Image recognition-based printing quality detection correction method and system

Abstract

The present application relates to the technical field of printing quality inspection, and provides a printing quality detection correction method and system based on image recognition, which comprises the following steps: after each color area is divided, the brightness difference value of each color area is calculated to obtain the maximum color difference influence caused by uneven light, then the detection image of the actual printed matter is divided according to the color area, the printed matter is preliminarily judged to be qualified through a preset model combined with the brightness difference value, the brightness is corrected when the color difference value meets the model, and whether the actual color difference passes the printing quality inspection is judged, so that the quality inspection efficiency and accuracy of the printed matter are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of printing quality inspection, and in particular to a printing quality detection correction method and system based on image recognition. BACKGROUND

[0002] In the printing industry, lighting is an important guarantee for the quality of printed matter, so in the printing workshop, light source uniformity is very important, and the difference in illuminance needs to be controlled within a certain range. If the lighting is not uniform, when the printed matter is finally inspected, the color difference caused by the difference in light will cause errors in color judgment during the quality inspection of the printed matter, affecting the accuracy of the quality inspection and causing false positives. SUMMARY

[0003] The present application provides a printing quality detection correction method based on image recognition, which solves the problem of printing quality inspection accuracy affected by lighting in the prior art.

[0004] The present application provides a printing quality detection correction method based on image recognition, which solves the problem of printing quality inspection accuracy affected by lighting in the prior art.

[0005] A standard printed image is obtained, and the image is divided according to different colors to obtain a plurality of color regions of interest; after receiving a printing start signal, a first illumination image is obtained, and the first maximum brightness difference of each color region of interest in the first illumination image is calculated;

[0006] An image of the current printed matter to be inspected is obtained to obtain a first detection image, and a plurality of points are sampled in each color region of interest in the first detection image, the Lab channel value of each point is calculated, and the brightness correction judgment model is substituted into the preset brightness correction judgment model to determine whether it meets the model, and the brightness correction judgment model is specifically:

[0007]

[0008] Wherein, DE1 is the maximum color difference value, DE2 is the minimum color difference value, Delta L is the printing brightness difference, Delta a is the red-green color difference, Delta b is the yellow-blue color difference, Delta l is the first maximum brightness difference, and e is the color difference threshold value that can be recognized by the human eye.

[0009] If the brightness correction judgment model is met, the corresponding sampling point position is obtained, the sampling points of the first detection image are corrected in brightness according to the color step of the sampling point position in the first illumination image, the sampling correction value is obtained, and the color difference is calculated to determine whether the obtained color difference value is less than the color difference threshold value, if so, the color difference of the current printed matter meets the requirements.

[0010] Optionally, the plurality of points are sampled in each color region of interest in the first detection image, and the method further comprises:

[0011] Obtaining a pixel density corresponding to a camera image size, and calculating a ratio of the first maximum luminance difference value and the luminance range; multiplying the pixel density and the ratio of the first maximum luminance difference value and the luminance range to obtain a first detection image sampling point distribution density; obtaining a corresponding sampling point number according to the sampling point distribution density and the area of each color region of interest, and sampling.

[0012] Optionally, after judging whether the color difference value is less than the color difference threshold value, the method further comprises:

[0013] If the color difference value is greater than the color difference threshold value, it is judged that the color difference of the current printed matter does not meet the requirements, and the corresponding color region of interest and the color difference value are recorded and sent to the quality inspection personnel.

[0014] The second aspect of the application provides a printing quality detection correction system based on image recognition, comprising:

[0015] The area luminance calculation module is configured to obtain a standard printed image, divide the image according to different colors, and obtain a plurality of color regions of interest; after receiving a printing start signal, obtain a first illumination image, and calculate a first maximum luminance difference value of each color region of interest in the first illumination image.

[0016] The color difference preliminary judgment module is configured to obtain an image of a current printed matter to be inspected, obtain a first detection image, perform multi-point sampling on each color region of interest in the first detection image, calculate Lab channel values of each point, and substitute the values into a preset luminance correction judgment model to judge whether the values meet the model, wherein the luminance correction judgment model is specifically:

[0017]

[0018] wherein ΔE1 is a corrected maximum color difference value, ΔE2 is a corrected minimum color difference value, ΔL is a printing luminance difference, Δa is a red-green chroma difference, Δb is a yellow-blue chroma difference, Δl is the first maximum luminance difference value, and e is a color difference threshold value recognizable by the human eye.

[0019] The printing quality inspection correction module is configured to, if the luminance correction judgment model is met, obtain a corresponding sampling point position, perform luminance correction on the sampling points of the first detection image according to the sampling point position color step in the first illumination image, obtain a sampling correction value, perform color difference calculation, judge whether a obtained color difference value is less than a color difference threshold value, and if so, the color difference of the current printed matter meets the requirements.

[0020] Optionally, in the color difference preliminary judgment module, the multi-point sampling according to each color region of interest in the first detection image further comprises:

[0021] Obtain the pixel density corresponding to the camera image size, and calculate the ratio of the first maximum brightness difference value and the brightness range; multiply the pixel density and the ratio of the first maximum brightness difference value and the brightness range to obtain the first detection image sampling point distribution density; obtain the corresponding sampling point number according to the sampling point distribution density and the area of each color region of interest, and perform sampling.

[0022] Optionally, in the printed quality inspection correction module, after judging whether the color difference value is less than the color difference threshold, the method further comprises:

[0023] If the color difference value is greater than the color difference threshold, it is judged that the color difference of the current printed product does not meet the requirements, and the corresponding color region of interest and color difference value are recorded and sent to the quality inspection personnel.

[0024] The third aspect of the present application provides a printed quality detection correction method and device based on image recognition, the device comprising a processor and a memory:

[0025] The memory is used to store program code and transmit the program code to the processor;

[0026] The processor is used to execute the method according to the instructions in the program code.

[0027] The fourth aspect of the present application provides a computer readable storage medium, characterized in that the computer readable storage medium is used to store program code, and the program code is used to execute the method according to any one of the first aspect of the present application.

[0028] From the above technical solutions, the present application has the following advantages: after dividing each color region, the brightness difference value of each color region is calculated to obtain the maximum color difference influence caused by uneven light, and then the detection image of the actual printed product is divided according to the color region, the printed product is preliminarily judged to be qualified by the preset model combined with the brightness difference value, the brightness is corrected according to the color difference value meeting the model, and it is judged whether the actual color difference passes the printed quality inspection, thereby improving the quality inspection efficiency and accuracy of the printed product. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0030] Figure 1A first flow chart of a printing quality detection correction method based on image recognition;

[0031] Figure 2 A system structure diagram of a printing quality detection correction system based on image recognition. DETAILED DESCRIPTION

[0032] In order to make the inventive purposes, features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the following described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0033] The present application provides a printing quality detection correction method based on image recognition, which is used to solve the problem that the printing quality detection accuracy is affected by illumination in the prior art.

[0034] Please refer to Figure 1 , Figure 1 A first flow chart of a printing quality detection correction method based on image recognition is provided in the embodiments of the present application.

[0035] S100, a standard printing image is acquired, and image division is performed according to different colors to obtain a plurality of color regions of interest; after receiving a printing start signal, a first illumination image is acquired, and a first maximum brightness difference value of each color region of interest in the first illumination image is calculated;

[0036] It should be noted that the standard printing image is set before printing work, and the printing machine performs printing according to the standard printing image. The standard printing image is a pattern composed of a plurality of colors, so the image can be segmented according to the color, and different colors are segmented. Parts of the same color are divided into the same region. The algorithm principle of color segmentation can be realized based on color space change. After the standard printing image is divided based on the color, a plurality of color regions are obtained. In addition to the region shape and size in the image, there is also a corresponding position in the image. At this time, each color region of interest is actually a mask at the corresponding position in the image, which covers the image regions of different colors. Only the region where a single color is located can be subjected to image processing and recognition.

[0037] The camera position for shooting the image in the printing machine is fixed, and when the printing machine is started, the standard color calibration plate can be placed under the shooting position when the first light image is shot and acquired, and the first light image is directly shot and acquired, at this time, the first light image can reflect the ambient light distribution during subsequent printing quality inspection; the image recognition is performed on the first light image using the color area of interest in sequence, the color area of interest is covered on the first light image as a mask, the light intensity of the same color area can be analyzed, the color step distribution can be used, the first maximum brightness difference value is obtained according to the maximum value and the minimum value in the histogram of the color step distribution, the color step refers to the brightness, and is irrelevant to the color, but the brightest white and the darkest black; the abscissa axis-X axis of the color step distribution corresponds to the pixel brightness, and the ordinate axis-Y axis represents the number of pixels of each color tone brightness in the image, the maximum value of the brightness in the color step distribution is the maximum value on the x-axis, the minimum value of the brightness is the minimum value on the x-axis, the first light image under the color calibration plate should be concentrated on a same color step pixel brightness when the light is uniform, the first maximum brightness difference value in the color area of interest of each color reflects the maximum light difference of the same color difference that can be caused during the printing quality inspection.

[0038] S200, an image of a current printing product to be inspected is acquired to obtain a first detection image, each color area of interest in the first detection image is multi-point sampled, the Lab channel value of each point is calculated, and the brightness correction judgment model is substituted into a preset brightness correction judgment model to judge whether it is consistent with the model, and the brightness correction judgment model is specifically:

[0039]

[0040] Wherein, ΔE1 is a corrected maximum color difference value, ΔE2 is a corrected minimum color difference value, ΔL is a printing brightness difference, Δa is a red-green chroma difference, Δb is a yellow-blue chroma difference, Δl is a first maximum brightness difference value, and e is a color difference threshold value that can be recognized by the human eye.

[0041] It should be noted that after printing, the printed matter is detected for printing quality, and the printed matter is imaged at the same shooting position as the first light image to obtain a first detection image of the printed matter to be detected; the first detection image is divided into color regions of interest, each color region of interest is used as a mask to identify only the same color region of the first detection image, the light source brightness on the first detection image is smooth and there is no sudden change in brightness, therefore the sampling points are uniformly spaced in the color region of interest of the same color, the Lab channel values are obtained at each sampling point, the Lab color space is a color space based on human eye perception, which can well describe the color perception of the printed matter in the user's eyes; the color difference is also calculated based on the Lab color space, the spatial distance of the Lab channel values of two points in the Lab color space is the color difference between the two points in the same color region; the color step value of the sampling point in the first detection image is between 0-255, while the brightness of the Lab color space is between 0-100, the difference value in the color step chart can be scaled to obtain the first maximum brightness difference value;

[0042] Each sampling point in the same color region is substituted into the brightness correction judgment model, and the first maximum brightness difference value corresponding to the color region of interest is also substituted into the model, when the first formula of the brightness correction judgment model is satisfied, it means that the color difference of the two points substituted into the model is large in the detection image, which has reached the degree that can be recognized by the human eye, and when the second formula of the brightness correction judgment model is satisfied, it means that although the color difference of the two points is obvious under the current ambient light, it is possible to be less than the color difference threshold of the human eye after correcting the ambient brightness, and the first detection image in this case needs to be corrected for brightness; if the first formula of the brightness correction judgment model is satisfied but the second formula is not satisfied, it means that the color difference of the two points is large, and even if the influence of uneven ambient light is excluded, it will not be less than the color difference threshold, so there is no need to correct it subsequently, and the color difference of the printed matter is necessarily unqualified;

[0043] When the color difference values of any two points in each color region of interest in the first detection image do not satisfy the first formula of the brightness correction judgment model, it indicates that even if the environmental brightness distribution is opposite to the actual color difference, the color difference in the image is weakened by the environmental brightness, and there is no obvious color difference in the corrected image that can be recognized by the human eye, and it can be determined that the color difference of the printed matter meets the quality inspection requirements. When there are two points in a color region of interest in the first detection image, the color difference values do not satisfy the second formula of the brightness correction judgment model, it indicates that even if the environmental brightness distribution is the same as the actual color difference, the color difference in the image is increased by the environmental brightness, and there is still obvious color difference in the corrected image that can be recognized by the human eye, and it can be determined that the color difference of the printed matter does not meet the quality inspection requirements. When the brightness correction judgment model is met, and only the sampling points that meet the brightness correction judgment model and the sampling points that do not satisfy the second formula of the model are in the first detection image, the unevenness of the environmental brightness may affect the quality inspection accuracy of the printed matter, leading to misjudgment of the color difference of the printed matter, and it is necessary to perform subsequent brightness correction of the first detection image to obtain the real color difference on the printed matter for quality inspection.

[0044] In S300, if the brightness correction judgment model is met, the corresponding sampling point position is obtained, the sampling points of the first detection image are corrected in brightness according to the color step situation of the sampling point position in the first illumination image, the sampling correction value is obtained, and the color difference is calculated. It is determined whether the obtained color difference value is less than the color difference threshold value. If yes, the color difference of the current printed matter meets the requirements.

[0045] It should be noted that in the foregoing step S200, sampling is performed in the image according to the pixel coordinates, and each sampling point has a corresponding image pixel coordinate. Since the shooting conditions of the first illumination image and the first detection image are consistent, the environmental illumination at the same pixel coordinate of the sampling points is consistent. The color uniformity step in the first illumination image only reflects the brightness difference, while the Lab channel value in the first detection image reflects the color situation of the printed matter observed by the human eye, which combines the light and dark situations. In the foregoing step S200, only the most extreme value situation is calculated according to the first maximum brightness difference value of each color region of interest and the Lab channel value of each sampling point, and the actual environmental light distribution situation is not considered. That is, the color difference in the Lab color space in the first detection image may be smaller or larger. When it is identified in step S200 that the color difference value is affected by the first maximum brightness difference value and is less than the color difference threshold value, subsequent correction calculation is needed.

[0046] The color step difference values at the coordinates of the two sampling points in the first illumination image are obtained, and the actual color difference value is calculated in combination with the brightness difference values in the first detection image. For example, the two sampling points are m point and n point. In the sampling correction value, the red-green chroma and the yellow-blue chroma are unchanged, and the printed brightness difference is changed to (L m -L n )+(lm —l n ), and the color difference value obtained at this time is accurate and excludes the influence of uneven illumination. Comparison with the color difference threshold value can improve the accuracy of print quality inspection and avoid false inspection and missed inspection. The printing workshop is generally a closed room, and the ambient light is not affected by time and outdoor light. The ambient light of the first illumination image detected when the foregoing steps are started will also be maintained in subsequent quality inspection. If the printing duration is relatively long or the printing workshop is affected by outdoor changing light, step S100 can be performed again at a preset time to acquire the first illumination image and correct the first maximum brightness difference value.

[0047] In this embodiment, by dividing each color region, the brightness difference value of each color region is calculated to obtain the maximum color difference influence caused by uneven light. Then, in the detection image of the actual printed matter, the color regions are divided, and the printed matter is preliminarily judged according to the preset model combined with the brightness difference value. The color difference value that meets the model is subjected to brightness correction, and it is judged whether the actual color difference passes the print quality inspection. The print quality inspection efficiency and accuracy are improved.

[0048] The foregoing is a detailed description of a first embodiment of a print quality detection correction method based on image recognition provided by the present application. The following is a detailed description of a second embodiment of a print quality detection correction method based on image recognition provided by the present application.

[0049] In this embodiment, a print quality detection correction method based on image recognition is further provided. In the foregoing step S200, each color region of interest is subjected to multi-point sampling in the first detection image, and the specific steps further include:

[0050] The pixel point density corresponding to the camera image size is acquired, and the ratio of the first maximum brightness difference value to the brightness range is calculated. The pixel point density is multiplied by the ratio of the first maximum brightness difference value to the brightness range to obtain the sampling point distribution density of the first detection image. The corresponding sampling point number is obtained according to the sampling point distribution density and the area of each color region of interest, and sampling is performed.

[0051] It should be noted that the number of pixels in the color step graph reflects the brightness of each pixel, and the size of the first detection image captured by the camera is fixed, and the number of pixels in the image is also fixed; the first maximum brightness difference of the color step distribution reflects the brightness interval of the pixel, and the brightness change in the image is smooth and will not be abrupt, and in the case of constant image size, the larger the brightness change interval, the greater the brightness difference between the same distance pixels in the image, and the sampling points in the first detection image need to be selected as much as possible to the maximum and minimum values of the brightness, so the distribution density of the sampling points needs to be proportional to the brightness interval, for example, the image pixel density is a per square meter, the color step is concentrated in 150-200 in the color step graph, and the color step brightness range is 0-255, the ratio of the first maximum brightness difference to the brightness range is 50 / 255, and the sampling point distribution density of the first detection image is 10a / 51 per square meter, and the sampling point distribution density is rounded up.

[0052] After dividing each color region of interest, the corresponding area is obtained, the number of sampling points in each color region of interest can be calculated according to the area and the sampling point distribution density, and the distance between each sampling point can be obtained according to the sampling point distribution density, and the sampling points are uniformly arranged in each color region of interest; and to ensure subsequent color difference comparison, at least two sampling points are needed in one color region of interest.

[0053] Further, after the step S300 of determining whether the color difference value is less than the color difference threshold, the method further comprises: if the color difference value is greater than the color difference threshold, determining that the color difference of the current printed matter does not meet the requirements, recording the corresponding color region of interest and the color difference value, and sending to the quality inspection personnel; after the printed matter has obvious color difference, in order to avoid the quality problem of subsequent printed matters or to repair the current defective printed matter, the position area of the current color difference and how to adjust the uniformity to repair the color difference can be sent to the quality inspection personnel, and the maintenance efficiency of the quality inspection personnel on the subsequent printing is improved.

[0054] The above is a detailed description of the first aspect of the application, and the following is a detailed description of an embodiment of a printing quality detection and correction system based on image recognition provided by the second aspect of the application.

[0055] Please refer to Figure 2 , Figure 2 is a structure diagram of a printing quality detection and correction system based on image recognition. The embodiment provides a printing quality detection and correction system based on image recognition, which comprises:

[0056] The regional brightness calculation module 10 is configured to acquire a standard printed image, divide the image according to different colors, and obtain a plurality of color regions of interest; after receiving a printing start signal, acquire a first illumination image, and calculate a first maximum brightness difference of each color region of interest in the first illumination image;

[0057] The color difference preliminary judgment module 20 is configured to acquire an image of a current printed product to be detected, obtain a first detection image, perform multi-point sampling on each color region of interest in the first detection image, calculate Lab channel values of each point, and input the Lab channel values into a preset brightness correction judgment model to determine whether the Lab channel values meet the model, wherein the brightness correction judgment model is specifically:

[0058]

[0059] wherein ΔE1 is a corrected maximum color difference value, ΔE2 is a corrected minimum color difference value, ΔL is a printing brightness difference, Δa is a red-green chroma difference, Δb is a yellow-blue chroma difference, Δl is the first maximum brightness difference value, and e is a color difference threshold value that can be recognized by a human eye.

[0060] The printed product quality inspection correction module 30 is configured to, if the Lab channel values meet the brightness correction judgment model, acquire corresponding sampling point positions, correct brightness of the sampling points in the first detection image according to a color step of the sampling point positions in the first illumination image, obtain a sampling correction value, and perform color difference calculation to determine whether a color difference value obtained is less than the color difference threshold value, and if so, the color difference of the current printed product meets the requirement.

[0061] Further, in the color difference preliminary judgment module 20, the multi-point sampling on each color region of interest in the first detection image specifically further includes:

[0062] acquiring a pixel point density corresponding to a camera image size, calculating a ratio of the first maximum brightness difference value to a brightness range, multiplying the pixel point density and the ratio of the first maximum brightness difference value to the brightness range to obtain a sampling point distribution density of the first detection image, and obtaining a corresponding sampling point number according to the sampling point distribution density and an area of each color region of interest, and performing sampling.

[0063] Further, in the printed product quality inspection correction module 30, after determining whether the color difference value is less than the color difference threshold value, the module further includes:

[0064] If the color difference value is greater than the color difference threshold value, it is determined that the color difference of the current printed product does not meet the requirement, and corresponding color regions of interest and the color difference value are recorded and sent to a quality inspector.

[0065] The third aspect of the present application further provides a printing quality detection correction method based on image recognition, comprising a processor and a memory: the memory is used for storing program codes and transmitting the program codes to the processor; the processor is used for executing the printing quality detection correction method based on image recognition according to the instructions in the program codes.

[0066] The fourth aspect of the present application provides a computer readable storage medium, characterized in that the computer readable storage medium is used for storing program codes, and the program codes are used for executing the printing quality detection correction method based on image recognition.

[0067] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described device and equipment can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0068] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0069] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. According to actual needs, part or all of the units can be selected to achieve the purpose of the present embodiment.

[0070] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of software functional unit.

[0071] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in other words, the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0072] The above-described embodiments are merely used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still make modifications to the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some of the technical features; and these modifications or replacements do not cause the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An image recognition-based printed quality detection correction method, characterized by The method comprises the following steps: acquire a standard printing image, and divide the image according to different colors to obtain a plurality of color regions of interest; after receiving a printing start signal, a first light image is acquired, the first light image is an image of ambient light irradiating on a standard pure color calibration board, image recognition is performed on the first light image by sequentially covering each color region of interest as a mask, and a first maximum brightness difference value of each color region of interest in the first light image is calculated; an image of a current printing product to be detected is acquired to obtain a first detection image, and a plurality of sampling points are sampled in each color region of interest in the first detection image, specifically including: acquiring a pixel density corresponding to the image size of the camera, and calculating a ratio of the first maximum brightness difference value to the brightness range; multiplying the pixel density and the ratio of the first maximum brightness difference value to the brightness range to obtain a first detection image sampling point distribution density; obtaining the corresponding sampling point number according to the sampling point distribution density and the area of each color region of interest, and sampling; calculating the Lab channel value of each point, and substituting each pixel point in the same color region into a preset brightness correction judgment model to judge whether it conforms to the model, the brightness correction judgment model is specifically: wherein ΔE1 is a corrected maximum color difference value, ΔE2 is a corrected minimum color difference value, ΔL is a printing brightness difference, Δa is a red-green chroma difference, Δb is a yellow-blue chroma difference, Δl is a first maximum brightness difference value, and e is a color difference threshold value that can be recognized by the human eye; if any of the brightness correction judgment model is not met, the corresponding sampling point does not need to be corrected in brightness; if the brightness correction judgment model is met, the position of the corresponding sampling point is acquired, the sampling points of the first detection image are corrected in brightness according to the color step situation of the sampling point position in the first light image to obtain a sampling correction value, and color difference calculation is performed to judge whether the obtained color difference value is less than the color difference threshold value, if yes, the color difference of the current printing product meets the requirements.

2. The image recognition-based printed quality detection correction method according to claim 1, characterized in that, After judging whether the obtained color difference value is less than the color difference threshold value, the method further comprises the following steps: if the color difference value is greater than the color difference threshold value, it is judged that the color difference of the current printing product does not meet the requirements, and the corresponding color region of interest and the color difference value are recorded and sent to a quality inspection personnel.

3. An image recognition-based printed quality detection correction system, characterized by, The method comprises the following steps: a region brightness calculation module is configured to acquire a standard printing image, and divide the image according to different colors to obtain a plurality of color regions of interest; after receiving a printing start signal, a first light image is acquired, the first light image is an image of ambient light irradiating on a standard pure color calibration board, image recognition is performed on the first light image by sequentially covering each color region of interest as a mask, and a first maximum brightness difference value of each color region of interest in the first light image is calculated; The color difference preliminary judgment module is configured to obtain an image of a current printed product to be detected to obtain a first detection image, and to perform multi-point sampling on each color region of interest in the first detection image. Specifically, the color difference preliminary judgment module is configured to obtain a pixel density corresponding to a camera image size, and to calculate a ratio of a first maximum brightness difference value to a brightness range; to multiply the pixel density by the ratio of the first maximum brightness difference value to the brightness range to obtain a first detection image sampling point distribution density; to obtain a corresponding sampling point number according to the sampling point distribution density and an area of each color region of interest, and to perform sampling; to calculate Lab channel values of each point, and to input each pixel point in the same color region into a preset brightness correction judgment model to judge whether the pixel point meets the model, the brightness correction judgment model being specifically: where ΔE1 is a corrected maximum color difference value, ΔE2 is a corrected minimum color difference value, ΔL is a printing brightness difference, Δa is a red-green chroma difference, Δb is a yellow-blue chroma difference, Δl is a first maximum brightness difference value, and e is a color difference threshold value recognizable by a human eye. The printing quality detection correction module is configured to, if any of the brightness correction judgment model is not met, not perform brightness correction on the corresponding sampling point; and if the brightness correction judgment model is met, to obtain a corresponding sampling point position, to perform brightness correction on the sampling point of the first detection image according to a color step situation of the sampling point position in the first illumination image to obtain a sampling correction value, and to perform color difference calculation to judge whether a color difference value obtained is less than the color difference threshold value, if so, the color difference of the current printed product meets the requirements.

4. The image recognition-based printing quality detection correction system according to claim 3, wherein In the printing quality detection correction module, after judging whether the color difference value is less than the color difference threshold value, the following is further included: If the color difference value is greater than the color difference threshold value, it is judged that the color difference of the current printed product does not meet the requirements, and a corresponding color region of interest and color difference value are recorded and sent to a quality inspector.

5. An image recognition-based printed matter quality detection correction apparatus characterized by comprising: The device includes a processor and a memory: The memory is configured to store program code and transmit the program code to the processor; The processor is configured to execute the instructions in the program code to perform the printing quality detection correction method based on image recognition according to any one of claims 1-2.

6. A computer-readable storage medium, characterized in that, The computer readable storage medium is configured to store program code, and the program code is configured to perform the printing quality detection correction method based on image recognition according to any one of claims 1-2.

Images