Packaging box printed matter printing quality detection method

Abstract

The invention discloses a packaging box printed matter printing quality detection method, which comprises the following steps of: acquiring a high-resolution printing image through an image acquisition module, preprocessing and segmenting the printing image by an image processing module, optimizing image quality, classifying the preprocessed image by an image classification module according to characters, patterns and color blocks, and detecting the printing quality of a packaging box printed matter. The defect detection module performs multi-dimensional detection on the classified preprocessed images based on image parameter data, the quality analysis module quantitatively evaluates quality scores of characters, patterns and color blocks based on detection results, and the computer terminal generates a comprehensive score through weighting calculation and automatically generates a detection report; according to the packaging box printed matter printing quality detection method, complex defects such as font errors, stroke breakage, pattern deviation, color deviation and stains are effectively recognized through multi-dimensional defects, the manual reinspection cost is reduced through full-process automatic processing, and the quality of printed matter is visually reflected through a quantitative evaluation system.

Description

Technical Field

[0001] The invention belongs to the technical field of printed matter quality detection, and specifically discloses a method for detecting the printing quality of printed matter on a packaging box. Background Art

[0002] The clarity of patterns and color consistency of printed materials directly affect consumers' first impression of the product. Blurred text or incorrect anti-counterfeiting marks may damage the brand image. Quality inspection can detect material or process problems in advance and reduce the waste of raw materials caused by batch defects.

[0003] Traditional printing quality inspection mainly relies on manual visual inspection, which is not only inefficient but also easily affected by subjective factors, resulting in unreliable and unstable inspection results. With the advancement of technology, although some automated printing quality inspection methods have been developed, these methods can often only detect single printing defects, such as font errors, pattern missing or color deviation, and lack a comprehensive assessment of the overall quality of printed products.

[0004] Therefore, there is an urgent need to invent a method for detecting the printing quality of printed materials on packaging boxes to solve the problems existing in the prior art, such as low detection efficiency, detection results being easily affected by subjective factors, and lack of comprehensive evaluation means. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method for detecting the printing quality of printed materials on packaging boxes, which obtains high-resolution printed images through an image acquisition module, and the image processing module preprocesses the printed images, segments and optimizes the image quality. The image classification module classifies and preprocesses the images according to text, patterns and color blocks, and the defect detection module performs multi-dimensional detection on the classified preprocessed images based on image parameter data. The quality analysis module quantitatively evaluates the quality scores of text, patterns and color blocks based on the detection results, and the computer terminal generates a comprehensive score through weighted calculation; and automatically generates a detection report; the printing quality detection method for printed materials on packaging boxes proposed by the present invention significantly improves the detection accuracy and efficiency through multi-dimensional defect precision identification, full-process automated processing and quantitative evaluation system, and at the same time effectively reduces the defective rate and manual re-inspection cost through high-precision defect positioning and real-time feedback; effectively solves the problems mentioned in the background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solution: including a data acquisition module, an image processing module, an image classification module, a defect detection module, a quality analysis module, a database, and a computer terminal, specifically including the following steps:

[0007] S1. The data acquisition module performs a data acquisition to collect image data of the printed matter to be inspected;

[0008] S2, the image processing module preprocesses the acquired image data to obtain preprocessed image data;

[0009] S3, the image classification module classifies the pre-processed image data and divides the pre-processed image into text area images, pattern area images and color block area images;

[0010] S4. The data acquisition module performs secondary data acquisition and collects image parameter data. The defect detection module performs multi-dimensional detection on the classified pre-processed image based on the image parameter data and outputs the defect detection results to the quality analysis module.

[0011] S5. The quality analysis module evaluates the printing quality based on the defect detection results and outputs the printing quality evaluation results to the computer terminal;

[0012] S6. The computer terminal calculates the printing quality score based on the printing quality assessment results and generates a test report.

[0013] Based on the above embodiment, the image data is a digital image of the printed surface of the packaging box; the image parameter data includes: the starting and ending positions of the text in the text area image, the feature point positions, image diagonal length and total pattern area of ​​the pattern area image, and the total area of ​​the color blocks and sampling point color depth of the color block area image.

[0014] Based on the above embodiment, it is characterized in that: the preprocessing includes: edge sharpening, histogram equalization and Gaussian filtering.

[0015] Based on the above embodiment, the specific process of the multi-dimensional detection is as follows:

[0016] A1. Perform OCR character recognition on the text area image and compare the recognition results with the standard vector font library in the database to detect font errors and stroke breaks, and obtain the text error rate and text offset of the text area image;

[0017] A2. Match the pattern area image with the pattern templates in the database, perform structural similarity analysis, detect pattern offset and missing, and obtain structural similarity, position offset, and missing area ratio;

[0018] A3. Perform Lab color difference analysis combined with morphological operations on the color block area image to detect color deviation and stains, and obtain the ΔE color difference value and the stain area.

[0019] Based on the above embodiment, the printing quality assessment includes: text quality assessment, image quality assessment and color block quality assessment.

[0020] Based on the above embodiment, the specific method of the text quality assessment is: calculating the text quality score based on the text error rate, text offset, and character recognition accuracy. The calculation formula of the text quality score is: Q in the formula t is the text quality score, E is the text error rate, D is the text offset, R is the character recognition accuracy, ω1 is the error rate weight coefficient, ω2 is the offset weight coefficient, λ is the defect penalty coefficient, and the variable definition domain is: E, D, R ∈ [0, 1].

[0021] Based on the above embodiment, the specific method of image quality assessment is: calculating the image quality score based on pattern structure similarity, position offset, and missing area ratio. The calculation formula of the image quality score is: Q in the formula i is the image quality score, SSIM(I d ,I r ) represents the structural similarity of the pattern, I d is the actual image, I r is the standard image in the database, ΔP i is the position deviation of the i-th feature point, P0 is the length of the image diagonal, A m is the area of ​​the missing region, A t is the total area of ​​the pattern, ω3 and ω4 are weight coefficients, and n is the total number of image feature points.

[0022] Based on the above embodiment, the specific method of the color block quality assessment is: calculating the color block quality score according to the ΔE color difference value and the stain area. The calculation formula of the color block quality score is: Q in the formula c is the color block quality score, N is the total number of sampling points, L j is the Lab value of the jth sampling point, L j0 is the standard Lab value of the jth sampling point in the database, ΔE 00 (L j ,L j0 ) is the ΔE color difference value of the j-th sampling point, S d is the area of ​​the stain, S t is the total area of ​​the color block, ω5 and ω6 are weight coefficients.

[0023] Based on the above embodiment, the calculation formula for the printing quality score is: In the formula, Q is the printing quality score, Q lim The single quality qualification threshold, ω7, ω8 and ω9 are weight coefficients.

[0024] Technical effects and advantages of the present invention:

[0025] 1. Accurate multi-dimensional defect identification: Through classification detection and targeted algorithms, defect detection accuracy is significantly improved. It can effectively identify complex defects such as font errors, broken strokes, pattern offsets, color deviations, and stains, reducing missed detections and false detections, significantly reducing the risk of defective products entering the market, while saving manual re-inspection costs and improving overall production efficiency.

[0026] 2. Automation and efficiency: Full process automation reduces manual intervention, and inspection efficiency is several times higher than traditional manual visual inspection, making it suitable for rapid quality inspection of large-scale printed product lines;

[0027] 3. Quantitative evaluation and standardized scoring: A normalized formula is used to quantify and score text quality, image quality, and color block quality. A weighted comprehensive score is generated to achieve objectivity and consistency in quality assessment, facilitating horizontal comparison and quality traceability. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The present invention is further described with reference to the accompanying drawings. However, the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative effort.

[0029] Figure 1 It is a schematic diagram of the overall structure of the present invention.

[0030] Figure 2 It is an overall step diagram of the present invention. DETAILED DESCRIPTION

[0031] The technical solutions 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 only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0032] The present invention provides Figure 1 A printing quality inspection method for printed matter on a packaging box is shown, comprising a data acquisition module, an image processing module, an image classification module, a defect detection module, a quality analysis module, a database, and a computer terminal;

[0033] In a more specific application of the present invention, the data acquisition module is used to acquire image data and image parameter data collected by the primary data acquisition and the secondary data acquisition, wherein the image data is a digital image of the printed surface of the packaging box; the image parameter data includes: the start and end positions of the text in the text area image, the feature point positions, the image diagonal length and the total area of ​​the pattern in the pattern area image, and the total area of ​​the color blocks and the color depth of the sampling points in the color block area image;

[0034] In a more specific application of the present invention, the image processing module is used to pre-process the data acquired by the data acquisition module, specifically including edge sharpening, histogram equalization and Gaussian filtering.

[0035] In a more specific application of the present invention, the image classification module is used to divide the pre-processed image into a text area image, a pattern area image and a color block area image.

[0036] In a more specific application of the present invention, the defect detection module is used to perform multi-dimensional detection on the classified pre-processed image and output the defect detection results to the quality analysis module;

[0037] Furthermore, in the above technical solution, the specific process of the multi-dimensional detection is as follows:

[0038] A1. Perform OCR character recognition on the text area image and compare the recognition results with the standard vector font library in the database to detect font errors and stroke breaks, and obtain the text error rate and text offset of the text area image;

[0039] A2. Match the pattern area image with the pattern templates in the database, perform structural similarity analysis, detect pattern offset and missing, and obtain structural similarity, position offset, and missing area ratio;

[0040] A3. Perform Lab color difference analysis combined with morphological operations on the color block area image to detect color deviation and stains, and obtain the ΔE color difference value and the stain area.

[0041] In a more specific application of the present invention, the quality analysis module is used to perform printing quality assessment based on the defect detection results and output the printing quality assessment results to a computer terminal;

[0042] Furthermore, in the above technical solution, the printing quality assessment includes: text quality assessment, image quality assessment and color block quality assessment.

[0043] In a more specific application of the present invention, the database is used to store standard images of printed products, parameter data of the standard images, analysis results of the quality analysis module, defect detection results of the defect detection module, and image quality scores calculated by image quality assessment.

[0044] In a more specific application of the present invention, the computer terminal is used to calculate a printing quality score and generate a test report based on the printing quality evaluation result.

[0045] like Figure 2 As shown, the specific steps include:

[0046] S1. The data acquisition module performs a data acquisition to collect image data of the printed matter to be inspected;

[0047] In a preferred technical solution of the present application, the image data is a digital image of the printed surface of the packaging box; the image parameter data includes: the starting and ending positions of the characters in the text area image, the feature point positions, the image diagonal length and the total pattern area of ​​the pattern area image, and the total area of ​​the color blocks and the color depth of the sampling points in the color block area image;

[0048] Furthermore, in the above technical solution, image data is acquired through a high-precision industrial CCD camera array in conjunction with a multi-spectral ring light source system, and image parameter data is acquired through an embedded image processing system, wherein the starting and ending positions of the text are located through edge detection and semantic segmentation neural network, the feature point positions are matched based on the SIFT feature extraction algorithm, the image diagonal length is measured by sub-pixel image analysis software, the total area of ​​the pattern is acquired through pixel clustering statistics, the total area of ​​the color block is determined by a region growing algorithm, and the color depth of the sampling point is quantitatively collected in Lab color space using a spectrophotometer.

[0049] S2, the image processing module preprocesses the acquired image data to obtain preprocessed image data;

[0050] In a preferred technical solution of the present application, the preprocessing includes: edge sharpening, histogram equalization and Gaussian filtering;

[0051] Furthermore, in the above technical solution, the specific method of edge sharpening is: using a combination of the Sobel operator and the Laplace edge detection algorithm, extracting image edge features through the Sobel horizontal and vertical gradient operators with a convolution kernel size of 3×3, superimposing the Laplace second-order differential operation to enhance the edge contrast, and adjusting the sharpening intensity; the specific method of histogram equalization is: based on the limited contrast adaptive histogram equalization algorithm, the image is divided into 8×8 sub-regions, the contrast limit threshold is set to 2.0, and bilinear interpolation is used to eliminate the grayscale steps between sub-regions to achieve local contrast optimization; the specific method of Gaussian filtering is: using a two-dimensional discrete Gaussian convolution kernel with a standard deviation σ=1.5, the kernel size is set to 5×5, and linear filtering is performed in the row direction and then in the column direction of the image through a separable convolution method, combined with a boundary mirror filling strategy to suppress image noise and retain detail features.

[0052] S3, the image classification module classifies the pre-processed image data and divides the pre-processed image into text area images, pattern area images and color block area images;

[0053] S4. The data acquisition module performs secondary data acquisition and collects image parameter data. The defect detection module performs multi-dimensional detection on the classified pre-processed image based on the image parameter data and outputs the defect detection results to the quality analysis module.

[0054] In the preferred technical solution of this application, the specific process of the multi-dimensional detection is as follows:

[0055] A1. Perform OCR character recognition on the text area image and compare the recognition results with the standard vector font library in the database to detect font errors and stroke breaks, and obtain the text error rate and text offset of the text area image;

[0056] A2. Match the pattern area image with the pattern templates in the database, perform structural similarity analysis, detect pattern offset and missing, and obtain structural similarity, position offset, and missing area ratio;

[0057] A3. Perform Lab color difference analysis combined with morphological operations on the color block area image to detect color deviation and stains, and obtain the ΔE color difference value and the stain area.

[0058] S5. The quality analysis module evaluates the printing quality based on the defect detection results and outputs the printing quality evaluation results to the computer terminal;

[0059] In a preferred technical solution of the present application, the printing quality assessment includes: text quality assessment, image quality assessment and color block quality assessment;

[0060] Furthermore, in the above technical solution, the specific method of the text quality assessment is: calculating the text quality score based on the text error rate, text offset, and character recognition accuracy. The calculation formula of the text quality score is: Q in the formula t is the text quality score, E is the text error rate, D is the text offset, R is the character recognition accuracy, ω1 is the error rate weight coefficient, ω2 is the offset weight coefficient, λ is the defect penalty coefficient, and the variable definition domain is: E, D, R ∈ [0, 1];

[0061] Furthermore, in the above technical solution, the error rate weight coefficient ω1 and the offset weight coefficient ω2 are obtained by a mapping set of historical data and weight coefficients established in a database, that is, the corresponding weight coefficients are obtained according to the current data; the defect penalty coefficient λ is determined according to comparative experiments;

[0062] It should be noted that ω3, ω4, ω5, ω6, ω7, ω 8和 ω9 is also obtained through the mapping set of historical data and weight coefficients established in the database, that is, the corresponding weight coefficient is obtained according to the current data;

[0063] Furthermore, in the above technical solution, the specific method of image quality assessment is: calculating the image quality score based on pattern structure similarity, position offset, and missing area ratio. The calculation formula of the image quality score is: Q in the formula i is the image quality score, SSIM(I d ,I r ) represents the structural similarity of the pattern, I d is the actual image, I r is the standard image in the database, ΔP i is the position deviation of the i-th feature point, P0 is the length of the image diagonal, A m is the area of ​​the missing region, A t is the total area of ​​the pattern, ω3 and ω4 are weight coefficients, and n is the total number of image feature points;

[0064] Furthermore, in the above technical solution, the specific method of the color block quality assessment is: calculating the color block quality score according to the ΔE color difference value and the stain area. The calculation formula of the color block quality score is: Q in the formula c is the color block quality score, N is the total number of sampling points, L j is the Lab value of the jth sampling point, L j0 is the standard Lab value of the jth sampling point in the database, ΔE 00 (L j ,L j0 ) is the ΔE color difference value of the j-th sampling point, S d is the area of ​​the stain, S t is the total area of ​​the color block, ω5 and ω6 are weight coefficients;

[0065] Calculation example:

[0066] Example of text quality score calculation:

[0067] Assume parameters:

[0068] Text error rate E = 0.05, text offset D = 0.02, character recognition accuracy R = 0.98, weight coefficient: ω1 = 0.7, ω2 = 0.3, defect penalty coefficient: λ = 1.5;

[0069] Calculation process:

[0070] Q t =[(1-0.7×0.05)×(1-0.3×0.02)×0.98] 1 / 3 ×e -1.5×0.05 ×100≈0.98×

[0071] 0.93×100=91.14;

[0072] Conclusion: The text quality score is 91.14;

[0073] Image quality score calculation example:

[0074] Assume parameters:

[0075] Structural similarity SSIM (I d ,I r )=0.92, position offset Amount = 5, image diagonal length P0 = 100, missing area ratio A m / A t =0.01, weight coefficient: ω3=0.5, ω4=0.5, total number of image feature points n=10;

[0076] Calculation process: Q i =(0.92-5 / (100×10)×0.5-0.01×0.5)×100=91.25;

[0077] Conclusion: The image quality score is 91.25;

[0078] Example of calculating the color patch quality score:

[0079] Parameter assumptions: the total number of sampling points N = 10, the sum of the squares of the ΔE color difference values ​​of the 10 sampling points = 25, Stain area ratio S d / S t =0.02, weight coefficient: ω5=0.6, ω6=0.4;

[0080] Calculation process:

[0081] Conclusion: The color patch quality score is 98.42;

[0082] S6. The computer terminal calculates the printing quality score based on the printing quality assessment results and generates a test report;

[0083] In the preferred technical solution of this application, the calculation formula for the printing quality score is: In the formula, Q is the printing quality score, Q lim The qualified threshold of single quality, ω7, ω8 and ω9 are weight coefficients;

[0084] Example of print quality score calculation:

[0085] Parameter assumptions:

[0086] Single quality threshold Q lim=70, weight coefficients: ω7=0.4, ω8=0.4, ω9=0.2;

[0087] Image 1 scoring parameters: text quality score Q t =91, image quality score Q i =92, color block quality score Q c =98;

[0088] Image 2 scoring parameters: text quality score Q t =60, image quality score Q i =65, color block quality score Q c =75;

[0089] Calculation process:

[0090] Picture 1 printing quality score: Q1=91 0.4 ×92 0.4 ×98 0.2 ≈92.76;

[0091] Image 2 printing quality score: Q2 = min(60,65,75) = 60;

[0092] Conclusion: The printing quality score of Picture 1 is 92.76, and the printing quality score of Picture 2 is 60;

[0093] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for detecting the printing quality of printed matter on a packaging box, characterized in that: It includes a data acquisition module, an image processing module, an image classification module, a defect detection module, a quality analysis module, a database, and a computer terminal, and specifically includes the following steps: S1. The data acquisition module performs a data acquisition to collect image data of the printed matter to be inspected; S2, the image processing module preprocesses the acquired image data to obtain preprocessed image data; S3, the image classification module classifies the pre-processed image data and divides the pre-processed image into text area images, pattern area images and color block area images; S4. The data acquisition module performs secondary data acquisition and collects image parameter data. The defect detection module performs multi-dimensional detection on the classified pre-processed image based on the image parameter data and outputs the defect detection results to the quality analysis module. S5. The quality analysis module evaluates the printing quality based on the defect detection results and outputs the printing quality evaluation results to the computer terminal; S6. The computer terminal calculates the printing quality score based on the printing quality assessment results and generates a test report.

2. The method for detecting printing quality of printed matter on a packaging box according to claim 1, wherein: The image data is a digital image of the printed surface of the packaging box; The image parameter data includes: the starting and ending positions of the characters in the character area image, the feature point positions, the image diagonal length and the total pattern area of ​​the pattern area image, and the total color block area and sampling point color depth of the color block area image.

3. The method for detecting printing quality of printed matter on a packaging box according to claim 1, wherein: The preprocessing includes edge sharpening, histogram equalization and Gaussian filtering.

4. The method for detecting printing quality of printed matter on a packaging box according to claim 1, wherein: The specific process of the multi-dimensional detection is as follows: A1. Perform OCR character recognition on the text area image and compare the recognition results with the standard vector font library in the database to detect font errors and stroke breaks, and obtain the text error rate and text offset of the text area image; A2. Match the pattern area image with the pattern templates in the database, perform structural similarity analysis, detect pattern offset and missing, and obtain structural similarity, position offset, and missing area ratio; A3. Perform Lab color difference analysis combined with morphological operations on the color block area image to detect color deviation and stains, and obtain the ΔE color difference value and the stain area.

5. The method for detecting printing quality of printed matter on a packaging box according to claim 1, wherein: The printing quality assessment includes: text quality assessment, image quality assessment and color block quality assessment.

6. A method for inspecting printing quality of printed matter on a packaging box according to claim 5, characterized in that: The specific method of the text quality assessment is: calculating the text quality score based on the text error rate, text offset, and character recognition accuracy. The calculation formula of the text quality score is: Q in the formula t is the text quality score, E is the text error rate, D is the text offset, R is the character recognition accuracy, ω1 is the error rate weight coefficient, ω2 is the offset weight coefficient, λ is the defect penalty coefficient, and the variable definition domain is: E, D, R ∈ [0, 1].

7. The method for detecting printing quality of printed matter on a packaging box according to claim 5, wherein: The specific method of image quality assessment is: calculating the image quality score based on pattern structure similarity, position offset, and missing area ratio. The calculation formula of the image quality score is: Q in the formula i is the image quality score, SSIM(I d ,I r ) represents the structural similarity of the pattern, I d is the actual image, I r is the standard image in the database, ΔP i is the position deviation of the i-th feature point, P0 is the length of the image diagonal, A m is the area of ​​the missing region, A t is the total area of ​​the pattern, ω3 and ω4 are weight coefficients, and n is the total number of image feature points.

8. The method for detecting printing quality of printed matter on a packaging box according to claim 5, wherein: The specific method of the color block quality assessment is: the color block quality score is calculated based on the ΔE color difference value and the stain area. The calculation formula of the color block quality score is: Q in the formula c is the color block quality score, N is the total number of sampling points, L j is the Lab value of the jth sampling point, L j0 is the standard Lab value of the jth sampling point in the database, ΔE 00 (L j ,L j0 ) is the ΔE color difference value of the j-th sampling point, S d is the area of ​​the stain, S t is the total area of ​​the color block, ω5 and ω6 are weight coefficients.

9. The method for detecting printing quality of printed matter on a packaging box according to claim 1, wherein: The calculation formula for the printing quality score is: In the formula, Q is the printing quality score, Q lim The single quality qualification threshold, ω7, ω8 and ω9 are weight coefficients.

Images