Image detection method and system for correcting wrong binding of books and periodicals

Abstract

The invention relates to the technical field of image detection, in particular to an image detection method and system for correcting wrong binding of books and periodicals. The method comprises the following steps: performing image scanning on books and periodicals to be corrected to generate a book and periodical image set to be corrected; space coordinate system construction is carried out on the to-be-corrected book and periodical image set, and a book and periodical theme-background space coordinate system is obtained; performing composition visual coordination perception analysis on the to-be-corrected book and periodical image set according to the book and periodical theme-background space coordinate system, and performing visual transition color conflict identification to obtain visual transition color conflict data; based on the visual transition color conflict data, carrying out binding wrong pasting mode identification, and carrying out coordinate orientation adjustment on a book theme-background space coordinate system to obtain image binding wrong pasting orientation adjustment data; and carrying out automatic execution program design on the wrong binding and pasting direction adjustment data. According to the invention, the image detection technology is more perfect through accurate recognition of the binding image.

Description

Technical Field

[0001] The invention relates to the technical field of image detection, and in particular to an image detection method and system for correcting mis-sticking of book and periodical binding. Background Art

[0002] There are a large number of books and periodicals with various types. It is easy for the content to be wrongly pasted or missed during the binding process. If these problems are not discovered in time during the inspection of the finished product, it will lead to a decline in product quality and even cause serious economic losses and customer complaints. Therefore, it is urgent to introduce an automated detection method to improve the binding quality and reduce production costs. Automatic detection of binding errors in books and periodicals is achieved through machine vision and image processing algorithms. In this technical context, the system usually uses a high-precision camera or scanning device to capture images of each bound book and periodical. Through OCR (optical character recognition), image matching, feature extraction and other technical means, the bound image data is compared with the pre-stored standard page image to detect wrong pages, duplicate pages or missed pages. In particular, modern image detection systems can recognize text content in different formats, languages ​​and fonts, and have strong adaptability. The system can also be linked with the production line. When a binding error is found, it will automatically alarm and stop production, or automatically correct it through a robot. However, a conventional image detection method for correcting book binding errors has problems such as inaccurate visual color conflict recognition of image binding errors and large errors in image orientation adjustment. Summary of the invention

[0003] Based on this, it is necessary to provide an image detection method and system for correcting mis-sticking of book binding to solve at least one of the above technical problems.

[0004] To achieve the above object, an image detection method for correcting mis-sticking of book binding is provided, the method comprising the following steps:

[0005] Step S1: Scan the image of the book or periodical to be corrected to generate a set of book or periodical images to be corrected; calculate the subject-background ratio of the book or periodical image set to be corrected to obtain subject-background ratio data; construct a spatial coordinate system for the book or periodical image set to be corrected according to the subject-background ratio data to obtain a book or periodical subject-background spatial coordinate system;

[0006] Step S2: performing composition visual coordination perception analysis on the book image set to be corrected according to the book theme-background space coordinate system to obtain composition visual coordination perception data; performing visual transition color conflict recognition according to the composition visual coordination perception data to obtain visual transition color conflict data;

[0007] Step S3: performing binding error pattern recognition based on the visual transition color conflict data to obtain image binding error pattern data; adjusting the coordinate orientation of the book theme-background space coordinate system according to the image binding error pattern data to obtain image binding error orientation adjustment data;

[0008] Step S4: Design an automated execution program for the binding mis-adherence orientation adjustment data to obtain an automated binding mis-adherence adjustment program; embed the automated binding mis-adherence adjustment program into the image detection system to execute the binding mis-adherence image detection method.

[0009] The present invention first generates a set of images of books and periodicals to be corrected by performing high-quality image scanning on the books and periodicals to be corrected. This process not only ensures the digital preservation of the contents of the books and periodicals, but also lays a foundation for subsequent data analysis. Subsequently, by calculating the proportion of the subject and the background, the structure and visual layout of the contents of the books and periodicals can be deeply understood. These data can not only reveal the relative relationship between the main subject and the background, but also provide a scientific basis for constructing a spatial coordinate system. The constructed book and periodical theme-background spatial coordinate system makes the subsequent analysis more accurate and can effectively quantify the visual content of the books and periodicals. The beneficial effect of this series of steps is that the visual structure of the books and periodicals can be systematically understood, data support can be provided for subsequent design and correction, and the visual effect of the books and periodicals and the effectiveness of content communication can be improved. Composition visual coordination perception analysis of the set of images of books and periodicals to be corrected is performed according to the book and periodical theme-background spatial coordinate system, which helps to systematically evaluate the visual balance and attractiveness of the books and periodicals. By analyzing the visual coordination of the composition, it can be identified which elements complement each other and which elements will cause visual confusion or disharmony, thereby helping designers to optimize typesetting and design. Further, visual transition color conflict recognition is performed based on the composition visual coordination perception data, so that the designer can accurately identify the conflict that exists in the color transition process. This conflict recognition is crucial to improving the overall aesthetics and readability of books and periodicals, and can effectively reduce visual fatigue, so that readers can have a more pleasant experience when reading. Finally, this analysis process provides a theoretical basis for the design of books and periodicals, making the correction work more scientific, reasonable and efficient. Binding mis-pasting pattern recognition based on visual transition color conflict data is an important link to ensure the binding quality of books and periodicals. By identifying the binding mis-pasting pattern in the image, it is possible to accurately determine which pages have mis-pasting phenomena, thereby obtaining the binding mis-pasting pattern data. This data provides the necessary information basis for the subsequent coordinate orientation adjustment, so that the theme-background space coordinate system of the book and periodical can be adjusted more accurately. By adjusting the coordinate orientation, image binding mis-pasting orientation adjustment data can be generated. The beneficial effect of this step is that through scientific analysis and adjustment, the overall binding accuracy and visual consistency of books and periodicals can be greatly improved, thereby improving the quality of books and periodicals, so that they can reach higher standards in appearance and readability. At the same time, this also lays the foundation for subsequent automated adjustments, making the entire correction process more efficient and accurate. Designing an automated execution program for binding mis-adherence orientation adjustment data is the key to improving correction efficiency and accuracy. This process can achieve automatic correction and repair of books and periodicals by converting the identified binding mis-adherence information into a specific automated adjustment program. Embedding the automated adjustment program into the image detection system allows the entire system to quickly and accurately execute adjustment measures when binding mis-adherence problems are detected in real time, avoiding errors and time consumption caused by manual intervention. This automated detection method not only improves processing efficiency, but also ensures the consistency and reliability of the correction results.At the same time, after the implementation of automated adjustment, the binding quality of books and periodicals will be greatly improved, which will help improve user experience and ensure the competitiveness of books and periodicals in the market. Overall, this series of steps forms an efficient and accurate book and periodical correction system, providing important support for industry standardization. Therefore, the present invention is an optimization of a traditional image detection method for correcting mis-sticking of book and periodical binding, which solves the problems of inaccurate visual color conflict recognition of image binding mis-sticking and large image orientation adjustment errors in a traditional image detection method for correcting mis-sticking of book and periodical binding, thereby improving the accuracy of visual color conflict recognition and reducing the error of image orientation adjustment.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: Scan the image of the books and periodicals to be corrected to generate an image set of the books and periodicals to be corrected;

[0012] Step S12: performing subject-background separation on the image set of books and periodicals to be corrected to obtain subject-background separation data;

[0013] Step S13: calculating the subject-background ratio of the book image set to be corrected according to the subject-background separation data to obtain subject-background ratio data;

[0014] Step S14: constructing a spatial coordinate system for the book image set to be corrected according to the subject-background ratio data to obtain a book subject-background spatial coordinate system.

[0015] The present invention gradually realizes the comprehensive analysis and processing of the books and periodicals to be corrected through the refinement process. First, in step S11, the books and periodicals to be corrected are image scanned to generate the image set of the books and periodicals to be corrected, thereby ensuring the high-quality digital preservation of the contents of the books and periodicals. This step lays the foundation for subsequent data processing and provides accurate image data for content analysis. Next, in step S12, subject-background separation is performed, which can effectively distinguish the main information in the books and periodicals from the background elements and generate subject-background separation data. This separation process enables the subsequent analysis to focus more on the subject content, reduces background interference, and improves the accuracy of information extraction. Subsequently, in step S13, the proportion of subject and background is calculated based on the subject-background separation data, which can deeply understand the relative importance and layout effect of the subject content in the books and periodicals, and provide a basis for design optimization. Finally, in step S14, the subject-background space coordinate system of the books and periodicals is constructed according to the subject-background proportion data, which not only provides a quantitative analysis tool for the visual design of the books and periodicals, but also provides a clear structural framework for the subsequent correction work. The combined benefit of this series of steps is that, through systematic analysis and processing, the visual quality and information delivery effect of books and periodicals can be significantly improved, bringing readers a better reading experience.

[0016] Preferably, step S2 comprises the following steps:

[0017] Step S21: performing composition visual coordination perception analysis on the book image set to be corrected according to the book theme-background space coordinate system to obtain composition visual coordination perception data;

[0018] Step S22: performing visual center of gravity offset recognition on the composition visual coordination perception data to obtain composition visual center of gravity offset data;

[0019] Step S23: performing color gradient level perception analysis on the book image set to be corrected according to the composition visual center of gravity offset data to obtain visual center of gravity color gradient level data;

[0020] Step S24: performing visual transition color conflict recognition on the visual center of gravity color gradient level data to obtain visual transition color conflict data.

[0021] The present invention conducts in-depth research on the composition and visual effects of the books and periodicals to be corrected through systematic analysis and processing, so as to enhance the overall visual coordination and aesthetic feeling. First, in step S21, the composition visual coordination perception analysis is performed according to the book and periodical theme-background space coordinate system to obtain the composition visual coordination perception data. The main function of this step is to evaluate the relationship between the elements in the books and periodicals and their visual balance, and provide basic information for subsequent optimization. Then, in step S22, the composition visual coordination perception data is identified by visual center of gravity offset, which can clarify the change of the visual center of gravity in the books and periodicals, and then identify which elements will cause visual imbalance or unclear focus. This identification helps designers understand the visual flow of readers and adjust the position of elements to achieve a more ideal visual effect. Subsequently, in step S23, the color gradient level perception analysis is performed according to the composition visual center of gravity offset data to generate visual center of gravity color gradient level data. This analysis provides a basis for understanding the relationship between color change and visual center of gravity in books and periodicals, so that designers can better master the use of colors and gradient effects. Finally, in step S24, visual transition color conflict recognition is performed on the visual center color gradient level data, which can effectively detect the conflicts and disharmony in color transition. The ultimate goal of this process is to achieve higher coordination and attractiveness in the visual presentation of books and periodicals through systematic analysis and optimization, thereby enhancing readers' reading experience and understanding of the content of books and periodicals. Overall, step S2 not only provides data support for book and periodical design, but also improves the effectiveness of visual communication and ensures the scientific nature of the design.

[0022] Preferably, step S21 includes the following steps:

[0023] Step S211: performing theme-background space relative coordinate mapping on the book theme-background space coordinate system to obtain theme-background relative coordinate mapping data;

[0024] Step S212: Calculating the geometric curvature of the outline of the theme element of the book or periodical theme-background space coordinate system according to the theme-background relative coordinate mapping data to obtain the geometric curvature data of the outline of the theme element;

[0025] Step S213: defining the background depth range of the book theme-background space coordinate system based on the theme-background relative coordinate mapping data to obtain background depth range data;

[0026] Step S214: performing diagonal visual tension evaluation on the image set of books and periodicals to be corrected according to the geometric curvature data of the subject element contour and the background depth of field range data, and obtaining image diagonal visual tension evaluation data;

[0027] Step S215: performing composition visual coordination perception analysis on the book and periodical image set to be corrected based on the image diagonal visual tension evaluation data to obtain composition visual coordination perception data.

[0028] In step S21, the present invention aims to improve the visual coordination and aesthetic feeling of books and periodicals through a series of detailed analyses. First, in step S211, the theme-background space coordinate system of the book and periodical is subjected to theme-background space relative coordinate mapping to generate theme-background relative coordinate mapping data. This process allows the relationship between the theme and the background to be expressed in coordinate form, providing a quantitative basis for subsequent analysis. Next, in step S212, the geometric curvature of the outline of the theme element of the book and periodical is calculated according to the theme-background relative coordinate mapping data to obtain the geometric curvature data of the outline of the theme element. The importance of this step is that it can reveal the shape characteristics and changes of the theme element and provide necessary information for optimizing the design. Then, in step S213, the background depth of field range of the book and periodical is defined based on the theme-background relative coordinate mapping data to generate background depth of field range data. The introduction of this data enables the designer to clearly define the level and depth of the background element and enhance the overall spatial effect. Then, in step S214, the diagonal visual tension is evaluated according to the geometric curvature data of the outline of the theme element and the background depth of field range data to obtain the image diagonal visual tension evaluation data. Through the analysis of diagonal visual tension, the dynamic sense and visual impact of the book image can be evaluated, providing guidance for the adjustment of the composition. Finally, in step S215, the composition visual coordination perception analysis of the book image is performed based on the image diagonal visual tension evaluation data to obtain the composition visual coordination perception data. The comprehensive effect of this series of steps is that through precise spatial mapping, geometric analysis and tension evaluation, the visual characteristics of books can be deeply understood, and comprehensive data support can be provided for subsequent design optimization, ensuring that the visual effects of books are more harmonious and enhancing the readers' visual experience. Overall, step S21 forms a systematic analysis process to ensure that the visual design of books and periodicals reaches higher standards in structure and aesthetics.

[0029] Preferably, performing diagonal visual tension evaluation on the image set of books and periodicals to be corrected according to the geometric curvature data of the subject element contour and the background depth range data comprises the following steps:

[0030] The curvature variation interval of the diagonal area of ​​the image set of books and periodicals to be corrected is calculated according to the geometric curvature data of the outline of the subject element, and the curvature variation interval of the diagonal area of ​​the element is obtained;

[0031] Performing diagonal area deformation analysis on the book and periodical image set to be corrected according to the background depth range data, obtaining background diagonal area depth deformation data;

[0032] Based on the curvature change interval of the element diagonal area and the depth deformation data of the background diagonal area, the visual perception tension of the diagonal local area is simulated to obtain the diagonal local visual perception tension data;

[0033] The primary and secondary structure continuity data of the diagonal local visual perception tension is analyzed to obtain the primary and secondary structure continuity data of the diagonal global;

[0034] According to the diagonal local visual perception tension data and the diagonal global primary and secondary structure continuity data, the diagonal visual tension of the book and periodical image set to be corrected is evaluated to obtain the image diagonal visual tension evaluation data.

[0035] In the diagonal visual tension assessment of the present invention, the curvature change interval of the diagonal area of ​​the image set of the book or periodical to be corrected is first calculated according to the geometric curvature data of the outline of the subject element, and the curvature change interval of the diagonal area of ​​the element is generated. This calculation is crucial for identifying and quantifying the shape changes of the subject elements in the book or periodical, and helps to understand how the elements visually affect the tension and dynamic sense of the overall composition. Next, the diagonal area deformation analysis is performed based on the background depth of field range data to obtain the background diagonal area depth of field deformation data. This step enables the designer to identify the depth change of the background element and evaluate its impact on the overall visual effect, thereby ensuring that the background does not interfere with the presentation of the subject element. Subsequently, the visual perception tension of the diagonal local area is simulated based on the curvature change interval of the element diagonal area and the background diagonal area depth of field deformation data to obtain the diagonal local visual perception tension data. This simulation provides an important reference for understanding the visual performance of the local area and helps the designer identify potential visual conflicts or disharmony. Next, the primary and secondary structure continuity analysis of the diagonal local visual perception tension data is performed to obtain the diagonal global primary and secondary structure continuity data. This step can ensure that the hierarchical relationship of visual elements in books and periodicals is reasonably maintained and improve the coherence of the overall composition. Finally, based on the diagonal local visual perception tension data and the diagonal global primary and secondary structure continuity data, the diagonal visual tension of the book and periodical image set to be corrected is evaluated to obtain the image diagonal visual tension evaluation data. Through this series of analysis and evaluation, designers can achieve comprehensive visual optimization of book and periodical images, ensure that each visual element coexists harmoniously in the overall design, and analyze the overall appeal and visual impact of books and periodicals.

[0036] Preferably, step S24 includes the following steps:

[0037] Step S241: Calculate the pixel difference between adjacent pixels of the visual center of gravity color gradient level data to obtain the color gradient adjacent pixel difference;

[0038] Step S242: performing a color wheel analysis of the contrast between adjacent colors on the color gradient level data of the visual center of gravity according to the color gradient adjacent pixel difference, to obtain the adjacent color contrast color wheel data;

[0039] Step S243: Calculate the hue angle of the adjacent color contrast color wheel data to obtain the adjacent color hue angle data;

[0040] Step S244: performing visual fatigue simulation based on adjacent color hue angle data, adjacent color contrast color wheel data, and color gradient adjacent pixel difference values ​​to obtain color gradient visual fatigue data;

[0041] Step S245: performing visual transition color conflict recognition on the visual center of gravity color gradient level data according to the color gradient visual fatigue data to obtain visual transition color conflict data.

[0042] Step S24 of the present invention improves the visual consistency and aesthetic effect of the image through a series of fine color analysis. First, in step S241, the pixel difference between adjacent pixels of the visual center of gravity color gradient level data is calculated to generate the color gradient adjacent pixel difference. This process can quantify the color changes between adjacent pixels, provide basic data for subsequent analysis, and help designers identify subtle differences in vision. Then, step S242 performs a contrast color wheel analysis between adjacent colors based on the color gradient adjacent pixel difference, thereby obtaining adjacent color contrast color wheel data. The key to this step is to clarify the contrast relationship between adjacent colors and help designers understand which color combinations are visually conflicting or inharmonious. Then, in step S243, the hue angle of the adjacent color contrast color wheel data is calculated to obtain the adjacent color hue angle data. This step can not only clearly display the angle relationship between colors, but also further help designers judge the similarity or difference of these colors in visual performance. Next, step S244 simulates visual fatigue based on the adjacent color hue angle data, the adjacent color contrast color wheel data, and the color gradient adjacent pixel difference to obtain the color gradient visual fatigue data. This simulation can point out the areas where the color gradient causes visual discomfort by evaluating the sense of visual fatigue, thereby providing a basis for adjusting the color scheme. Finally, in step S245, the visual transition color conflict is identified for the visual center of gravity color gradient level data based on the color gradient visual fatigue data to generate visual transition color conflict data. The goal of this step is to accurately locate the visual conflicts that occur during the color transition process to ensure the overall harmony and beauty of the book image. Through this series of analyses, the entire step S24 forms an effective image detection framework, which provides data support and optimization suggestions for correcting the problem of mis-sticking in book binding, and improves the visual effect of the image and the reader's visual experience.

[0043] Preferably, the visual fatigue simulation based on adjacent color hue angle data, adjacent color contrast color wheel data and color gradient adjacent pixel difference includes the following steps:

[0044] Performing color gamut distortion analysis based on adjacent color hue angle data, adjacent color contrast color wheel data, and color gradient adjacent pixel difference values ​​to obtain color gamut distortion data;

[0045] Performing color detail loss analysis on the color gamut distortion data to obtain color detail loss data;

[0046] Visual stimulation cumulative effect analysis is performed based on color detail loss data to obtain visual stimulation cumulative effect data;

[0047] Visual fatigue degree simulation is performed based on the color detail loss data and the visual stimulation cumulative effect data to obtain the color gradient visual fatigue degree data.

[0048] The step of simulating visual fatigue based on adjacent color hue angle data, adjacent color contrast color wheel data and color gradient adjacent pixel difference of the present invention has significant application value. First, color gamut distortion analysis is performed, and color gamut distortion data is obtained based on adjacent color hue angle data and color gradient adjacent pixel difference. This analysis reveals the distribution changes of colors in different regions, helping designers to identify areas that cause visual inconsistency, thereby improving the overall quality and consistency of the image. Next, color detail loss analysis is performed on the color gamut distortion data to obtain color detail loss data. The core of this step is to identify the loss of details that occur during the color gradient process, ensuring that important visual information is not ignored, thereby providing a basis for subsequent adjustments. Then, visual stimulation cumulative effect analysis is performed based on the color detail loss data to obtain visual stimulation cumulative effect data. This analysis can quantify the long-term impact of visual stimulation on the audience, reveal the degree of influence of color on visual fatigue after long-term reading or observation, thereby effectively preventing the occurrence of visual fatigue. Finally, visual fatigue simulation is performed based on the color detail loss data and the visual stimulation cumulative effect data to obtain color gradient visual fatigue data. Through this simulation, designers can accurately understand the degree of visual fatigue in readers under specific color combinations, and then optimize the design to ensure that the color configuration of books and periodicals is both beautiful and does not cause visual burden. The whole process not only improves the visual quality of book and periodical images, but also provides a scientific basis for correcting book and periodical binding errors.

[0049] Preferably, step S3 comprises the following steps:

[0050] Step S31: performing image edge continuity detection on the book image set to be corrected according to the visual transition color conflict data to obtain image edge continuity data;

[0051] Step S32: performing binding error pattern recognition based on the image edge continuity data and the visual transition color conflict data to obtain image binding error pattern data;

[0052] Step S33: adjusting the coordinate orientation of the book subject-background space coordinate system according to the image binding mis-pasting mode data to obtain image binding mis-pasting orientation adjustment data.

[0053] Step S3 of the present invention significantly improves the quality and consistency of the image through a series of efficient detection and adjustment processes. First, step S31 performs image edge continuity detection on the image set of books and periodicals to be corrected based on the visual transition color conflict data to generate image edge continuity data. The purpose of this detection process is to identify the smoothness and consistency of the edges in the image and ensure the natural transition of colors and shapes. By obtaining the edge continuity data, the designer can effectively identify those incoherent edges caused by color conflicts, thereby providing an important basis for subsequent adjustments. Then, in step S32, the binding mis-pasting pattern recognition is performed based on the image edge continuity data and the visual transition color conflict data to obtain the image binding mis-pasting pattern data. This process can accurately identify the potential binding error types in the book and periodical images, including position offset and color mismatch, etc., by comprehensively considering edge continuity and color conflict. Through pattern recognition, the designer can clearly understand the nature and location of the problem, so as to quickly make corresponding adjustments and optimizations to improve the accuracy and visual appeal of the overall image. Finally, in step S33, the coordinate orientation of the book theme-background space coordinate system is adjusted according to the image binding mis-pasting mode data to obtain the image binding mis-pasting orientation adjustment data. The core of this step is to ensure the coordination and harmony between the theme and background of the book by adjusting the coordinate system. Through precise orientation adjustment, the designer can optimize the visual layout of the book, ensure the clear communication of information and visual beauty, thereby enhancing the overall reading experience of the reader. In summary, step S3 realizes the precise correction of the book image through image edge detection, binding mode recognition and spatial coordinate adjustment, which provides a scientific basis and practical method for effectively solving the binding mis-pasting problem.

[0054] Preferably, step S4 comprises the following steps:

[0055] Step S41: normalizing the binding mis-pasting orientation adjustment data to obtain orientation adjustment normalized data;

[0056] Step S42: Designing an automatic execution program for the orientation adjustment normalization data to obtain an automatic adjustment program for mis-adhering of the binding;

[0057] Step S43: embedding the automatic binding mis-adherence adjustment program into the image detection system to execute the binding mis-adherence image detection method.

[0058] Step S4 of the present invention ensures efficient correction of mis-adherence in binding through normalization processing, automated program design and system embedding. First, in step S41, the orientation adjustment data of mis-adherence in binding is normalized to obtain orientation adjustment normalized data. The main purpose of this processing is to eliminate the scale differences existing in the data so that different adjustment data can be compared and analyzed under the same standard. Through normalization, the designer can more intuitively identify the relative importance of each adjustment, thereby improving the accuracy and reliability of subsequent steps. This process ensures the consistency of the data and lays a solid foundation for the subsequent automated program design. Next, in step S42, the orientation adjustment normalized data is used to design an automated execution program, thereby obtaining an automated adjustment program for mis-adherence in binding. This step uses programming technology to convert the normalized data into an operable program, which can automatically identify and adjust the mis-adherence problem in the book and periodical image. Through the design of the automated program, the correction process is significantly simplified, the necessity of manual intervention is reduced, and the work efficiency and correction accuracy are improved. In addition, the automated adjustment program can be used repeatedly, is suitable for different types of books and periodicals, and has good expansibility and adaptability. Finally, in step S43, the automatic adjustment program for binding errors is embedded in the image detection system, so that it can execute the image detection method for binding errors in real time. Through this integration, the image detection system can not only automatically identify the situation of binding errors during the detection process, but also quickly start the corresponding adjustment program when the problem is found. This automated integration improves the intelligence level of the entire system, reduces human operation errors, and improves the efficiency and consistency of image correction, thereby significantly optimizing the visual effects of books and periodicals, ensuring that readers have a better reading experience. Overall, the implementation of step S4 effectively promotes the automation and efficiency of the image detection method, and provides comprehensive technical support for solving the problem of binding errors in books and periodicals.

[0059] Preferably, the present invention further provides an image detection system for correcting mis-sticking in book binding, which is used to execute the image detection method for correcting mis-sticking in book binding as described above. The image detection system for correcting mis-sticking in book binding comprises:

[0060] The coordinate system construction module is used to scan the image of the book to be corrected to generate a set of images of the book to be corrected; calculate the subject-background ratio of the book image to be corrected to obtain subject-background ratio data; construct a spatial coordinate system for the book image to be corrected according to the subject-background ratio data to obtain a book subject-background spatial coordinate system;

[0061] The visual color conflict recognition module is used to perform composition visual coordination perception analysis on the image set of books and periodicals to be corrected according to the book and periodical theme-background space coordinate system to obtain composition visual coordination perception data; perform visual transition color conflict recognition according to the composition visual coordination perception data to obtain visual transition color conflict data;

[0062] The mis-pasting orientation adjustment module is used to identify the binding mis-pasting pattern based on the visual transition color conflict data to obtain the image binding mis-pasting pattern data; adjust the coordinate orientation of the book theme-background space coordinate system according to the image binding mis-pasting pattern data to obtain the image binding mis-pasting orientation adjustment data;

[0063] The automatic program design module is used to automatically execute program design for the binding mis-sticking orientation adjustment data to obtain the binding mis-sticking automatic adjustment program; the binding mis-sticking automatic adjustment program is embedded in the image detection system to execute the binding mis-sticking image detection method.

[0064] The beneficial effect of the present invention is that, firstly, a high-quality image scanning is performed on the books and periodicals to be corrected to generate a set of images of books and periodicals to be corrected. This process not only ensures the digital preservation of the contents of the books and periodicals, but also lays a foundation for subsequent data analysis. Subsequently, by calculating the proportion of the theme and the background, the structure and visual layout of the contents of the books and periodicals can be deeply understood. These data can not only reveal the relative relationship between the main theme and the background, but also provide a scientific basis for constructing a spatial coordinate system. The constructed book and periodical theme-background spatial coordinate system makes the subsequent analysis more accurate and can effectively quantify the visual content of the books and periodicals. The beneficial effect of this series of steps is that the visual structure of the books and periodicals can be systematically understood, data support can be provided for subsequent design and correction, and the effectiveness of the visual effect and content communication of the books and periodicals can be improved. Composition visual coordination perception analysis of the set of images of books and periodicals to be corrected is performed according to the book and periodical theme-background spatial coordinate system, which helps to systematically evaluate the visual balance and attractiveness of the books and periodicals. By analyzing the visual coordination of the composition, it can be identified which elements complement each other and which elements will cause visual confusion or disharmony, thereby helping designers to optimize typesetting and design. Further, visual transition color conflict recognition is performed based on the composition visual coordination perception data, so that the designer can accurately identify the conflict that exists in the color transition process. This conflict recognition is crucial to improving the overall aesthetics and readability of books and periodicals, and can effectively reduce visual fatigue, so that readers can have a more pleasant experience when reading. Finally, this analysis process provides a theoretical basis for the design of books and periodicals, making the correction work more scientific, reasonable and efficient. Binding mis-pasting pattern recognition based on visual transition color conflict data is an important link to ensure the binding quality of books and periodicals. By identifying the binding mis-pasting pattern in the image, it is possible to accurately determine which pages have mis-pasting phenomena, thereby obtaining the binding mis-pasting pattern data. This data provides the necessary information basis for the subsequent coordinate orientation adjustment, so that the theme-background space coordinate system of the book and periodical can be adjusted more accurately. By adjusting the coordinate orientation, image binding mis-pasting orientation adjustment data can be generated. The beneficial effect of this step is that through scientific analysis and adjustment, the overall binding accuracy and visual consistency of books and periodicals can be greatly improved, thereby improving the quality of books and periodicals, so that they can reach higher standards in appearance and readability. At the same time, this also lays the foundation for subsequent automated adjustments, making the entire correction process more efficient and accurate. Designing an automated execution program for binding mis-adherence orientation adjustment data is the key to improving correction efficiency and accuracy. This process can achieve automatic correction and repair of books and periodicals by converting the identified binding mis-adherence information into a specific automated adjustment program. Embedding the automated adjustment program into the image detection system allows the entire system to quickly and accurately execute adjustment measures when binding mis-adherence problems are detected in real time, avoiding errors and time consumption caused by manual intervention. This automated detection method not only improves processing efficiency, but also ensures the consistency and reliability of the correction results.At the same time, after the implementation of automated adjustment, the binding quality of books and periodicals will be greatly improved, which will help improve user experience and ensure the competitiveness of books and periodicals in the market. Overall, this series of steps forms an efficient and accurate book and periodical correction system, providing important support for industry standardization. Therefore, the present invention is an optimization of a traditional image detection method for correcting mis-sticking of book and periodical binding, which solves the problems of inaccurate visual color conflict recognition of image binding mis-sticking and large image orientation adjustment errors in a traditional image detection method for correcting mis-sticking of book and periodical binding, thereby improving the accuracy of visual color conflict recognition and reducing the error of image orientation adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 A schematic diagram of the steps of an image detection method for correcting mis-sticking in book and periodical binding;

[0066] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0067] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0068] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0069] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0070] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0071] To achieve this, please refer to Figure 1 to Figure 2 , an image detection method for correcting mis-sticking of book binding, the method comprising the following steps:

[0072] Step S1: Scan the image of the book or periodical to be corrected to generate a set of book or periodical images to be corrected; calculate the subject-background ratio of the book or periodical image set to be corrected to obtain subject-background ratio data; construct a spatial coordinate system for the book or periodical image set to be corrected according to the subject-background ratio data to obtain a book or periodical subject-background spatial coordinate system;

[0073] Step S2: performing composition visual coordination perception analysis on the book image set to be corrected according to the book theme-background space coordinate system to obtain composition visual coordination perception data; performing visual transition color conflict recognition according to the composition visual coordination perception data to obtain visual transition color conflict data;

[0074] Step S3: performing binding error pattern recognition based on the visual transition color conflict data to obtain image binding error pattern data; adjusting the coordinate orientation of the book theme-background space coordinate system according to the image binding error pattern data to obtain image binding error orientation adjustment data;

[0075] Step S4: Design an automated execution program for the binding mis-adherence orientation adjustment data to obtain an automated binding mis-adherence adjustment program; embed the automated binding mis-adherence adjustment program into the image detection system to execute the binding mis-adherence image detection method.

[0076] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of an image detection method for correcting mis-sticking in book binding according to the present invention. In this example, the image detection method for correcting mis-sticking in book binding includes the following steps:

[0077] Step S1: Scan the image of the book or periodical to be corrected to generate a set of book or periodical images to be corrected; calculate the subject-background ratio of the book or periodical image set to be corrected to obtain subject-background ratio data; construct a spatial coordinate system for the book or periodical image set to be corrected according to the subject-background ratio data to obtain a book or periodical subject-background spatial coordinate system;

[0078] In the embodiment of the present invention, the image scanning of the books and periodicals to be corrected is performed, and a high-resolution flatbed scanner is used to scan the entire books and periodicals page by page to ensure the clarity of the image and the integrity of each page. The resolution of the scanner is set at 600dpi or higher to avoid the loss of details caused by insufficient resolution. Each scanned image is stored in a lossless TIFF format to form a set of books and periodicals images to be corrected. Subsequently, an image processing algorithm is used to calculate the proportion of the subject-background for each image in the book and periodical image set. In order to complete this calculation, the image needs to be segmented first, and the boundaries of the main area (subject) and the secondary area (background) in the image are identified by the image histogram analysis method. The subject and the background are classified according to the brightness, color and texture features in the image by using a threshold segmentation-based technology. Then, the proportion of the subject and the background is calculated respectively by pixel area calculation, and the subject-background proportion data of each page is accurately recorded. Next, based on the subject-background proportion data of each image, a spatial coordinate system is constructed for the book and periodical image by using a mathematical geometric modeling method. The coordinate system needs to be based on the geometric center of the subject and background in each image, and calibrated by a polar coordinate or Cartesian coordinate system on a two-dimensional plane. This process uses geometric transformation technology to accurately locate the subject and background areas in the image to specific locations in the coordinate system, generating a book subject-background spatial coordinate system. This coordinate system will serve as a spatial reference frame for subsequent analysis to ensure that the positional relationship between the subject and background can be accurately expressed and used in subsequent steps.

[0079] Step S2: performing composition visual coordination perception analysis on the book image set to be corrected according to the book theme-background space coordinate system to obtain composition visual coordination perception data; performing visual transition color conflict recognition according to the composition visual coordination perception data to obtain visual transition color conflict data;

[0080] In an embodiment of the present invention, after obtaining the book theme-background space coordinate system, the book image set is subjected to composition visual coordination perception analysis based on this coordinate system. The visual coordination of the composition can be evaluated by analyzing the visual features such as the symmetry, balance and golden ratio of the image. The method combining Fourier transform and spatial frequency analysis is adopted to analyze the balanced distribution of the image in the frequency domain, and then judge the symmetry of its composition. In addition, the relative position of the center of gravity position of the image theme part and the entire image frame is calculated by mathematical statistical analysis to determine whether the image has visual balance. The result data of each page of the image is output in the form of composition visual coordination perception data. Subsequently, based on the composition visual coordination perception data, visual transition color conflict recognition is performed. The recognition of color conflict depends on the quantitative analysis of color space, and the RGB color mode is converted into the CIE Lab color space, which is closer to the human eye's perception of color. Using the color difference formula (ΔE), by calculating the color difference values ​​of different areas in the image, the parts where the color changes are too drastic or disharmonious are identified, especially between the theme and the background or in the visual transition area of ​​the image. The conflicting areas are classified using color harmony theory (such as the adjacent colors or complementary colors principle on the color wheel). The identified conflicting areas and the amount of color difference are recorded as visual transition color conflict data, which will serve as a key basis for subsequent corrections.

[0081] Step S3: performing binding error pattern recognition based on the visual transition color conflict data to obtain image binding error pattern data; adjusting the coordinate orientation of the book theme-background space coordinate system according to the image binding error pattern data to obtain image binding error orientation adjustment data;

[0082] In the embodiment of the present invention, first, the binding mis-pasting pattern recognition is performed based on the visual transition color conflict data. The key to the binding mis-pasting pattern recognition is to determine whether the image has a mis-pasting phenomenon by analyzing the arrangement of the subject and background areas in the book image and the visual transition abnormal points. A pattern recognition algorithm (such as a support vector machine or a Bayesian classifier) ​​is used to classify the abnormal areas in the visual transition color conflict data to identify the existing mis-pasting pattern. This process compares each page image with the standard template of a normal book, and detects whether each page image has significant differences in subject, background, and color transition from the previous and next pages, thereby locating the specific position of the mis-pasting. The recognition result forms the image binding mis-pasting pattern data, which not only includes the specific mis-pasting page, but also marks the specific pattern features of the mis-pasting (such as image inversion, page order reversal or wrong page, etc.). Then, according to the image binding mis-pasting pattern data, the coordinate orientation of the book theme-background space coordinate system is adjusted. The main purpose of the orientation adjustment is to correct the image direction, order, and position offset caused by mis-pasting. This adjustment uses a geometric transformation algorithm, including rotation, mirror reversal, and image order rearrangement. According to the misalignment characteristics in the misplacement pattern data, each page image is first rotated or flipped accordingly to ensure that the coordinate systems of the subject part and the background part are realigned to the standard direction. Subsequently, the image registration technology is applied to optimize the global position of the adjusted image to ensure that the relative position of the misplaced page in the entire book is corrected. Through a series of geometric transformations and image alignment operations, the image binding misplacement orientation adjustment data is output, which records the new orientation coordinates of each page image after the misplacement correction.

[0083] Step S4: Design an automated execution program for the binding mis-adherence orientation adjustment data to obtain an automated binding mis-adherence adjustment program; embed the automated binding mis-adherence adjustment program into the image detection system to execute the binding mis-adherence image detection method.

[0084] In an embodiment of the present invention, after obtaining the image binding mis-pasting orientation adjustment data, the next step is to design an automated execution program for the data. The design of the program relies on embedded control logic, and the correction operation of binding mis-pasting is performed by writing an automated script. The core of the automated execution program is to call a pre-designed image processing function according to the geometric transformation parameters (such as rotation angle, translation distance, flip direction, etc.) in the image binding mis-pasting orientation adjustment data, and perform the adjustment operation page by page. In order to ensure the automation and accuracy of the operation, the control flow logic is adopted in the program, and the corresponding processing subroutines are called for different mis-pasting modes. For example, for the page inversion situation, the rotation function is called, and for the wrong page situation, the image reordering function is called. The program corrects the mis-pasting problem of the entire book page by page through a series of batch operations, and records each operation step performed. Finally, the binding mis-pasting automated adjustment program is embedded in the image detection system to perform the binding mis-pasting image detection method. The embedding process relies on the integrated environment of the image detection system, and the automated program script is directly loaded in the system to make it seamlessly connected with the existing image detection process. The system automatically triggers the binding error correction process by monitoring the processing status of the book image in real time, and calls the automated program to make adjustments after the wrong page is found. At the same time, the system generates an operation log to record the detailed process of each detection and correction for subsequent review and proofreading. By embedding the binding error adjustment program, the entire book correction process is highly automated and accurate, ensuring that the image detection method can efficiently perform the error correction task.

[0085] Preferably, step S1 comprises the following steps:

[0086] Step S11: Scan the image of the books and periodicals to be corrected to generate an image set of the books and periodicals to be corrected;

[0087] Step S12: performing subject-background separation on the image set of books and periodicals to be corrected to obtain subject-background separation data;

[0088] Step S13: calculating the subject-background ratio of the book image set to be corrected according to the subject-background separation data to obtain subject-background ratio data;

[0089] Step S14: constructing a spatial coordinate system for the book image set to be corrected according to the subject-background ratio data to obtain a book subject-background spatial coordinate system.

[0090] In the embodiment of the present invention, the image scanning step of the books and periodicals to be corrected is performed by using professional high-precision scanning equipment to ensure that the content of each page of the books and periodicals can be completely and losslessly digitized. The scanning operation requires the use of a scanner with an optical resolution of 600dpi or higher to ensure accurate capture of all details in the books and periodicals (including images, text, page margins, binding lines, etc.). Each scanned page will be automatically cropped to a uniform size to ensure clear image boundaries and avoid redundant blank areas. In order to generate a uniform image set of books and periodicals to be corrected, all scanned images are saved in a lossless compression format (such as TIFF format) to avoid loss of details caused by compression during image processing. This process is strictly carried out in a page-by-page scanning order, and the generated image set will serve as the basic data for analysis and processing in subsequent steps. After completing the image scanning, the image set of books and periodicals to be corrected needs to undergo a subject-background separation operation. This operation is implemented through an image segmentation algorithm, and the specific method is to apply edge detection and region growth technology to separate the subject part (such as text, image body) from the background part (such as page blank, decorative elements) in the image. First, the Canny edge detection algorithm is used to identify the significant edges in the image and locate the outlines of the text and image. Then, the region growing algorithm based on pixel similarity is used to aggregate similar pixel groups to ensure the integrity of the entire subject area. For images with complex backgrounds, the color histogram segmentation technology is used to accurately separate the subject and background parts based on the color differences in the image. The final subject-background separation data includes the accurate segmentation results of each part in each page image and is saved in the form of a binary image (black and white segmentation map) or annotated map for use in subsequent steps. Based on the subject-background separation data, the proportion of the subject and background of each image in the book to be corrected is calculated. This step is completed using a regional statistical algorithm, which specifically traverses each separated image pixel by pixel and calculates the total number of pixels in the subject part and the background part. By comparing the number of subject pixels with the total number of pixels in the image, the area ratio data of the subject-background can be obtained. In order to improve the accuracy of the calculation, the morphological filtering technology is also used to preprocess the segmentation results to remove noise interference and small irregular areas to ensure the accuracy of the ratio calculation. The calculation results are output as percentages of the subject and background, and recorded as subject-background ratio data for subsequent construction of the spatial coordinate system. The spatial coordinate system is constructed for each image based on the subject-background ratio data generated in the previous step. First, the geometric center of the subject of the image is identified and used as the origin of the coordinate system. Combined with the aspect ratio of the image, the Cartesian coordinate system is used to accurately locate the image. In order to ensure that the relative positions of the subject and background in the coordinate system are consistent, the image's axis of symmetry and bounding box are further introduced. By calculating the edge position of each image and its relative distance from the subject center point, these position points are mapped to a unified spatial coordinate system.In addition, in order to detect mis-binding, the image coordinates are adjusted using an affine transformation matrix to ensure that image comparisons in the subsequent detection process can be performed in the same coordinate reference system. The constructed book theme-background space coordinate system is output in the form of a coordinate array, including the theme area and background area of ​​each image and their specific position on the page.

[0091] Preferably, step S2 comprises the following steps:

[0092] Step S21: performing composition visual coordination perception analysis on the book image set to be corrected according to the book theme-background space coordinate system to obtain composition visual coordination perception data;

[0093] Step S22: performing visual center of gravity offset recognition on the composition visual coordination perception data to obtain composition visual center of gravity offset data;

[0094] Step S23: performing color gradient level perception analysis on the book image set to be corrected according to the composition visual center of gravity offset data to obtain visual center of gravity color gradient level data;

[0095] Step S24: performing visual transition color conflict recognition on the visual center of gravity color gradient level data to obtain visual transition color conflict data.

[0096] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0097] Step S21: performing composition visual coordination perception analysis on the book image set to be corrected according to the book theme-background space coordinate system to obtain composition visual coordination perception data;

[0098] In an embodiment of the present invention, according to the book theme-background space coordinate system, the process of performing composition visual coordination perception analysis on the book image set to be corrected adopts a visual evaluation method based on geometric composition. First, the geometric position and proportion information in the theme-background space coordinate system is used to determine whether the subject part of each image is on the visual golden section line or the composition thirds line of the image, and evaluate its relative position relationship with the background part. The analysis is mainly performed by calculating the proportion of the subject in the image, the balance of the edge, and the distance between the subject area and the image boundary. The horizontal and vertical symmetry axes in the image are analyzed using a symmetry measurement algorithm, and combined with the regional shape matching technology, it is ensured that the composition between the subject and the background has good balance and coordination. Finally, the composition visual coordination perception data is output as a structured evaluation index, mainly including the deviation value of the subject position from the golden section line, the symmetry coefficient of the composition, and the overall balance score.

[0099] Step S22: performing visual center of gravity offset recognition on the composition visual coordination perception data to obtain composition visual center of gravity offset data;

[0100] In an embodiment of the present invention, after obtaining the composition visual coordination perception data, it is necessary to further identify the visual center of gravity offset in the image. This operation is achieved by calculating the visual center of gravity position of each image. First, based on the geometric proportions of the subject and background in the image, the visual center of gravity is calculated using the mass center formula, that is, the "center of mass" position of the subject area and the background area is determined in the coordinate system. In order to ensure accurate visual center of gravity calculation, the subject and background areas in the image need to be processed by center of gravity correction, and the edge detection algorithm is used to filter out noise interference. By calculating the degree of deviation of the visual center of gravity, it is determined whether the subject area in the image is excessively offset to a certain direction. The recognition process outputs the composition visual center of gravity offset data, which mainly includes the center of gravity coordinate offset value of each image and its distance deviation relative to the center of the image.

[0101] Step S23: performing color gradient level perception analysis on the book image set to be corrected according to the composition visual center of gravity offset data to obtain visual center of gravity color gradient level data;

[0102] In an embodiment of the present invention, a color gradient level perception analysis is performed on a set of books and periodicals to be corrected according to the composition visual center of gravity offset data to determine the smoothness and level of color transition in the image. First, based on the identified visual center of gravity, the image is divided into different color level regions, and the color gradient gradient in each region is calculated. The analysis uses a color histogram equalization technique to evaluate the smoothness of color transition by analyzing the variation amplitude of the three RGB channels in the image. At the same time, a color clustering algorithm is used to group the main colors in the image to detect the level of each color region and the color gradient relationship. By calculating the gradient value of the color change, it is determined whether the color level is coordinated with the distribution of the visual center of gravity. Finally, the visual center of gravity color gradient level data is output, which mainly includes the gradient coefficient of the color gradient in each image and the depth distribution of the color level.

[0103] Step S24: performing visual transition color conflict recognition on the visual center of gravity color gradient level data to obtain visual transition color conflict data.

[0104] In an embodiment of the present invention, after obtaining the visual center of gravity color gradient level data, it is necessary to identify the visual transition color conflict existing in the image. This step is completed by color contrast detection and color harmony calculation. First, based on the color distribution and gradient information in the previous step, the color contrast between the main color areas is analyzed, and the color differences between different areas in the image are calculated using the CIELAB color model. Then, a color harmony evaluation algorithm is used to evaluate whether the transition between adjacent color areas meets the visual aesthetic standards, especially for those areas with strong color contrast, conflicts are identified. In this process, the specific manifestation of identifying visual conflicts is that the color contrast is too large or the gradient is not coordinated, resulting in visual discomfort. Finally, the visual transition color conflict data is obtained, which mainly includes the positioning information of the color conflict area and the conflict level evaluation.

[0105] Preferably, step S21 includes the following steps:

[0106] Step S211: performing theme-background space relative coordinate mapping on the book theme-background space coordinate system to obtain theme-background relative coordinate mapping data;

[0107] Step S212: Calculating the geometric curvature of the outline of the theme element of the book or periodical theme-background space coordinate system according to the theme-background relative coordinate mapping data to obtain the geometric curvature data of the outline of the theme element;

[0108] Step S213: defining the background depth range of the book theme-background space coordinate system based on the theme-background relative coordinate mapping data to obtain background depth range data;

[0109] Step S214: performing diagonal visual tension evaluation on the image set of books and periodicals to be corrected according to the geometric curvature data of the subject element contour and the background depth of field range data, and obtaining image diagonal visual tension evaluation data;

[0110] Step S215: performing composition visual coordination perception analysis on the book and periodical image set to be corrected based on the image diagonal visual tension evaluation data to obtain composition visual coordination perception data.

[0111] In the embodiment of the present invention, the process of subject-background space relative coordinate mapping of the subject-background space coordinate system of the book and periodical first needs to clarify the spatial boundary between the subject and the background in the book and periodical image, and accurately delineate the subject area and the background area through the image segmentation algorithm. This step maps the image information of the book and periodical into a standardized relative coordinate system by calculating the geometric position of the subject part relative to the background part, and adopts a two-dimensional coordinate transformation method. The proportion and relative position of the subject area in the entire image space are determined by using the relative position relationship of the pixel points and the boundary detection technology. After the relative coordinate mapping process, the subject-background relative coordinate mapping data finally generated includes the precise position of each subject point and background point in the spatial coordinate system, as well as the distance and relative position parameters between the two. According to the subject-background relative coordinate mapping data, it is necessary to further calculate the geometric curvature of the outline of the subject part of the book and periodical. By analyzing the edge contour of the subject area, the geometric shape features of the subject contour are extracted by using a geometric curvature calculation algorithm based on edge detection. This calculation step uses the second-order differential method, and by numerically analyzing the continuity of the subject boundary, the curvature change of the edge is calculated, and the high curvature area is marked. This process ensures accurate recognition of complex boundaries such as arcs and corners, avoiding geometric shape distortion. At the same time, combined with the relative position of the background, the curvature value of the geometric contour is further corrected, and the output geometric curvature data of the subject element contour includes the curvature value of each contour point and the curvature distribution of the overall contour. Based on the subject-background relative coordinate mapping data, the depth range of the background part in the book image needs to be defined next. This operation determines the depth range of different levels in the background by analyzing the image depth information and using an algorithm based on image gradient and depth estimation. First, the gradient changes of different color areas in the image are calculated to identify the boundaries of the depth level, and then these boundaries are depth mapped to obtain the depth range of the background. By comparing the relative position of the subject area, the background depth range data finally generated includes the depth value and hierarchy of each background area, as well as the depth comparison relationship with the subject area. According to the geometric curvature data of the subject element contour and the background depth range data, the diagonal visual tension of the book image needs to be evaluated. First, based on the geometric curvature data of the subject area and the change trend of the background depth, the diagonal tension distribution of the image is analyzed by the tension field calculation algorithm. The algorithm determines whether the visual tension between the subject and the background in the image is balanced by calculating the tension difference in the diagonal direction of the image. In this process, special attention should be paid to the effect of the geometric curvature of the subject outline on the visual tension, especially the tension concentration caused by the high curvature area. The final output of the image diagonal visual tension evaluation data includes the tension value in the diagonal direction, the tension concentration area, and the overall tension distribution. Based on the image diagonal visual tension evaluation data, it is necessary to conduct a composition visual coordination perception analysis on the image set of books and periodicals to be corrected.First, based on the tension distribution data, the visual balance between the subject and the background in the image is judged, especially the impact of the diagonal tension change on the overall composition. The visual coordination algorithm based on the balance theory is used to evaluate the visual importance of the subject in the image space and its coordination with the background. By analyzing the tension concentration area in the image, it is determined whether the image composition needs to be adjusted. The final output of the composition visual coordination perception data includes the composition balance, the displacement of the visual center of gravity, and the score of the overall coordination of the image, ensuring that the book image achieves the best effect in visual perception.

[0112] Preferably, performing diagonal visual tension evaluation on the image set of books and periodicals to be corrected according to the geometric curvature data of the subject element contour and the background depth range data comprises the following steps:

[0113] The curvature variation interval of the diagonal area of ​​the image set of books and periodicals to be corrected is calculated according to the geometric curvature data of the outline of the subject element, and the curvature variation interval of the diagonal area of ​​the element is obtained;

[0114] Performing diagonal area deformation analysis on the book and periodical image set to be corrected according to the background depth range data, obtaining background diagonal area depth deformation data;

[0115] Based on the curvature change interval of the element diagonal area and the depth deformation data of the background diagonal area, the visual perception tension of the diagonal local area is simulated to obtain the diagonal local visual perception tension data;

[0116] The primary and secondary structure continuity data of the diagonal local visual perception tension is analyzed to obtain the primary and secondary structure continuity data of the diagonal global;

[0117] According to the diagonal local visual perception tension data and the diagonal global primary and secondary structure continuity data, the diagonal visual tension of the book and periodical image set to be corrected is evaluated to obtain the image diagonal visual tension evaluation data.

[0118] In the embodiment of the present invention, the curvature change of each diagonal area in the image set of books and periodicals to be corrected is accurately calculated according to the geometric curvature data of the subject element outline. In this process, the specific numerical value of the curvature of the subject outline needs to be analyzed first, and these curvature values ​​are projected to the diagonal area of ​​the image. The curvature change of each diagonal area is divided into intervals by a mathematical algorithm based on the curvature gradient, that is, the significant change points of the curvature are identified and calculated in sections. The size, direction and change amplitude of each curvature change point are calculated using the derivative method to generate the curvature change interval data of the element diagonal area. The data accurately represents the change pattern of the curvature of the image in different diagonal areas, thereby providing a basis for subsequent tension analysis. According to the background depth of field range data, it is necessary to further perform deformation analysis on the diagonal area of ​​the background part of the image set of books and periodicals to be corrected. First, by extracting the deformation features of different depth levels in the background layer by layer, the deformation degree of the background in the diagonal area is analyzed by using an algorithm based on the depth gradient and the deformation matrix. This process calculates the distribution difference of depth of field levels, identifies the key positions of background deformation, and outputs the depth of field deformation data of the background diagonal area in combination with the change trend of geometric morphology. This data can clearly show the deformation of the background in the diagonal area, especially the deformation amplitude in the area where the depth of field changes significantly, providing a three-dimensional depth reference for the overall tension assessment. Based on the curvature change interval of the element diagonal area and the depth of field deformation data of the background diagonal area generated in the first two steps, the visual perception tension of the diagonal local area is simulated. This step adopts the tension field simulation algorithm to construct a tension field model by comprehensively calculating the deformation and curvature difference of the subject and background in the diagonal direction of the image. Based on the visual tension theory, this model analyzes the relationship between the visual attraction and oppression of the subject and background in the local area, and simulates the visual perception tension in the diagonal direction of the image. The generated diagonal local visual perception tension data accurately describes the tension distribution in each local area, and marks the peak and low tension areas of visual tension, thus laying the foundation for global tension assessment. The generated diagonal local visual perception tension data is subjected to a global primary-secondary structure continuity analysis. The tension of each diagonal local area in the image is integrated and analyzed mainly through the primary-secondary structure recognition algorithm to identify the primary-secondary visual structure in the image. This step identifies the continuity relationship between the dominant visual elements and the secondary visual elements in the image by comparing the strength difference of the local tension. The gradient continuity analysis method is used to calculate the degree of connection between the primary and secondary structures. The final output of the diagonal global primary-secondary structure continuity data includes the distribution of the primary and secondary structures in the image, the continuity score, and the tension difference between each area. This data provides an important reference for the overall visual balance and coherence of the image. According to the diagonal local visual perception tension data and the diagonal global primary-secondary structure continuity data, the overall diagonal visual tension evaluation of the book and periodical image set to be corrected needs to be performed.This evaluation step uses a comprehensive visual tension analysis method to calculate the overall visual tension distribution of the image by unifying the local tension with the global primary and secondary structures. The distribution curve of the tension in the diagonal direction is analyzed by the numerical integration method, the peak and trough points of the tension in the image are identified, and the balance and uniformity of the overall tension are calculated. The final image diagonal visual tension evaluation data includes the image's visual tension score, tension distribution map, and global visual balance evaluation, which provides an important basis for the detection of mis-sticking in book binding.

[0119] Preferably, step S24 includes the following steps:

[0120] Step S241: Calculate the pixel difference between adjacent pixels of the visual center of gravity color gradient level data to obtain the color gradient adjacent pixel difference;

[0121] Step S242: performing a color wheel analysis of the contrast between adjacent colors on the color gradient level data of the visual center of gravity according to the color gradient adjacent pixel difference, to obtain the adjacent color contrast color wheel data;

[0122] Step S243: Calculate the hue angle of the adjacent color contrast color wheel data to obtain the adjacent color hue angle data;

[0123] Step S244: performing visual fatigue simulation based on adjacent color hue angle data, adjacent color contrast color wheel data, and color gradient adjacent pixel difference values ​​to obtain color gradient visual fatigue data;

[0124] Step S245: performing visual transition color conflict recognition on the visual center of gravity color gradient level data according to the color gradient visual fatigue data to obtain visual transition color conflict data.

[0125] In an embodiment of the present invention, when processing the visual center of gravity color gradient level data, it is first necessary to accurately calculate the color values ​​between adjacent pixels. To this end, a pixel difference calculation method is used. This method extracts the RGB color values ​​of each pixel and its adjacent pixels by scanning the visual center of gravity area of ​​the image pixel by pixel, and then calculates the color difference between the adjacent pixels. Specifically, based on the Euclidean distance formula, the difference between each pair of adjacent pixels in the RGB three-dimensional space is calculated, and these differences are recorded as color gradient adjacent pixel difference data. By analyzing these differences, the precise change gradient of the color gradient in the image can be quantified, providing a basis for the subsequent calculation of color contrast and hue angle. According to the color gradient adjacent pixel difference data, it is necessary to perform color contrast analysis on the visual center of gravity color gradient level, using a contrast color wheel analysis method. In this step, each pixel in the color gradient is first mapped to the standard color wheel, and the angle and contrast are calculated according to its color position and the color position of the adjacent pixel. By analyzing the contrast relationship of these colors in the color wheel, especially analyzing the relationship of color complementation, analogy or tri-color, adjacent color contrast color wheel data is generated. These data represent the color contrast between adjacent pixels and their color wheel relationship, which helps to further understand the law of color change in the image. The hue angle calculation is performed on the generated adjacent color contrast color wheel data, and this step is achieved through a mathematical angle transformation algorithm. In this process, the hue angle is first calculated according to the color contrast position of adjacent pixels in the color wheel, and the angle of each pair of adjacent colors is accurately measured using the angle calculation formula in geometry. Through this calculation method, the adjacent color hue angle data is generated. This data provides the angle relationship of color change between each adjacent pixel, which can reflect the directionality and smoothness of color gradient in the image and lay the foundation for the subsequent visual fatigue simulation. Based on the adjacent color hue angle data, adjacent color contrast color wheel data and the color gradient adjacent pixel difference, the visual fatigue simulation is performed. The process of visual fatigue simulation adopts tension field analysis and visual stimulation model, and simulates the visual fatigue of the human eye under different color gradients by comprehensively calculating the hue angle, color contrast and color change gradient. Specifically, the psychological visual theory is used to analyze the influence of color change frequency and contrast strength on visual fatigue, and then the color gradient visual fatigue data is generated. This data quantifies the fatigue effect of different gradient areas on the human eye and helps identify areas that cause visual discomfort. Based on the generated color gradient visual fatigue data, the visual center color gradient level data is further used to identify visual transition color conflicts. This step performs conflict detection on color changes in the color gradient area and uses color harmony and color contrast theory to analyze whether the color transition meets the principle of visual comfort. By comparing the color conflict data and the fatigue data of the gradient area, the visually inharmonious color transition areas in the image are identified, and finally the visual transition color conflict data is generated.The data indicates the color conflict areas in the image, which affect the visual beauty and smoothness of the overall image, and provide an intuitive color adjustment basis for correcting binding mis-attachment.

[0126] Preferably, the visual fatigue simulation based on adjacent color hue angle data, adjacent color contrast color wheel data and color gradient adjacent pixel difference includes the following steps:

[0127] Performing color gamut distortion analysis based on adjacent color hue angle data, adjacent color contrast color wheel data, and color gradient adjacent pixel difference values ​​to obtain color gamut distortion data;

[0128] Performing color detail loss analysis on the color gamut distortion data to obtain color detail loss data;

[0129] Visual stimulation cumulative effect analysis is performed based on color detail loss data to obtain visual stimulation cumulative effect data;

[0130] Visual fatigue degree simulation is performed based on the color detail loss data and the visual stimulation cumulative effect data to obtain the color gradient visual fatigue degree data.

[0131] In the embodiment of the present invention, in the preliminary stage of visual fatigue simulation, it is necessary to perform color gamut distortion analysis based on adjacent color hue angle data, adjacent color contrast color wheel data, and color gradient adjacent pixel difference. First, each color point is mapped from the RGB color space to the hue, saturation, brightness (HSL) color space through mathematical color space mapping technology, and the color distortion is calculated. In specific implementation, the nonlinear geometric transformation theory is used to perform distortion analysis on the change area of ​​the color space to identify deformation phenomena such as stretching and compression of the color gamut boundary, and generate color gamut distortion data. The color gamut distortion data reflects the hue deformation of the colors in different areas of the image during the gradient process, and helps to quantify the visual impact of the colors in the gradient area. After the color gamut distortion data is generated, it is necessary to perform color detail loss analysis on these data. This step decomposes the frequency domain of the color through mathematical tools such as Fourier transform or wavelet transform, so as to analyze the degree of loss of color details in the image. First, the color information of the high-frequency area in the image is extracted to identify the area where the details are lost during the gradient process, especially the high-frequency color loss caused by the color gamut distortion. Next, combined with the changes in color boundary sharpness and transition smoothness, color detail loss data is calculated and generated. This data shows which parts of the gradient area have lost color details, providing a basis for the subsequent assessment of visual fatigue. Based on the color detail loss data, a visual stimulation cumulative effect analysis is performed. This step simulates the cumulative visual fatigue of the human eye when observing the same gradient color area for a long time by combining the psychological visual theory with the comprehensive effects of color contrast, frequency change and detail loss. In specific implementation, the cumulative stimulation model of visual neurology is adopted to quantify the cumulative effect of stimulation of each color area on the human eye by analyzing factors such as the frequency of color contrast changes and the duration of visual stimulation intensity, and generate visual stimulation cumulative effect data. This data is used to quantify the visual fatigue in the color gradient area, thereby identifying potential sources of visual fatigue. Finally, based on the color detail loss data and the visual stimulation cumulative effect data, an overall visual fatigue simulation is performed. This step simulates the degree of visual fatigue of the human eye to the gradient color area through a comprehensive analysis of the cumulative effect analysis and color details. The perception model is used to quantify the human eye's response to high-contrast colors, detail loss and long-term stimulation to form an estimate of fatigue for different color areas. By integrating the visual fatigue of each area, color gradient visual fatigue data is generated. This data provides the specific area that causes visual fatigue in the image and provides a quantitative basis for correcting the visual discomfort in the image, thereby improving the accuracy of image correction in book binding inspection.

[0132] Preferably, step S3 comprises the following steps:

[0133] Step S31: performing image edge continuity detection on the book image set to be corrected according to the visual transition color conflict data to obtain image edge continuity data;

[0134] Step S32: performing binding error pattern recognition based on the image edge continuity data and the visual transition color conflict data to obtain image binding error pattern data;

[0135] Step S33: adjusting the coordinate orientation of the book subject-background space coordinate system according to the image binding mis-pasting mode data to obtain image binding mis-pasting orientation adjustment data.

[0136] In the embodiment of the present invention, it is first necessary to perform image edge continuity detection on the image set of books and periodicals to be corrected according to the visual transition color conflict data. In the specific implementation, an edge detection algorithm, such as the Canny edge detection algorithm, is used to extract the edge in the image. Then, the continuity of the image edge is calculated by analyzing the overlap between the extracted edge and the color conflict area. The core of this detection is that the definition of edge continuity refers to whether the lines or shapes that should be visually continuous in the image are broken or distorted due to color conflict. In order to enhance the accuracy of the continuity detection, the mathematical gradient calculation method can also be combined to analyze the change direction of adjacent pixels in the image to ensure the smooth transition of the edge. The image edge continuity data finally generated is used to measure the integrity and continuity of the edge area in the image of the book and periodical to be corrected, reflecting the degree of damage to the visual boundary caused by the color conflict. Binding error pattern recognition is performed based on the image edge continuity data and the visual transition color conflict data. The specific implementation process is to identify the edge dislocation or irregularity caused by binding errors by performing pattern matching analysis on the image edge continuity data. This process relies on the geometric morphology and position information of the edge. By comparing the abnormal areas in the edge continuity data and combining the visual incoherence caused by color conflict, the geometric pattern recognition algorithm is applied to identify the specific pattern of binding mis-pasting. Through the pattern recognition algorithm, such as contour matching or topological analysis based on shape features, the features of different mis-pasting types are classified and labeled to generate image binding mis-pasting pattern data. This data reflects the specific type of binding mis-pasting problem in each image to be corrected, such as page number misalignment, binding line distortion or overall image offset. According to the generated image binding mis-pasting pattern data, the coordinate orientation of the book theme-background space coordinate system is adjusted. In specific implementation, the offset information in the mis-pasting pattern data is first used to calculate the mis-pasting orientation offset between the subject and background in the book image. The coordinates in the image are adjusted through coordinate transformation technology, such as two-dimensional or three-dimensional affine transformation. This adjustment not only includes translation, rotation, and scaling of the image, but also requires more complex geometric transformations based on the specific characteristics of the mis-pasting pattern to ensure that the relative position of each element in the image in the subject-background coordinate system is corrected. In the implementation process, matrix operations are used to reposition the positions of each pixel of the image to generate image binding mis-pasting orientation adjustment data, which represents the relative position of the subject and background after the image is corrected, ensuring that the binding mis-pasting problem is effectively corrected.

[0137] Preferably, step S4 comprises the following steps:

[0138] Step S41: normalizing the binding mis-pasting orientation adjustment data to obtain orientation adjustment normalized data;

[0139] Step S42: Designing an automatic execution program for the orientation adjustment normalization data to obtain an automatic adjustment program for mis-adhering of the binding;

[0140] Step S43: embedding the automatic binding mis-adherence adjustment program into the image detection system to execute the binding mis-adherence image detection method.

[0141] In an embodiment of the present invention, the binding mis-pasting orientation adjustment data is normalized. In the specific implementation process, it is first necessary to clearly adjust the orientation offset of each book image in the data, including parameters of geometric transformations such as image translation, rotation, and scaling. In order to ensure that these data can be used for subsequent automated program execution, they need to be normalized. This step uses a mathematical normalization method to uniformly map the offsets of different amplitudes to the same scale range. For example, for the rotation angle, it can be standardized to an interval of [-1,1], and the translation distance can be scaled according to the image resolution to ensure that the orientation adjustment of all book images is processed under the same reference frame. The main purpose of this normalization operation is to reduce the impact of differences in parameters such as book size and image resolution on the orientation adjustment accuracy. The output after normalization is orientation adjustment normalized data, and the unified and standardized values ​​are helpful for the implementation of automated programs in subsequent steps. It is necessary to design an automated execution program for the orientation adjustment normalized data. In the specific implementation, the various adjustment parameters in the normalized data are first analyzed, and the corresponding geometric transformation algorithm is designed based on these parameters. Geometric transformation tools based on matrix operations, such as affine transformation or perspective transformation, can be used to reposition the coordinates of each pixel point of the image. In order to achieve automation, these geometric transformation steps need to be arranged into a continuous operation process to ensure that the orientation adjustment of different book and periodical images can be performed in batches. During the program design process, the execution order of the steps should be clarified, including how to obtain the orientation adjustment normalization data from the input image, how to apply specific transformation parameters, and ensure that the adjusted image is saved to the specified location. Through this design, the program can automatically perform the necessary geometric transformation after receiving the normalized data to form an automated adjustment program for binding errors. The automated adjustment program for binding errors is embedded in the image detection system to perform the image detection method for binding errors. The specific operation is to integrate the aforementioned designed automated adjustment program with the existing image detection system. First, it is necessary to ensure that the automated program is compatible with the image processing module in the detection system. The adjustment program can be embedded as a subroutine in the process of the detection system by calling the API interface in the image processing system or writing a script. Secondly, the system needs to pre-process the input image set to obtain the initial data required for mis-binding detection, and call the automatic adjustment program after detecting mis-binding to perform orientation correction operations. The embedded process must ensure that the mis-binding adjustment program can automatically run in the detection link of the system, process all images of books and periodicals detected to have mis-binding problems, and finally output the corrected image results.

[0142] Preferably, the present invention further provides an image detection system for correcting mis-sticking in book binding, which is used to execute the image detection method for correcting mis-sticking in book binding as described above. The image detection system for correcting mis-sticking in book binding comprises:

[0143] The coordinate system construction module is used to scan the image of the book to be corrected to generate a set of images of the book to be corrected; calculate the subject-background ratio of the book image to be corrected to obtain subject-background ratio data; construct a spatial coordinate system for the book image to be corrected according to the subject-background ratio data to obtain a book subject-background spatial coordinate system;

[0144] The visual color conflict recognition module is used to perform composition visual coordination perception analysis on the image set of books and periodicals to be corrected according to the book and periodical theme-background space coordinate system to obtain composition visual coordination perception data; perform visual transition color conflict recognition according to the composition visual coordination perception data to obtain visual transition color conflict data;

[0145] The mis-pasting orientation adjustment module is used to identify the binding mis-pasting pattern based on the visual transition color conflict data to obtain the image binding mis-pasting pattern data; adjust the coordinate orientation of the book theme-background space coordinate system according to the image binding mis-pasting pattern data to obtain the image binding mis-pasting orientation adjustment data;

[0146] The automatic program design module is used to automatically execute program design for the binding mis-sticking orientation adjustment data to obtain the binding mis-sticking automatic adjustment program; the binding mis-sticking automatic adjustment program is embedded in the image detection system to execute the binding mis-sticking image detection method.

[0147] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0148] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. An image detection method for correcting mis-sticking of books and periodicals, characterized in that: The following steps are involved: Step S1: Scan the image of the book or periodical to be corrected to generate a set of book or periodical images to be corrected; calculate the subject-background ratio of the book or periodical image set to be corrected to obtain subject-background ratio data; construct a spatial coordinate system for the book or periodical image set to be corrected according to the subject-background ratio data to obtain a book or periodical subject-background spatial coordinate system; Step S2: performing composition visual coordination perception analysis on the book image set to be corrected according to the book theme-background space coordinate system to obtain composition visual coordination perception data; performing visual transition color conflict recognition according to the composition visual coordination perception data to obtain visual transition color conflict data; Step S3: performing binding error pattern recognition based on the visual transition color conflict data to obtain image binding error pattern data; adjusting the coordinate orientation of the book theme-background space coordinate system according to the image binding error pattern data to obtain image binding error orientation adjustment data; Step S4: Design an automated execution program for the binding mis-adherence orientation adjustment data to obtain an automated binding mis-adherence adjustment program; embed the automated binding mis-adherence adjustment program into the image detection system to execute the binding mis-adherence image detection method.

2. The image detection method for correcting mis-sticking of books and periodicals according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Scan the books and periodicals to be corrected to generate an image set of the books and periodicals to be corrected; Step S12: performing subject-background separation on the book and periodical image set to be corrected to obtain subject-background separation data; Step S13: calculating the subject-background ratio of the book image set to be corrected according to the subject-background separation data to obtain subject-background ratio data; Step S14: constructing a spatial coordinate system for the book image set to be corrected according to the subject-background ratio data to obtain a book subject-background spatial coordinate system.

3. The image detection method for correcting mis-sticking of books and periodicals according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: performing composition visual coordination perception analysis on the book image set to be corrected according to the book theme-background space coordinate system to obtain composition visual coordination perception data; Step S22: performing visual center of gravity offset recognition on the composition visual coordination perception data to obtain composition visual center of gravity offset data; Step S23: performing color gradient level perception analysis on the book image set to be corrected according to the composition visual center of gravity offset data to obtain visual center of gravity color gradient level data; Step S24: performing visual transition color conflict recognition on the visual center of gravity color gradient level data to obtain visual transition color conflict data.

4. The image detection method for correcting mis-sticking of books and periodicals according to claim 3, characterized in that: Step S21 includes the following steps: Step S211: performing theme-background space relative coordinate mapping on the book theme-background space coordinate system to obtain theme-background relative coordinate mapping data; Step S212: Calculating the geometric curvature of the outline of the theme element in the book theme-background space coordinate system according to the theme-background relative coordinate mapping data to obtain the geometric curvature data of the outline of the theme element; Step S213: defining the background depth range of the book theme-background space coordinate system based on the theme-background relative coordinate mapping data to obtain background depth range data; Step S214: performing diagonal visual tension evaluation on the book and periodical image set to be corrected according to the geometric curvature data of the subject element contour and the background depth of field range data, to obtain image diagonal visual tension evaluation data; Step S215: performing composition visual coordination perception analysis on the book and periodical image set to be corrected based on the image diagonal visual tension evaluation data to obtain composition visual coordination perception data.

5. The image detection method for correcting mis-sticking of books and periodicals according to claim 3, characterized in that: The diagonal visual tension evaluation of the image set of books and periodicals to be corrected based on the geometric curvature data of the subject element contour and the background depth range data includes the following steps: The curvature variation interval of the diagonal area of ​​the image set of books and periodicals to be corrected is calculated according to the geometric curvature data of the outline of the subject element, and the curvature variation interval of the diagonal area of ​​the element is obtained; Performing diagonal area deformation analysis on the book and periodical image set to be corrected according to the background depth range data, obtaining background diagonal area depth deformation data; Based on the curvature change interval of the element diagonal area and the depth deformation data of the background diagonal area, the visual perception tension of the diagonal local area is simulated to obtain the diagonal local visual perception tension data; The diagonal local visual perception tension data is analyzed for primary and secondary structure continuity to obtain the diagonal global primary and secondary structure continuity data; According to the diagonal local visual perception tension data and the diagonal global primary and secondary structure continuity data, the diagonal visual tension of the book and periodical image set to be corrected is evaluated to obtain the image diagonal visual tension evaluation data.

6. The image detection method for correcting mis-sticking of books and periodicals according to claim 3, characterized in that: Step S24 includes the following steps: Step S241: Calculate the pixel difference between adjacent pixels of the visual center of gravity color gradient level data to obtain the color gradient adjacent pixel difference; Step S242: performing a color wheel analysis of the contrast between adjacent colors on the color gradient level data of the visual center of gravity according to the color gradient adjacent pixel difference, to obtain the adjacent color contrast color wheel data; Step S243: Calculate the hue angle of the adjacent color contrast color wheel data to obtain the adjacent color hue angle data; Step S244: performing visual fatigue simulation based on adjacent color hue angle data, adjacent color contrast color wheel data, and color gradient adjacent pixel difference values ​​to obtain color gradient visual fatigue data; Step S245: performing visual transition color conflict identification on the visual center of gravity color gradient level data according to the color gradient visual fatigue data to obtain visual transition color conflict data.

7. The image detection method for correcting mis-sticking of books and periodicals according to claim 6, characterized in that: The simulation of visual fatigue based on adjacent color hue angle data, adjacent color contrast color wheel data, and color gradient adjacent pixel difference includes the following steps: Performing color gamut distortion analysis based on adjacent color hue angle data, adjacent color contrast color wheel data, and color gradient adjacent pixel difference values ​​to obtain color gamut distortion data; Performing color detail loss analysis on the color gamut distortion data to obtain color detail loss data; Visual stimulation cumulative effect analysis is performed based on color detail loss data to obtain visual stimulation cumulative effect data; Visual fatigue simulation is performed based on the color detail loss data and the visual stimulation cumulative effect data to obtain the color gradient visual fatigue data.

8. The image detection method for correcting mis-sticking of books and periodicals according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: performing image edge continuity detection on the book image set to be corrected according to the visual transition color conflict data to obtain image edge continuity data; Step S32: performing binding error pattern recognition based on the image edge continuity data and the visual transition color conflict data to obtain image binding error pattern data; Step S33: adjusting the coordinate orientation of the book theme-background space coordinate system according to the image binding mis-pasting mode data to obtain the image binding mis-pasting orientation adjustment data.

9. The image detection method for correcting mis-sticking of book binding according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: normalizing the binding mis-pasting orientation adjustment data to obtain orientation adjustment normalized data; Step S42: Designing an automatic execution program for the orientation adjustment normalization data to obtain an automatic adjustment program for mis-adhering of the binding; Step S43: embedding the automatic binding mis-adherence adjustment program into the image detection system to execute the binding mis-adherence image detection method.

10. An image detection system for correcting mis-binding of books and periodicals, characterized in that: Used to execute the image detection method for correcting mis-sticking of book binding as claimed in claim 1, the image detection system for correcting mis-sticking of book binding comprises: The coordinate system construction module is used to scan the image of the book to be corrected to generate a set of images of the book to be corrected; calculate the subject-background ratio of the book image to be corrected to obtain subject-background ratio data; construct a spatial coordinate system for the book image to be corrected according to the subject-background ratio data to obtain a book subject-background spatial coordinate system; The visual color conflict recognition module is used to perform composition visual coordination perception analysis on the image set of books and periodicals to be corrected according to the book and periodical theme-background space coordinate system to obtain composition visual coordination perception data; perform visual transition color conflict recognition according to the composition visual coordination perception data to obtain visual transition color conflict data; The mis-pasting orientation adjustment module is used to identify the binding mis-pasting pattern based on the visual transition color conflict data to obtain the image binding mis-pasting pattern data; adjust the coordinate orientation of the book theme-background space coordinate system according to the image binding mis-pasting pattern data to obtain the image binding mis-pasting orientation adjustment data; The automatic program design module is used to automatically execute program design for the binding mis-sticking orientation adjustment data to obtain the binding mis-sticking automatic adjustment program; the binding mis-sticking automatic adjustment program is embedded in the image detection system to execute the binding mis-sticking image detection method.

Images