inkjet position status of each mounted paper inkjet image is associated with the inkjet position status of each mounted paper inkjet image, if ... There are The status of the inkjet printing position or If the status of more than one inkjet printer position is normal, it can be determined that the corresponding paper laminating machine is working normally during this evaluation period, and no correction operation is required, that is, no processing is performed. The value is preset by the operator based on the actual situation; If within an evaluation cycle Among the mounting paper inkjet coding images, the number of mounting paper inkjet coding images associated with the normal inkjet position status is less than , then it is determined that the working status of the corresponding paper laminating machine is abnormal during this evaluation period, and the coding position correction parameters are generated in this evaluation period in combination with the standard coding area pre-built by the operator. The specific steps are as follows: Compare this assessment to the Coding area Perform fitting (together with the mounted paper inkjet coding image within this evaluation period. The fitting technology is an existing technology and will not be further studied in this solution). Finally, a unique coding area is obtained, which is marked as , it should be noted that, Subscript in It only represents the only identified coding area and has no other special meaning; Then, any one of the two-dimensional coordinate systems constructed in Example 2 is extracted, and the annotated or mapped content therein is cleared, leaving only the two-dimensional coordinate system itself; The coding area The standard coding area is mapped separately in the cleared two-dimensional coordinate system (this two-dimensional coordinate system includes the two-dimensional coordinate system associated with the mounting coding image. In Example 1, it is determined that the pixel size of the mounting coding image is unique, and the relative position of the standard coding area is fixed compared to the mounting coding image. Therefore, the standard coding area can be mapped in the two-dimensional coordinate system associated with the mounting coding image). Extract coding areas separately The geometric center of the standard coding area; For the coding area : Construct a line through the coding area The geometric center of the inkjet printing area The straight lines of the upper and lower boundaries are denoted as ; Then build a line through the coding area The geometric center of the inkjet printing area The straight lines on the left and right boundaries are denoted as , and perpendicular to each other; Similarly, for the standard coding area: construct a straight line passing through the geometric center of the standard coding area and parallel to the upper and lower boundaries of the standard coding area, denoted as ; Then construct a straight line passing through the geometric center of the standard coding area and parallel to the left and right boundaries of the standard coding area, recorded as , and perpendicular to each other; The straight line Rotate to the line Parallel, straight line Rotate to the line Parallel, and record the rotation angle at this time (it is necessary to make sure that the rotation angle is the minimum, for example, after rotating 15 degrees, the straight line Rotate to the line Parallel, straight line Rotate to the line Parallel, straight line after rotating 375 degrees Rotate to the line Parallel, straight line Rotate to the line Parallel, then take 15 degrees as the rotation offset ), marked as the rotation offset ; Then set the coding area Adjust to the state with the largest intersection with the standard coding area, that is, the coding area The part that overlaps with the standard inkjet printing area is the largest (this step requires determining the rotation offset After that), record the minimum distance of translation on the horizontal axis scale and the minimum distance of translation on the vertical axis scale, and set the coding area The minimum distance of translation on the horizontal axis scale is recorded as the horizontal offset , coding area The minimum distance of translation on the vertical axis scale is recorded as the longitudinal offset ; Summarize the resulting rotation offsets , lateral offset , longitudinal offset , through the mapping relationship between the mounted paper inkjet coding image and the real object determined in Example 1, the rotation offset in reality is obtained , lateral offset , longitudinal offset The corresponding values ​​are recorded as: rotation offset parameters , lateral offset parameters , longitudinal offset parameters ; At this point, the obtained rotation offset parameters , lateral offset parameters , longitudinal offset parameters It is the inkjet position correction parameter for correcting the inkjet position of the paper laminating machine, based on the rotation offset parameter , lateral offset parameters , longitudinal offset parameters It is to calibrate the inkjet coding position of the paper laminating machine in the next evaluation cycle.

[0029] Some of the data in the formulas described above are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0030] The above contents are merely examples and explanations of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in similar ways. As long as they do not deviate from the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

[0031] It is important to note that all user data collected in this application is collected with the user's consent and authorization. Furthermore, the use of user data is legal and compliant, and the use and processing of user data complies with the relevant laws, regulations, and standards of the relevant regions.

Claims

1. A paper laminating machine inkjet position calibration method based on visual recognition, characterized in that: This method includes the following: Step 1: Determine an evaluation period. During the current evaluation period, use a pre-built industrial camera to photograph a number of mounted papers that have completed coding, obtaining a number of mounted paper coding images. The number of mounted paper coding images are recorded as a mounted paper coding image sequence according to the shooting timeline. Step 2: traverse the paper-mounted coding image sequence, extract the coding area in the paper-mounted coding image, perform position deviation analysis on the coding area, and determine the coding position status; Step 3: If the inkjet position status is determined to be abnormal, the inkjet position correction parameters are generated in this evaluation cycle in combination with the pre-built standard inkjet area, and the inkjet position in the next evaluation cycle is calibrated.

2. The method for calibrating the inkjet printer position of a paper laminating machine based on visual recognition according to claim 1, characterized in that: In step 1, the specific method of determining the evaluation period is: Determine a number of normal laminating machines and record the number of normal laminating machines The time it takes to complete the coding of the mounting paper is several times, among which, Preset values ​​for operators; Take the average of several durations as the duration of an evaluation cycle, recorded as .

3. The method for calibrating the inkjet printer position of a paper laminating machine based on visual recognition according to claim 2, characterized in that: In the step 1, the specific method of recording a plurality of mounting inkjet coding images as a mounting inkjet coding image sequence is as follows: The current moment is taken as the starting moment of the evaluation cycle, recorded as ; exist The evaluation cycle is superimposed on the The duration of ; Leverage pre-built industrial cameras to to The mounted paper after coding is photographed, and the total number of the photographed mounted paper coding images is recorded as ; In the order of shooting The paper-mounted inkjet coding images are recorded as: paper-mounted inkjet coding image sequence ; Among them, the mapping relationship between the mounted paper inkjet coding image and the real object is determined by the operator.

4. The method for calibrating the inkjet printer position of a paper laminating machine based on visual recognition according to claim 1, characterized in that: In the step 2, the specific method of extracting the inkjet coding area in the mounted paper inkjet coding image is as follows: Determine the mounting coding image sequence Any mounted paper inkjet image ,in, is the counting index, ; right Perform grayscale processing to obtain a grayscale image ; Get the segmentation pixel value threshold preset by the operator ; based on Generate grayscale image The associated binary image ; Extract binary image All connected areas with pixel values ​​of 225 are summarized as the connected area set ,in, is the total number of connected regions in C; Build The minimum bounding rectangle of each connected area is the set of minimum bounding rectangles ; Get the area range of the coding area preset by the operator ; The area of ​​the minimum bounding rectangle associated with the connected area selected from C satisfies A subset of connected regions , ; exist Extract the connected region with the largest area, and use the minimum circumscribed rectangle corresponding to the connected region with the largest area as The associated coding area is recorded as ; Repeat the above steps to get The inkjet printing area associated with all the paper-mounted inkjet printing images in The order of is sorted as follows: .

5. The method for calibrating the inkjet printer position of a paper laminating machine based on visual recognition according to claim 4, characterized in that: In the step 2, a grayscale image is generated The associated binary image The specific method is: Extracting grayscale images All pixels in the segmentation pixel value threshold are sequentially compared Make a comparison; If the pixel value of any pixel is less than , then the pixel value of this pixel is assigned to 0, otherwise it is assigned to 255.

6. The method for calibrating the inkjet printer position of a paper laminating machine based on visual recognition according to claim 4, characterized in that: In the second step, the position deviation analysis of the coding area is performed, and the specific method of determining the coding position status is as follows: Extract the determined mounting paper inkjet coding image and its associated coding area ; by The first pixel in the lower left corner is used as the origin of the two-dimensional coordinate system, the left boundary is used as the vertical axis, and the lower boundary is used as the horizontal axis to construct the two-dimensional coordinate system; Will Marking in a two-dimensional coordinate system; Obtain the standard coding area pre-constructed by the operator and map it in the constructed two-dimensional coordinate system; extract The intersection between the standard coding area and the total number of pixels in the intersection is counted and recorded as ; Get The total number of pixels in ; use get The total number of pixels in the difference between the standard coding area and the ; Get the total number of pixels in the standard coding area and record it as ; use get The deviation rate of the coding area position from the standard coding area ; Get the position deviation rate threshold of the coding area preset by the operator ,like ,determination The inkjet printing position status in is abnormal, otherwise it is judged to be normal.

7. The method for calibrating the inkjet printer position of a paper laminating machine based on visual recognition according to claim 6, characterized in that: In step 3, if it is determined that The corresponding image of the paper-mounted inkjet coding There are The status of each coding position and If the status of more than one inkjet printer position is normal, it is determined that the working status of the corresponding paper laminating machine during this evaluation period is normal, among which, Values ​​preset for the operator.

8. The method for calibrating the inkjet printer position of a paper laminating machine based on visual recognition according to claim 6, characterized in that: In the step 3, if it is determined that the number of paper-mounted inkjet images with normal inkjet position status within this evaluation period is less than , it is determined that the working status of the corresponding paper laminating machine in this evaluation cycle is abnormal, and the coding position correction parameters are generated in this evaluation cycle in combination with the pre-built standard coding area to calibrate the coding position in the next evaluation cycle.

9. The method for calibrating the inkjet printer position of a paper laminating machine based on visual recognition according to claim 8, characterized in that: In step 3, the specific method of generating the coding position correction parameter is as follows: Compare this assessment to the Coding area Perform fitting to obtain the coding area, which is recorded as ; Extract any constructed two-dimensional coordinate system, clear the coding area, and then And the standard coding area is mapped in it; Construction process Geometric center and parallel to Straight lines at the upper and lower boundaries ; Construction process Geometric center and parallel to Straight lines on the left and right boundaries ; and perpendicular to each other; Construct a straight line passing through the geometric center of the standard coding area and parallel to the upper and lower boundaries of the standard coding area ; Construct a straight line passing through the geometric center of the standard coding area and parallel to the left and right boundaries of the standard coding area ; and perpendicular to each other; Will Rotate to parallel, Rotate to Parallel, extract the rotation angle and record it as: rotation offset ; Will Adjust to the state with the largest intersection with the standard coding area and record During the adjustment process, the minimum distance of translation on the horizontal axis scale is recorded as the horizontal offset , the minimum distance of translation on the vertical axis scale, recorded as the longitudinal offset ; Based on the determined rotation offset , lateral offset , longitudinal offset Combining the mapping relationship between the mounted paper inkjet image and the physical object, the inkjet position correction parameter is obtained, which is expressed as: rotation offset parameter , lateral offset parameters , longitudinal offset parameters .

Images