A packaging process body production detection system and method

Abstract

The present invention relates to the technical field of production of package process forms, and in particular to a system and method for detecting production of package process forms, comprising an image acquisition module; an image processing and analysis module, the image processing and analysis module is used to receive package process form image data transmitted by the image acquisition module, and perform noise reduction and enhancement processing on the image data, and then analyze the processed data to determine whether the package process form has quality problems; a dimension measurement module; and a result display and alarm module. The present invention uses a high-resolution camera and image processing software to quickly and accurately detect the appearance of package products through the cooperation of the image acquisition module, the image processing and analysis module, the dimension measurement module and the result display and alarm module, and can automatically identify shape deviations, surface flaws, color differences, pattern defects and other problems, thereby improving detection efficiency and accuracy and reducing the influence of human factors.

Description

Technical Field

[0001] The present invention belongs to the technical field of bag process body production, and in particular relates to a bag process body production detection system and method. Background Art

[0002] Bag craft forms are the three-dimensional shapes of bags produced through a series of technological processes according to design requirements and uses; bag craft forms include handbags, backpacks, wallets and other types, each type has its own specific shape, structure and function; the production of craft forms involves multiple links such as material selection, cutting, sewing, and assembly, and each link requires delicate operation and strict quality control.

[0003] At present, the quality inspection method for bag craft body is usually to use appearance inspection, that is, the inspector directly observes the overall shape of the bag with the naked eye to check whether its shape meets the design requirements and whether there are any deformation, distortion and other problems; check the surface quality, including whether the color of the fabric is uniform, whether there is any color difference, flaws, stains, whether the hardware accessories have scratches, rust, etc.; check whether the stitching is uniform and tight, and whether there is any loose thread, jump thread and other phenomena.

[0004] However, manual inspection methods such as visual inspection are greatly affected by the inspectors' experience, skills, fatigue level and subjective judgment criteria. Different inspectors may have different judgment results for the same defect, which makes it difficult to ensure the consistency and accuracy of the inspection results. Therefore, we need to propose a package process body production inspection system and method to solve the above-mentioned problems, so that it can effectively improve the inspection efficiency and accuracy and reduce the impact of human factors on its results. Summary of the invention

[0005] In view of the above problems, the present invention provides a packaging process body production detection system, including an image acquisition module, which captures the image of the packaging process body through a high-definition camera and outputs a video signal to an image processing and analysis module;

[0006] An image processing and analysis module, which is used to receive the image data of the package process body transmitted by the image acquisition module, perform noise reduction and enhancement processing on the image data, and then analyze the processed data to determine whether there is any quality problem with the package process body;

[0007] The image processing and analysis module includes a preprocessing unit for preprocessing image data, an identification unit for detecting deformation and distortion of the preprocessed data, a color comparison unit for detecting color difference of the preprocessed data, a suture detection unit for detecting sutures of the preprocessed data, and an accessory detection unit for detecting hardware accessories in the preprocessed data, wherein the preprocessing unit is electrically connected to the identification unit, the color comparison unit, the suture detection unit, and the accessory detection unit, respectively;

[0008] A dimension measurement module is used to measure the dimensions of the package process body and then analyze the measured data to determine whether the dimensions meet the quality requirements;

[0009] The result display and alarm module is used to receive the detection result information of the image processing and analysis module and the dimension measurement module, and then display the result according to the detection result. At the same time, when the quality of the package process body is unqualified, an audible and visual alarm is issued to remind the staff to deal with it in time.

[0010] Furthermore, the image acquisition module acquires images of the packaging process body as follows:

[0011] A1. Calculate the light intensity of the lighting equipment according to the shooting requirements of the high-definition camera. The light intensity calculation formula is as follows:

[0012] Where E is the light intensity, I is the luminous intensity of the light source, and r is the distance from the illuminated surface to the light source. The light intensity is adjusted by changing the light source distance r, so that the lighting equipment can provide a uniform, sufficient and shadow-free light environment for the high-definition camera.

[0013] A2. By adjusting the position and angle of the shooting bracket, the high-definition camera can fully cover all parts of the package to obtain image information from different perspectives;

[0014] A3. Obtain the package body image under the corresponding viewing angle by taking the package body image information captured by the high-definition camera in real time;

[0015] A4. Transmit the package body image to the image processing and analysis module in real time.

[0016] Furthermore, the process of the preprocessing unit preprocessing the image data is as follows:

[0017] B11. Receive the appearance image data of the multi-view packaging process body from the image acquisition module, and use the noise reduction formula to perform noise reduction processing on the appearance image data. The noise reduction formula is as follows:

[0018] Where k and l are the values ​​of the window size, m and n are the length and width of the window size, I(x+i,y+j) is the grayscale value of the pixel with coordinates (x+i,y+j) in the neighborhood, and G(x,y) is the grayscale value of the image after denoising.

[0019] B12. Confirm the gray level i of the gray value G(x,y) of the image after denoising, and calculate the weight value Wi of the gray level i. The calculation formula of the weight value Wi is as follows:

[0020] Where N is the total number of pixels in the image, n i is the number of pixels of image gray level i;

[0021] B13. Map the gray level i to obtain a new gray level j. The mapping formula is as follows:

[0022] j=(L)*∑ i＝0 Wi, where L is the maximum value of the grayscale range;

[0023] B14. Calculate the grayscale value corresponding to the new grayscale level j according to the grayscale range of the original image. The calculation formula is as follows:

[0024] Where f(x,y) is the grayscale value corresponding to the new grayscale j, a is the minimum value of the grayscale range of the original image, L is the maximum value of the grayscale range of the original image, G(x,y) is the grayscale value of the original image, and M is the number of grayscale levels;

[0025] B15. Perform linear contrast stretching on the image and calculate the grayscale value after stretching. The calculation formula is as follows:

[0026] Among them, c and d are the minimum and maximum values ​​of the grayscale range after stretching, a and L are the minimum and maximum values ​​of the grayscale range of the original image, f(x, y) is the image grayscale value corresponding to the new grayscale level j, and F(x, y) is the grayscale value of the image after stretching.

[0027] Furthermore, the process of the recognition unit performing deformation and distortion detection on the image is as follows:

[0028] B21. Sum the grayscale values ​​of all pixels in the image to extract the geometric features of the image. The summation formula is as follows:

[0029] M0=∑ x ∑ y F(x,y), where M0 is the sum of the grayscale values ​​of all pixels in the image, and F(x,y) is the grayscale value of the preprocessed image;

[0030] B22. Determine two feature vectors e and g according to the extracted actual geometric features, where e = (e1, e2, ..., en), g = (g1, g2, ..., gn);

[0031] B23. Calculate the distance D(e,g) between two eigenvectors. The calculation formula of distance D(e,g) is as follows:

[0032] D(e,g)=∑ n＝1 |en-gn|;

[0033] B24. Compare the calculated distance D(e,g) with the set threshold, identify the features exceeding the threshold as shape anomalies, and identify the features within the threshold range as normal.

[0034] Furthermore, the process of color difference detection by the color contrast unit is as follows:

[0035] B31. Convert the preprocessed image data into CIE Lab color space. The conversion formula is as follows:

[0036] Among them, Y n is the stimulus value of the reference white point; Y is the brightness corresponding to the gray value F(x,y), L * is the brightness value in the CIE Lab color space;

[0037] B32. Select different locations of the fabric as sampling areas, and evenly select several sampling points in each area to represent the color of the area;

[0038] B33. Extract the corresponding color feature value at the selected sampling point and obtain its L in the CIE Lab color space. * 、a * and b * The value of

[0039] B34. Calculate the color difference between different sampling points. The color difference calculation formula is as follows:

[0040] in, and L * The two color values ​​of the channel, and They are respectively * The two color values ​​of the channel, and b * Two color values ​​of the channel, ΔE is the color difference between the two sampling points;

[0041] B35. Compare the calculated color difference value with the standard color difference threshold to determine whether there is a color difference problem. If the calculated color difference value exceeds the standard color difference threshold, it is determined that there is color difference at the position. If the calculated color difference value is within the standard color difference threshold range, it is determined that there is no color difference at the position.

[0042] Furthermore, the process of the suture detection unit detecting the suture is as follows:

[0043] B41. Filter the preprocessed data. The filter processing formula is as follows:

[0044] Y n =F(x,y)*d, where d is the filter coefficient, Y n is the gray value of the output image, and F(x,y) is the gray value of the image after preprocessing;

[0045] B42. Perform edge detection algorithm on the filtered image to extract possible edge information of the seam in the image. The edge calculation formula is as follows:

[0046]

[0047] Among them, S x is the horizontal gradient, S y is the vertical gradient, G c is the gradient amplitude, θ is the gradient direction;

[0048] B43. The suture width is obtained by calculating the distance between the two sides of the contour according to the extracted edge information, the suture length is obtained by calculating the length along the contour path, the suture density is obtained by counting the number of sutures per unit area, and the continuity of the suture contour, whether there is abnormal interruption or branching, is observed to obtain the characteristics of the suture;

[0049] B44. Compare the suture characteristics with the standard suture characteristics to confirm whether the suture meets the quality requirements. If it meets the standard suture characteristics, it is judged to be normal. If it exceeds the standard suture characteristics, it is judged to be abnormal.

[0050] Furthermore, the accessory detection unit performs the following process on the accessory detection:

[0051] B51. Use the target detection algorithm to identify the location of the hardware accessories in the package image from the preprocessed image, and divide the area containing the hardware accessories;

[0052] B52. Calculate the area Area and the perimeter Per of the contour according to the divided areas, and use the area Area and the perimeter Per to calculate the circularity C of the area. The calculation formula is as follows:

[0053]

[0054] B53. Compare the calculated circularity C with the circularity of the standard accessory. When the circularity C does not conform to the circularity of the standard accessory, the hardware accessory is judged to be abnormal. When the circularity C conforms to the circularity of the standard accessory, the hardware accessory is judged to be normal.

[0055] Furthermore, the size measurement module includes a laser distance sensor for size measurement and a data processor for analyzing and processing the measurement data. When measuring the size, the laser distance sensor selects a relatively fixed and easily identifiable part of the package as a reference point, and then measures different positions of the package from the reference point to obtain the size characteristics of the package;

[0056] The process of analyzing and processing the package size characteristics by the data processor is as follows:

[0057] C1. Set the standard range of each size according to the package process design requirements and quality standards:

[0058] C2. Calculate the deviation between the actual measured size and the standard size, and determine whether the deviation is within the tolerance range. If the deviation is within the tolerance range, the package size is normal; if the deviation exceeds the tolerance range, the package size is abnormal.

[0059] Furthermore, the process of displaying the results by the display and alarm module is as follows:

[0060] D1. Establish appearance quality inspection result table and dimension inspection result table respectively according to the analysis categories of image processing and analysis module and dimension measurement module, and set the standard range of qualified display in appearance quality inspection result table and dimension inspection result table respectively;

[0061] D2. Disassemble the received data information according to different test items and categories, and then classify the parsed information into the appearance quality test result table and the size test result table respectively;

[0062] D3. When the values ​​input into the appearance quality test result table and the size test result table are within the standard range, they will be displayed as qualified; when one of the values ​​input into the appearance quality test result table and the size test result table does not meet the standard range, it will be displayed as unqualified, and the unqualified abnormal content will be displayed and the sound and light alarm will be performed synchronously.

[0063] Based on the above-described system for detecting the production of a package process type body, the present invention further provides a method for detecting the production of a package process type body, comprising the following steps:

[0064] S1. Use a high-definition camera to capture the image of the packaging process body, and transmit the captured image data to the image processing and analysis module in the form of a video signal;

[0065] S2, the image processing and analysis module receives the video signal and converts it into digital image data;

[0066] S3, performing noise reduction processing on the received image data, while enhancing the contrast and clarity of the image;

[0067] S4, analyzing the processed image data, and then judging whether there is any quality problem in the package process body through an algorithm;

[0068] S5. Measure the dimensions of the package process body through the dimension measurement module, analyze the measured data, and determine whether the dimensions meet the quality requirements;

[0069] S6, the display and alarm module receives the detection result information of the image analysis and analysis module and the dimension measurement module, and displays the result according to the detection result. The displayed result includes the specific information of the quality problem and the dimension problem. When the quality of the package process body is unqualified or the dimension does not meet the requirements, the sound and light alarm is triggered to remind the staff to deal with it in time;

[0070] S7. Record the problems found during the inspection process, including the type, location, and severity of the problem.

[0071] The beneficial effects of the present invention are:

[0072] 1. The present invention uses a high-resolution camera and image processing software to quickly and accurately detect the appearance of package products through the cooperation of an image acquisition module, an image processing and analysis module, a size measurement module, and a result display and alarm module. It can automatically identify problems such as shape deviations, surface flaws, color differences, and pattern defects, thereby improving detection efficiency and accuracy and reducing the impact of human factors.

[0073] 2. The present invention can more accurately identify quality problems and dimensional deviations of package process shapes and reduce the misjudgment rate by setting up image processing and analysis modules and combining noise reduction and contrast enhancement technologies.

[0074] 3. The image acquisition module of the present invention realizes comprehensive and clear image acquisition of the packaging process body through the coordinated work of components such as a high-definition camera, a shooting bracket and lighting equipment.

[0075] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0077] Figure 1 A block diagram of a detection system according to an embodiment of the present invention is shown;

[0078] Figure 2 It shows a flowchart of the image acquisition module for capturing images of a package-like process body according to an embodiment of the present invention;

[0079] Figure 3 A flowchart showing the preprocessing unit preprocessing image data according to an embodiment of the present invention is shown;

[0080] Figure 4 A flowchart of deformation and distortion detection of an image by a recognition unit according to an embodiment of the present invention is shown;

[0081] Figure 5 A flowchart of a suture detection unit detecting sutures according to an embodiment of the present invention is shown;

[0082] Figure 6 A flow chart of a detection method according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0083] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0084] The embodiment of the present invention provides a packaging process type body production detection system, such as Figure 1-5As shown, it includes an image acquisition module, an image processing and analysis module, a dimension measurement module and a result display and alarm module. The image acquisition module captures the image of the packaging process body through a high-definition camera and outputs a video signal to the image processing and analysis module;

[0085] The image acquisition module includes a high-definition camera, an adjustable shooting bracket and lighting equipment. The high-definition camera has a sufficiently high resolution and can clearly capture various details of the package process shape. The high-definition camera is installed on the shooting bracket, and the shooting bracket can flexibly adjust the angle and height to ensure that the high-definition camera can shoot the package from different directions. The lighting equipment is installed on the shooting bracket, and the lighting equipment provides a uniform, sufficient and shadow-free lighting environment to avoid image quality being affected by light problems.

[0086] like Figure 2 As shown, the image acquisition module acquires images of the packaging process body as follows:

[0087] A1. Calculate the light intensity of the lighting equipment according to the shooting requirements of the high-definition camera. The light intensity calculation formula is as follows:

[0088] Where E is the light intensity, I is the luminous intensity of the light source, and r is the distance from the illuminated surface to the light source. The light intensity is adjusted by changing the light source distance r, so that the lighting equipment can provide a uniform, sufficient and shadow-free light environment for the high-definition camera.

[0089] A2. By adjusting the position and angle of the shooting bracket, the high-definition camera can fully cover all parts of the package to obtain image information from different perspectives;

[0090] A3. Obtain the package body image under the corresponding viewing angle by taking the package body image information captured by the high-definition camera in real time;

[0091] A4. Transmit the package body image to the image processing and analysis module in real time to provide comprehensive and clear appearance image data for subsequent analysis and processing;

[0092] The image acquisition module realizes comprehensive and clear image acquisition of the packaging process shape through the coordinated work of components such as high-definition cameras, shooting brackets and lighting equipment.

[0093] The image processing and analysis module is used to receive the image data of the package process body transmitted by the image acquisition module, perform noise reduction and enhancement processing on the image data, and then analyze the processed data to determine whether there is a quality problem with the package process body;

[0094] The image processing and analysis module includes a preprocessing unit for preprocessing image data, an identification unit for detecting deformation and distortion of the preprocessed data, a color comparison unit for detecting color difference of the preprocessed data, a suture detection unit for detecting sutures of the preprocessed data, and an accessory detection unit for detecting hardware accessories in the preprocessed data, wherein the preprocessing unit is electrically connected to the identification unit, the color comparison unit, the suture detection unit, and the accessory detection unit, respectively;

[0095] like Figure 3 As shown, the process of the preprocessing unit preprocessing the image data is as follows:

[0096] B11. Receive the appearance image data of the multi-view packaging process body from the image acquisition module, and use the noise reduction formula to perform noise reduction processing on the appearance image data. The noise reduction formula is as follows:

[0097] Where k and l are the values ​​of the window size, m and n are the length and width of the window size, I(x+i,y+j) is the grayscale value of the pixel with coordinates (x+i,y+j) in the neighborhood, and G(x,y) is the grayscale value of the image after denoising.

[0098] B12. Confirm the gray level i of the gray value G(x,y) of the image after denoising, and calculate the weight value Wi of the gray level i. The calculation formula of the weight value Wi is as follows:

[0099] Where N is the total number of pixels in the image, n i is the number of pixels of image gray level i;

[0100] B13. Map the gray level i to obtain a new gray level j. The mapping formula is as follows:

[0101] j=(L)*∑ i＝0 Wi, where L is the maximum value of the grayscale range;

[0102] B14. Calculate the grayscale value corresponding to the new grayscale level j according to the grayscale range of the original image. The calculation formula is as follows:

[0103] Where f(x,y) is the grayscale value corresponding to the new grayscale j, a is the minimum value of the grayscale range of the original image, L is the maximum value of the grayscale range of the original image, G(x,y) is the grayscale value of the original image, and M is the number of grayscale levels;

[0104] B15. Perform linear contrast stretching on the image and calculate the grayscale value after stretching. The calculation formula is as follows:

[0105] Where c and d are the minimum and maximum values ​​of the grayscale range after stretching, a and L are the minimum and maximum values ​​of the grayscale range of the original image, f(x, y) is the image grayscale value corresponding to the new grayscale level j, and F(x, y) is the grayscale value of the image after stretching;

[0106] like Figure 4 As shown, the process of the recognition unit performing deformation and distortion detection on the image is as follows:

[0107] B21. Sum the grayscale values ​​of all pixels in the image to extract the geometric features of the image. The summation formula is as follows:

[0108] M0=∑ x ∑ y F(x,y), where M0 is the sum of the grayscale values ​​of all pixels in the image, and F(x,y) is the grayscale value of the preprocessed image;

[0109] B22. Determine two feature vectors e and g according to the extracted actual geometric features, where e = (e1, e2, ..., en), g = (g1, g2, ..., gn);

[0110] B23. Calculate the distance D(e,g) between two eigenvectors. The calculation formula of distance D(e,g) is as follows:

[0111] D(e,g)=∑ n＝1 |en-gn|;

[0112] B24. Compare the calculated distance D(e,g) with the set threshold (this set threshold is calculated in advance using the same package process type without abnormalities), identify the features exceeding the threshold as shape abnormalities, and identify the features within the threshold range as normal.

[0113] The process of color difference detection by the color contrast unit is as follows:

[0114] B31. Convert the preprocessed image data into CIE Lab color space. The conversion formula is as follows:

[0115] Among them, Y n is the stimulus value of the reference white point; Y is the brightness corresponding to the gray value F(x,y), L * is the brightness value in the CIE Lab color space;

[0116] B32. Select different locations of the fabric as sampling areas, and evenly select several sampling points in each area to represent the color of the area;

[0117] B33. Extract the corresponding color feature value at the selected sampling point and obtain its L in the CIE Lab color space. * 、a * and b * The value of

[0118] B34. Calculate the color difference between different sampling points. The color difference calculation formula is as follows:

[0119] in, and L * The two color values ​​of the channel, and They are respectively * The two color values ​​of the channel, and b * Two color values ​​of the channel, ΔE is the color difference between the two sampling points;

[0120] B35. Compare the calculated color difference value with the standard color difference threshold to determine whether there is a color difference problem. If the calculated color difference value exceeds the standard color difference threshold (the standard color difference threshold is obtained by sampling and calculating the same product without color difference in advance), it is determined that there is color difference at the position. If the calculated color difference value is within the standard color difference threshold, it is determined that there is no color difference at the position.

[0121] like Figure 5 As shown, the process of the suture detection unit detecting the suture is as follows:

[0122] B41. Filter the preprocessed data. The filter processing formula is as follows:

[0123] Y n =F(x,y)*d, where d is the filter coefficient, Y n is the gray value of the output image, and F(x,y) is the gray value of the image after preprocessing;

[0124] B42. Perform edge detection algorithm on the filtered image to extract possible edge information of the seam in the image. The edge calculation formula is as follows:

[0125]

[0126] Among them, S x is the horizontal gradient, S y is the vertical gradient, G c is the gradient amplitude, θ is the gradient direction;

[0127] B43. The suture width is obtained by calculating the distance between the two sides of the contour according to the extracted edge information, the suture length is obtained by calculating the length along the contour path, the suture density is obtained by counting the number of sutures per unit area, and the continuity of the suture contour, whether there is abnormal interruption or branching, is observed to obtain the characteristics of the suture;

[0128] B44. Compare the stitch characteristics with the standard stitch characteristics to confirm whether the stitches meet the quality requirements. If they meet the standard stitch characteristics, it is judged to be normal. If they exceed the standard stitch characteristics, it is judged to be abnormal. The standard stitch characteristics are the stitch quality judgment standards in the bag process type industry.

[0129] The process of accessory detection by the accessory detection unit is as follows:

[0130] B51. Use the target detection algorithm to identify the location of the hardware accessories in the package image from the preprocessed image, and divide the area containing the hardware accessories;

[0131] B52. Calculate the area Area and the perimeter Per of the contour according to the divided areas, and use the area Area and the perimeter Per to calculate the circularity C of the area. The calculation formula is as follows:

[0132]

[0133] B53. Compare the calculated circularity C with the circularity of the standard accessory. When the circularity C does not meet the circularity of the standard accessory, the hardware accessory is judged to be abnormal. When the circularity C meets the circularity of the standard accessory, the hardware accessory is judged to be normal.

[0134] By setting up the image processing and analysis module, combined with noise reduction and contrast enhancement technology, it is possible to more accurately identify quality problems and dimensional deviations of package process shapes and reduce the misjudgment rate.

[0135] The dimension measurement module is used to measure the dimension of the package process body, and then analyze the measured data to determine whether the dimension meets the quality requirements;

[0136] The size measurement module includes a laser distance sensor for size measurement and a data processor for analyzing and processing the measurement data. When measuring the size, the laser distance sensor selects a relatively fixed and easily identifiable part of the package as a reference point, and then measures different positions of the package from the reference point to obtain the size characteristics of the package.

[0137] The process of analyzing and processing the package size characteristics by the data processor is as follows:

[0138] C1. Set the standard range of each size according to the package process design requirements and quality standards:

[0139] C2. Calculate the deviation between the actual measured size and the standard size, and determine whether the deviation is within the tolerance range. If the deviation is within the tolerance range, the package size is normal; if the deviation exceeds the tolerance range, the package size is abnormal:

[0140] The display and alarm module is used to receive the detection result information of the image processing and analysis module and the size measurement module, and then display the results according to the detection results. At the same time, when the quality of the package process body is unqualified, an audible and visual alarm is issued to remind the staff to deal with it in time.

[0141] The process of displaying the results of the display and alarm module is as follows:

[0142] D1. Establish appearance quality inspection result table and dimension inspection result table respectively according to the analysis categories of image processing and analysis module and dimension measurement module, and set the standard range of qualified display in appearance quality inspection result table and dimension inspection result table respectively;

[0143] D2. Disassemble the received data information according to different test items and categories, and then classify the parsed information into the appearance quality test result table and the size test result table respectively;

[0144] D3. When the values ​​input into the appearance quality test result table and the size test result table are within the standard range, the qualified display is performed; when one of the values ​​input into the appearance quality test result table and the size test result table does not meet the standard range, the unqualified display is performed, and the unqualified abnormal content is displayed and the sound and light alarm is performed synchronously;

[0145] The following situations exist in the unqualified abnormal content display:

[0146] 1) If a color difference problem is detected in the color comparison unit, the word "unqualified" will be displayed in the specific color detection area on the display screen, and the specific parts where the color difference exists and the specific color difference values ​​calculated will be listed in detail, so that the staff can intuitively understand the color quality problems;

[0147] 2) For unqualified sutures found by the suture detection unit, the unqualified sutures are displayed in the suture detection area, and the location of the suture problem is displayed (such as the suture is off here, the length is about XX mm, etc.), as well as the characteristic parameters related to the suture (such as the comparison deviation of the suture width and density with the standard requirements, etc.);

[0148] 3) If defects are detected in hardware accessories, they will be displayed as unqualified in the accessory inspection area, and then the specific defects of the hardware accessories will be explained (for example, there is a scratch on a hardware buckle, the scratch length is about XX mm, and it is located at XX position on the accessory surface. At the same time, the scratched hardware accessories and their locations can be highlighted on the image to facilitate the staff to locate and check), and the corresponding appearance characteristics and differences from the standards will be given (such as non-conformity in shape, color, etc.);

[0149] 4) In the size inspection area of ​​the display screen, the word "unqualified" is displayed for each size item that exceeds the tolerance range, and the specific size name, actual measurement value, standard size range and deviation value are listed, so that the staff can clearly understand the size non-conformity.

[0150] Through the cooperation of image acquisition module, image processing and analysis module, dimension measurement module and result display and alarm module, using high-resolution camera and image processing software, it can conduct fast and accurate appearance inspection on package products, automatically identify problems such as shape deviation, surface flaws, color difference, pattern defects, etc., improve inspection efficiency and accuracy, and reduce the influence of human factors.

[0151] Based on the above-described packaging process type body production detection system, the present invention also provides a packaging process type body production detection method, such as Figure 6 As shown, the following steps are included:

[0152] S1. Use a high-definition camera to capture images of package process forms, and transmit the captured image data to the image processing and analysis module in the form of video signals; by capturing images with a high-definition camera and automatically analyzing them, the efficiency of production inspection of package process forms can be significantly improved. Automated processing reduces manual intervention and improves the operating speed and stability of the production line;

[0153] S2, the image processing and analysis module receives the video signal and converts it into digital image data;

[0154] S3, performing noise reduction processing on the received image data to remove noise and interference in the image, while enhancing the contrast and clarity of the image to improve the accuracy of the analysis;

[0155] S4. Analyze the processed image data and use algorithms to determine whether there are quality problems with the package process type. Using advanced image processing and analysis modules, combined with noise reduction and contrast enhancement technologies, it is possible to more accurately identify quality problems and dimensional deviations of package process types and reduce the misjudgment rate.

[0156] S5. The dimension measurement module is used to measure the dimension of the package process body, and the measured data is analyzed to determine whether the dimension meets the quality requirements. The dimension measurement module can promptly send out the dimension that does not meet the requirements, and can immediately trigger the sound and light alarm when it does not meet the requirements, reminding the staff to deal with it in time, thereby avoiding the occurrence of large-scale quality problems;

[0157] S6, the display and alarm module receives the detection result information of the image analysis and analysis module and the dimension measurement module, and displays the result according to the detection result. The displayed result includes the specific information of the quality problem and the dimension problem. When the quality of the package process body is unqualified or the dimension does not meet the requirements, the sound and light alarm is triggered to remind the staff to deal with it in time;

[0158] S7. Record the problems found during the inspection process, including the type, location, and severity of the problem. Automated inspection methods are used to reduce labor and time costs. At the same time, timely detection and resolution of problems avoids material waste and other additional expenses caused by defective products, thereby reducing overall production costs.

[0159] Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent substitutions for some of the technical features therein; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A packaging process body production detection system, characterized by: It includes an image acquisition module, which captures the image of the packaging process body through a high-definition camera and outputs a video signal to the image processing and analysis module; An image processing and analysis module, which is used to receive the image data of the packaging process body transmitted by the image acquisition module, perform noise reduction and enhancement processing on the image data, and then analyze the processed data to determine whether there is any quality problem with the packaging process body; The image processing and analysis module includes a preprocessing unit for preprocessing image data, an identification unit for detecting deformation and distortion of the preprocessed data, a color comparison unit for detecting color difference of the preprocessed data, a suture detection unit for detecting sutures of the preprocessed data, and an accessory detection unit for detecting hardware accessories in the preprocessed data, wherein the preprocessing unit is electrically connected to the identification unit, the color comparison unit, the suture detection unit, and the accessory detection unit, respectively; A dimension measurement module is used to measure the dimensions of the package process body and then analyze the measured data to determine whether the dimensions meet the quality requirements; The result display and alarm module is used to receive the detection result information of the image processing and analysis module and the dimension measurement module, and then display the result according to the detection result. At the same time, when the quality of the package process body is unqualified, an audible and visual alarm is issued to remind the staff to deal with it in time.

2. A packaging process body production detection system according to claim 1, characterized in that: The image acquisition module acquires images of the packaging process body as follows: A1. Calculate the light intensity of the lighting equipment according to the shooting requirements of the high-definition camera. The light intensity calculation formula is as follows: Where E is the light intensity, I is the luminous intensity of the light source, and r is the distance from the illuminated surface to the light source. The light intensity is adjusted by changing the light source distance r, so that the lighting equipment can provide a uniform, sufficient and shadow-free light environment for the high-definition camera. A2. By adjusting the position and angle of the shooting bracket, the high-definition camera can fully cover all parts of the package to obtain image information from different perspectives; A3. Obtain the package body image under the corresponding viewing angle by taking the package body image information captured by the high-definition camera in real time; A4. Transmit the package body image to the image processing and analysis module in real time.

3. A packaging process body production detection system according to claim 2, characterized in that: The process of the preprocessing unit preprocessing the image data is as follows: B11. Receive the appearance image data of the multi-view packaging process body from the image acquisition module, and use the noise reduction formula to perform noise reduction processing on the appearance image data. The noise reduction formula is as follows: Where k and l are the values ​​of the window size, m and n are the length and width of the window size, I(x+i,y+j) is the grayscale value of the pixel with coordinates (x+i,y+j) in the neighborhood, and G(x,y) is the grayscale value of the image after denoising. B12. Confirm the gray level i of the gray value G(x,y) of the image after denoising, and calculate the weight value Wi of the gray level i. The calculation formula of the weight value Wi is as follows: Where N is the total number of pixels in the image, n i is the number of pixels of image gray level i; B13. Map the gray level i to obtain a new gray level j. The mapping formula is as follows: j=(L)*∑ i＝0 Wi, where L is the maximum value of the grayscale range; B14. Calculate the grayscale value corresponding to the new grayscale level j according to the grayscale range of the original image. The calculation formula is as follows: Where f(x,y) is the grayscale value corresponding to the new grayscale level j, a is the minimum value of the grayscale range of the original image, L is the maximum value of the grayscale range of the original image, G(x,y) is the grayscale value of the original image, and M is the number of grayscale levels; B15. Perform linear contrast stretching on the image and calculate the grayscale value after stretching. The calculation formula is as follows: Among them, c and d are the minimum and maximum values ​​of the grayscale range after stretching, a and L are the minimum and maximum values ​​of the grayscale range of the original image, f(x, y) is the image grayscale value corresponding to the new grayscale level j, and F(x, y) is the grayscale value of the image after stretching.

4. A packaging process body production detection system according to claim 3, characterized in that: The process of the recognition unit detecting deformation and distortion of an image is as follows: B21. Sum the grayscale values ​​of all pixels in the image to extract the geometric features of the image. The summation formula is as follows: M0=∑ x ∑ y F(x,y), where M0 is the sum of the grayscale values ​​of all pixels in the image, and F(x,y) is the grayscale value of the preprocessed image; B22. Determine two feature vectors e and g according to the extracted actual geometric features, where e = (e1, e2, ..., en) and g = (g1, g2, ..., gn); B23. Calculate the distance D(e,g) between two eigenvectors. The calculation formula of distance D(e,g) is as follows: D(e,g)=∑ n＝1 |en-gn|; B24. Compare the calculated distance D(e,g) with the set threshold, identify the features exceeding the threshold as shape anomalies, and identify the features within the threshold range as normal.

5. A packaging process body production detection system according to claim 4, characterized in that: The process of color difference detection by the color contrast unit is as follows: B31. Convert the preprocessed image data into CIE Lab color space. The conversion formula is as follows: Among them, Y n is the stimulus value of the reference white point; Y is the brightness corresponding to the gray value F(x,y), L * is the brightness value in the CIE Lab color space; B32. Select different locations of the fabric as sampling areas, and evenly select several sampling points in each area to represent the color of the area; B33. Extract the corresponding color feature value at the selected sampling point and obtain its L in the CIE Lab color space. * 、a * and b * The value of B34. Calculate the color difference between different sampling points. The color difference calculation formula is as follows: in, and L * The two color values ​​of the channel, and They are respectively * The two color values ​​of the channel, and b * Two color values ​​of the channel, ΔE is the color difference between the two sampling points; B35. Compare the calculated color difference value with the standard color difference threshold to determine whether there is a color difference problem. If the calculated color difference value exceeds the standard color difference threshold, it is determined that there is color difference at the position. If the calculated color difference value is within the standard color difference threshold range, it is determined that there is no color difference at the position.

6. A packaging process body production detection system according to claim 5, characterized in that: The process of the suture detection unit detecting the suture is as follows: B41. Filter the preprocessed data. The filter processing formula is as follows: Y n =F(x,y)*d, where d is the filter coefficient, Y n is the gray value of the output image, and F(x,y) is the gray value of the image after preprocessing; B42. Perform edge detection algorithm on the filtered image to extract possible edge information of the seam in the image. The edge calculation formula is as follows: Among them, S x is the horizontal gradient, S y is the vertical gradient, G c is the gradient amplitude, θ is the gradient direction; B43. The suture width is obtained by calculating the distance between the two sides of the contour according to the extracted edge information, the suture length is obtained by calculating the length along the contour path, the suture density is obtained by counting the number of sutures per unit area, and the continuity of the suture contour, whether there is abnormal interruption or branching, is observed to obtain the characteristics of the suture; B44. Compare the suture characteristics with the standard suture characteristics to confirm whether the suture meets the quality requirements. If it meets the standard suture characteristics, it is judged to be normal. If it exceeds the standard suture characteristics, it is judged to be abnormal.

7. A packaging process body production detection system according to claim 6, characterized in that: The process of accessory detection by the accessory detection unit is as follows: B51. Use the target detection algorithm to identify the location of the hardware accessories in the package image from the preprocessed image, and divide the area containing the hardware accessories; B52. Calculate the area Area and the perimeter Per of the contour according to the divided areas, and use the area Area and the perimeter Per to calculate the circularity C of the area. The calculation formula is as follows: B53. Compare the calculated circularity C with the circularity of the standard accessory. When the circularity C does not conform to the circularity of the standard accessory, the hardware accessory is judged to be abnormal. When the circularity C conforms to the circularity of the standard accessory, the hardware accessory is judged to be normal.

8. A packaging process body production detection system according to claim 7, characterized in that: The size measurement module includes a laser distance sensor for size measurement and a data processor for analyzing and processing the measurement data. When measuring the size, the laser distance sensor selects a relatively fixed and easily identifiable part of the package as a reference point, and then measures different positions of the package from the reference point to obtain the size characteristics of the package. The process of analyzing and processing the package size characteristics by the data processor is as follows: C1. Set the standard range of each size according to the package process design requirements and quality standards: C2. Calculate the deviation between the actual measured size and the standard size, and determine whether the deviation is within the tolerance range. If the deviation is within the tolerance range, the package size is normal; if the deviation exceeds the tolerance range, the package size is abnormal.

9. A packaging process body production detection system according to claim 8, characterized in that: The process of displaying the results of the display and alarm module is as follows: D1. Establish appearance quality inspection result table and dimension inspection result table respectively according to the analysis categories of image processing and analysis module and dimension measurement module, and set the standard range of qualified display in appearance quality inspection result table and dimension inspection result table respectively; D2. Disassemble the received data information according to different test items and categories, and then classify the parsed information into the appearance quality test result table and the size test result table respectively; D3. When the values ​​input into the appearance quality test result table and the size test result table are within the standard range, they will be displayed as qualified; when one of the values ​​input into the appearance quality test result table and the size test result table does not meet the standard range, it will be displayed as unqualified, and the unqualified abnormal content will be displayed and the sound and light alarm will be performed synchronously.

10. A method for detecting the production of a package process type body, based on a system for detecting the production of a package process type body according to any one of claims 1 to 9, characterized in that: The steps include: S1. Use a high-definition camera to capture the image of the packaging process body, and transmit the captured image data to the image processing and analysis module in the form of a video signal; S2, the image processing and analysis module receives the video signal and converts it into digital image data; S3, performing noise reduction processing on the received image data, while enhancing the contrast and clarity of the image; S4, analyzing the processed image data, and then judging whether there is any quality problem in the package process body through an algorithm; S5. Measure the dimensions of the package process body through the dimension measurement module, analyze the measured data, and determine whether the dimensions meet the quality requirements; S6, the display and alarm module receives the detection result information of the image analysis and analysis module and the dimension measurement module, and displays the result according to the detection result. The displayed result includes the specific information of the quality problem and the dimension problem. When the quality of the package process body is unqualified or the dimension does not meet the requirements, the sound and light alarm is triggered to remind the staff to deal with it in time; S7. Record the problems found during the inspection process, including the type, location, and severity of the problem.

Images