Digital printing quality detection method based on image analysis

Abstract

The invention discloses a digital printing quality detection method based on image analysis, and belongs to the technical field of quality control, and the method specifically comprises the steps: monitoring a voltage value of a preset light source, and generating a fluctuation curve of the voltage value along with the time change; collecting fluorescence images of the same printed matter under the irradiation of a preset light source in a detection period in real time, and generating an observation fluorescence image sequence; analyzing the observation fluorescence image sequence to obtain mass fractions of all observation fluorescence images; extracting a voltage value of a corresponding time node from the voltage fluctuation image, and constructing a point location map by taking the voltage value as an abscissa and the mass fraction as an ordinate; fitting the point map to obtain a fitting function, correcting the quality score according to the fitting function, and if the corrected image quality score is greater than or equal to a preset threshold value, judging that the printed matter is qualified; according to the invention, misjudgment and missed judgment caused by voltage fluctuation factors are reduced, and the accuracy and reliability of quality detection are improved.

Description

Technical Field

[0001] The invention relates to the technical field of quality control, and in particular to a digital printing quality detection method based on image analysis. Background Art

[0002] The printing industry is one of the largest industries in the world and one of the important channels for the transmission and interpretation of large amounts of data. The social contribution of the printing industry is constantly increasing and has become an important factor in the social economy. Printing is a technology that transfers the ink from the original text, pictures, photos, anti-counterfeiting manuscripts to the surface of paper, textiles, plastics, leather, PVC, PC and other materials through processes such as plate making, inking and pressurization to copy the original content in batches. Printing is the process of transferring the approved printing plate to the substrate through printing machinery and special ink. The definition of printing in international standards is: the reproduction process of transferring the colorant / colorant (such as ink) to the substrate using analog or digital image carriers.

[0003] In the prior art, machine vision is usually used to control the quality of printed products, that is, the printed product image after being irradiated by a preset light source is proofread with the template image, so as to achieve efficient and accurate printing quality control. However, in the actual process, the voltage fluctuation of the power supply will affect the intensity and stability of the light source. The voltage fluctuation will change the brightness and color temperature of the light source. The changes in brightness and color temperature will affect the color performance of the image, and then affect the accuracy of subsequent image matching and defect detection, resulting in inconsistent acquisition effects of each printed product image, which will lead to errors in the actual printing quality judgment and affect the quality control results. Summary of the invention

[0004] The purpose of the present invention is to provide a digital printing quality detection method based on image analysis to solve the following technical problems:

[0005] In the prior art, machine vision is usually used to control the quality of printed products, that is, the printed product image after being irradiated by a preset light source is proofread with the template image, so as to achieve efficient and accurate printing quality control. However, in the actual process, the voltage fluctuation of the power supply will affect the intensity and stability of the light source. The voltage fluctuation will change the brightness and color temperature of the light source. The changes in brightness and color temperature will affect the color performance of the image, and then affect the accuracy of subsequent image matching and defect detection, resulting in inconsistent acquisition effects of each printed product image, which will lead to errors in the actual printing quality judgment and affect the quality control results.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A digital printing quality detection method based on image analysis, S1, within a preset detection period, monitoring the voltage value of a preset light source and generating a fluctuation curve of the voltage value over time; real-time acquisition of the fluorescence image of the same printed matter under the irradiation of the preset light source within the detection period, generating an observed fluorescence image sequence;

[0008] S2, analyzing the observed fluorescence image sequence to obtain the quality scores of all observed fluorescence images; obtaining the corresponding time node of any observed fluorescence image, extracting the voltage value of the corresponding time node from the voltage fluctuation image, and constructing a point map with the voltage value as the horizontal coordinate and the quality score as the vertical coordinate; fitting the point map to obtain a fitting function, and the fitting function is used to fit the relationship between the quality score and the voltage value;

[0009] S3, obtaining the quality score and voltage value of the fluorescent image corresponding to the current printed product, and correcting the fluorescent image quality score of the printed product according to the fitting function. If the corrected image quality score is greater than or equal to a preset threshold, the printed product is judged to be qualified; if the corrected image quality score is less than the preset threshold, the printed product is judged to be unqualified.

[0010] As a further solution of the present invention: in S2, the calculation process of the mass fraction is:

[0011] The observed fluorescence image is gray-processed to obtain a gray-scale image. A rectangular coordinate system is established with the central pixel of the gray-scale image as the origin to generate the coordinates (x, y) of all pixels in the gray-scale image. The gray-scale values ​​P(x, y) of all pixels in the gray-scale image are identified. For the corresponding pixels in the template image, their gray-scale values ​​P′(x′, y′) are recorded. The gray-scale difference D of the pixels at the same position between the gray-scale image and the template image is calculated by the gray-scale difference formula D=|Pi(x, y)-Pi'(x', y')|. The gray-scale differences of all pixels are accumulated to obtain the total gray-scale difference M.

[0012] The fluorescent image and the template image are respectively converted into the HIS color space to obtain the corresponding HSV color matrix, the Haar transform is applied to the HSV color matrix, and the matrix after the Haar transform is normalized to obtain the color feature vector a of the fluorescent head image and the color feature vector b of the template image; the vector cosine value N of the color feature vector a and the color feature vector b is calculated;

[0013] The quality score of any fluorescent image is calculated according to the calculation formula W=b*M+d*N; wherein b and d are preset coefficients.

[0014] As a further solution of the present invention: S2 also includes calculating the average of these mass scores if there are two or more mass scores corresponding to any voltage value, and taking the average as the unique mass score corresponding to the voltage value, and constructing a point map based on the unique mass score.

[0015] As a further solution of the present invention: the horizontal coordinate of the point corresponding to the maximum mass fraction is extracted from the point map, and the horizontal coordinate is set as the reference voltage value.

[0016] As a further solution of the present invention: if there are two or more points with the same quality score and both are maximum values, the points are calibrated as pending points, the coordinate positions of all points in the point map are obtained and the Euclidean distances between all points are calculated, and the Euclidean distance set U is generated. The clustering control radius R is set, and any pending point is taken as the center, and the point density P within the control radius R is calculated. The pending point with the largest point density is selected as the corresponding horizontal coordinate, and the horizontal coordinate is calibrated as the reference voltage value.

[0017] As a further solution of the present invention: the specific calculation process of the point density is:

[0018]

[0019] P=i / (πR 2 );

[0020] Among them, u is the sum of the Euclidean distance data values ​​of all points, I is the Euclidean distance between any points, and i is the number of points within the control radius R.

[0021] As a further solution of the present invention: in S2, the specific expression formula of the fitting function is:

[0022] The point map is fitted by the least square method to obtain the fitting line equation: W=k×X+b, where k represents the slope of the fitting line, b represents the intercept of the fitting line, and k and b are constants.

[0023] As a further solution of the present invention: in S3, the specific process of correction is:

[0024] The difference between the reference voltage value and the voltage value corresponding to the current print is calculated, and the difference is substituted into the fitting function to calculate the correction score. The fluorescent image quality score corresponding to the current print is summed with the correction score to obtain the corrected image quality score of the print.

[0025] Beneficial effects of the present invention:

[0026] The present invention first sets a detection cycle, then continuously detects the same printed product within the detection cycle, and constructs a fitting function of the relationship between the power supply voltage value and the printed flat image quality score by monitoring the power supply voltage. It can be understood that through the fitting function, the influence of different voltage values ​​on the image quality can be quantified, thereby providing a more accurate reference for quality control, and the quality of the printed product can be more accurately evaluated, thereby improving the accuracy of quality control. Traditional printed product quality control methods often rely on fixed image matching algorithms and defect detection algorithms, and are easily affected by changes in light sources, resulting in good printing quality. Affected by changes in light sources, there are deviations during color proofreading, which are mistakenly considered to be poor effects, resulting in misjudgment or missed judgment. The present invention can dynamically adjust the image quality evaluation standard by introducing a voltage monitoring and quality score correction mechanism, reduce misjudgment and missed judgment caused by factors such as voltage fluctuations, and can timely discover problems in the printing process and make adjustments through real-time monitoring and dynamic adjustment, thereby avoiding the generation of a large number of waste products. At the same time, due to the reduction of the possibility of misjudgment and missed judgment, the workload of manual re-inspection is also reduced, thereby improving production efficiency as a whole. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The present invention will be further described below in conjunction with the accompanying drawings.

[0028] Figure 1 It is a schematic flow chart of a digital printing quality detection method based on image analysis of the present invention. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0030] See also Figure 1 As shown, the present invention is a digital printing quality detection method based on image analysis, comprising the following steps:

[0031] S1, within a preset detection period, monitoring the voltage value of a preset light source and generating a fluctuation curve of the voltage value over time; collecting in real time the fluorescence image of the same printed matter under the irradiation of the preset light source within the detection period, generating an observed fluorescence image sequence;

[0032] S2, analyzing the observed fluorescence image sequence to obtain the quality scores of all observed fluorescence images; obtaining the corresponding time node of any observed fluorescence image, extracting the voltage value of the corresponding time node from the voltage fluctuation image, and constructing a point map with the voltage value as the horizontal coordinate and the quality score as the vertical coordinate; fitting the point map to obtain a fitting function, and the fitting function is used to fit the relationship between the quality score and the voltage value;

[0033] S3, obtaining the quality score and voltage value of the fluorescent image corresponding to the current printed product, and correcting the fluorescent image quality score of the printed product according to the fitting function. If the corrected image quality score is greater than or equal to a preset threshold, the printed product is judged to be qualified; if the corrected image quality score is less than the preset threshold, the printed product is judged to be unqualified.

[0034] The present invention first sets a detection cycle, then continuously detects the same printed product within the detection cycle, and constructs a fitting function of the relationship between the power supply voltage value and the printed flat image quality score by monitoring the power supply voltage. It can be understood that through the fitting function, the influence of different voltage values ​​on the image quality can be quantified, thereby providing a more accurate reference for quality control, and the quality of the printed product can be more accurately evaluated, thereby improving the accuracy of quality control. Traditional printed product quality control methods often rely on fixed image matching algorithms and defect detection algorithms, and are easily affected by changes in light sources, resulting in good printing quality. Affected by changes in light sources, there are deviations during color proofreading, which are mistakenly considered to be poor effects, resulting in misjudgment or missed judgment. The present invention can dynamically adjust the image quality evaluation standard by introducing a voltage monitoring and quality score correction mechanism, reduce misjudgment and missed judgment caused by factors such as voltage fluctuations, and can timely discover problems in the printing process and make adjustments through real-time monitoring and dynamic adjustment, thereby avoiding the generation of a large number of waste products. At the same time, due to the reduction of the possibility of misjudgment and missed judgment, the workload of manual re-inspection is also reduced, thereby improving production efficiency as a whole.

[0035] It is worth noting that voltage fluctuations of the light source will cause dynamic changes in the brightness and color temperature of the light. Changes in color temperature will directly affect the hue of the image, and hue differences will usually cause changes in the image brightness distribution, which in turn affects the calculation of pixel brightness values. This difference will cause the pixel values ​​at the same position to change under the illumination of light sources with different color temperatures, thereby causing errors in the comparison or matching process. It is understandable that hue is a key factor in determining the color of an image, and hue differences will directly affect the color value of each pixel, causing the color of the pixels in the image to change. At the same time, changes in hue are usually accompanied by changes in saturation, which reflects the concentration or purity of the color. When the saturation is higher, the color appears brighter, while when the saturation is lower, the color appears darker. Therefore, differences in hue and saturation will not only change the color performance of the pixel, but also affect the overall visual effect and pixel value of the image, thereby affecting the accuracy and contrast of the image, resulting in differences in the quality scores of the same print in actual situations.

[0036] In a preferred case of this embodiment, in S2, the calculation process of the mass fraction is:

[0037] The observed fluorescence image is gray-processed to obtain a gray-scale image. A rectangular coordinate system is established with the central pixel of the gray-scale image as the origin to generate the coordinates (x, y) of all pixels in the gray-scale image. The gray-scale values ​​P(x, y) of all pixels in the gray-scale image are identified. For the corresponding pixels in the template image, their gray-scale values ​​P′(x′, y′) are recorded. The gray-scale difference D of the pixels at the same position between the gray-scale image and the template image is calculated by the gray-scale difference formula D=|Pi(x, y)-Pi'(x', y')|. The gray-scale differences of all pixels are accumulated to obtain the total gray-scale difference M.

[0038] The fluorescent image and the template image are respectively converted into the HIS color space to obtain the corresponding HSV color matrix, the Haar transform is applied to the HSV color matrix, and the matrix after the Haar transform is normalized to obtain the color feature vector a of the fluorescent head image and the color feature vector b of the template image; the vector cosine value N of the color feature vector a and the color feature vector b is calculated;

[0039] The quality score of any fluorescent image is calculated according to the calculation formula W=b*M+d*N; wherein b and d are preset coefficients.

[0040] In another preferred situation of the present embodiment, S2 further includes calculating the average of these mass scores if there are two or more mass scores corresponding to any voltage value, and taking the average as the unique mass score corresponding to the voltage value, and constructing a point map according to the unique mass score.

[0041] In another preferred embodiment of the present invention, the abscissa of the point corresponding to the maximum value of the mass fraction is extracted from the point map, and the abscissa is set as the reference voltage value.

[0042] By using the voltage value corresponding to the maximum quality score as the reference voltage, it can be ensured that the evaluation is performed under the condition of optimal image quality, which helps to reduce the impact of image quality changes caused by voltage fluctuations on the evaluation results, thereby improving the accuracy of the overall evaluation.

[0043] In another preferred situation of the present embodiment, if there are two or more points with the same quality scores and both are maximum values, the points are calibrated as pending points, the coordinate positions of all points in the point map are obtained and the Euclidean distances between all points are calculated, and the Euclidean distance set U is generated. The clustering control radius R is set, and any pending point is taken as the center. The point density P within the control radius R is calculated, and the horizontal coordinate corresponding to the pending point with the largest point density is selected, and the horizontal coordinate is calibrated as the reference voltage value.

[0044] In another preferred embodiment of the present invention, the specific calculation process of the point density is as follows:

[0045]

[0046] P=i / (πR 2 );

[0047] Among them, u is the sum of the Euclidean distance data values ​​of all points, I is the Euclidean distance between any points, and i is the number of points within the control radius R.

[0048] In another preferred embodiment of the present invention, in S2, the specific expression formula of the fitting function is:

[0049] The point map is fitted by the least square method to obtain the fitting line equation: W=k×X+b, where k represents the slope of the fitting line, b represents the intercept of the fitting line, and k and b are constants.

[0050] In another preferred embodiment of the present invention, in S3, the specific process of correction is as follows:

[0051] The difference between the reference voltage value and the voltage value corresponding to the current print is calculated, and the difference is substituted into the fitting function to calculate the correction score. The fluorescent image quality score corresponding to the current print is summed with the correction score to obtain the corrected image quality score of the print.

[0052] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A digital printing quality detection method based on image analysis, characterized in that: The following steps are involved: S1, within a preset detection period, monitoring the voltage value of a preset light source and generating a fluctuation curve of the voltage value over time; collecting in real time the fluorescence image of the same printed matter under the irradiation of the preset light source within the detection period, generating an observed fluorescence image sequence; S2, analyzing the observed fluorescence image sequence to obtain the quality scores of all observed fluorescence images; obtaining the corresponding time node of any observed fluorescence image, extracting the voltage value of the corresponding time node from the voltage fluctuation image, and constructing a point map with the voltage value as the horizontal coordinate and the quality score as the vertical coordinate; Fitting the point map to obtain a fitting function, wherein the fitting function is used to fit the relationship between the mass fraction and the voltage value; S3, obtaining the quality score of the fluorescent image corresponding to the current printed product, and correcting the quality score of the fluorescent image of the printed product according to the fitting function, and if the corrected image quality score is greater than or equal to a preset threshold, the printed product is determined to be qualified; If the corrected image quality score is less than a preset threshold, the printed product is judged to be unqualified.

2. The digital printing quality detection method based on image analysis according to claim 1, characterized in that: In S2, the calculation process of the mass fraction is: The observed fluorescence image is gray-processed to obtain a gray-scale image. A rectangular coordinate system is established with the central pixel of the gray-scale image as the origin to generate the coordinates (x, y) of all pixels in the gray-scale image. The gray-scale values ​​P(x, y) of all pixels in the gray-scale image are identified. For the corresponding pixels in the template image, their gray-scale values ​​P′(x′, y′) are recorded. The gray-scale difference D of the pixels at the same position between the gray-scale image and the template image is calculated by the gray-scale difference formula D=|Pi(x, y)-Pi'(x', y')|. The gray-scale differences of all pixels are accumulated to obtain the total gray-scale difference M. The fluorescent image and the template image are respectively converted into the HIS color space to obtain the corresponding HSV color matrix, the Haar transform is applied to the HSV color matrix, and the matrix after the Haar transform is normalized to obtain the color feature vector a of the fluorescent head image and the color feature vector b of the template image; the vector cosine value N of the color feature vector a and the color feature vector b is calculated; The quality score of any fluorescent image is calculated according to the calculation formula W=b*M+d*N; wherein b and d are preset coefficients.

3. The digital printing quality detection method based on image analysis according to claim 1, characterized in that: In the S2, if there are two or more mass scores corresponding to any voltage value, the average of these mass scores is calculated, and the average is used as the unique mass score corresponding to the voltage value, and a point map is constructed according to the unique mass score.

4. The digital printing quality detection method based on image analysis according to claim 3 is characterized in that: The horizontal coordinate of the point corresponding to the maximum mass fraction is extracted from the point map, and the horizontal coordinate is set as the reference voltage value.

5. The digital printing quality detection method based on image analysis according to claim 4 is characterized in that: If there are two or more points with the same quality score and both are maximum values, the points are calibrated as pending points, the coordinate positions of all points in the point map are obtained, and the Euclidean distances between all points are calculated to generate the Euclidean distance set U, set the clustering control radius R, and take any pending point as the center to calculate the point density P within the control radius R, select the horizontal coordinate corresponding to the pending point with the largest point density, and calibrate the horizontal coordinate as the reference voltage value.

6. The digital printing quality detection method based on image analysis according to claim 5, characterized in that: The specific calculation process of point density is: P=i / (πR 2 ); Among them, u is the sum of the Euclidean distance data values ​​of all points, I is the Euclidean distance between any points, and i is the number of points within the control radius R.

7. The digital printing quality detection method based on image analysis according to claim 1 is characterized in that: In S2, the specific expression formula of the fitting function is: The point map is fitted by the least square method to obtain the fitting line equation: W=k×X+b, where k represents the slope of the fitting line, b represents the intercept of the fitting line, and k and b are constants.

8. The digital printing quality detection method based on image analysis according to claim 1, characterized in that: In S3, the specific process of correction is: The difference between the reference voltage value and the voltage value corresponding to the current print is calculated, and the difference is substituted into the fitting function to calculate the correction score. The fluorescent image quality score corresponding to the current print is summed with the correction score to obtain the corrected image quality score of the print.

Images