An AI-based graphic and text layout method

Abstract

The present invention relates to the technical field of graphic and text typesetting, and discloses a graphic and text typesetting method based on artificial intelligence. This method highly integrates artificial intelligence technology, effectively improving the typesetting quality and efficiency. Especially in step S2, through the weight adjustment based on context relevance and emotional relevance, the method can more accurately express the text emotion, optimize the connection between paragraphs, and enable children to be more engaged when reading. In addition, through the comprehensive analysis of the picture emotion value, hue value, and intervention value, the method is more accurate in terms of the relevance between graphics and texts, adding much luster to the typesetting, making the pictures consistent with the text paragraphs, enhancing children's empathy when reading, increasing their eagerness for the subsequent content of the reading material, and to a certain extent, balancing the distribution uniformity of text and pictures, making children's reading smoother.

Description

Technical Field

[0001] The present invention relates to the technical field of graphic layout, and specifically provides a graphic layout method based on artificial intelligence. Background Art

[0002] Graphic layout refers to the orderly organization of elements such as text, pictures, and charts on a page to create a beautiful, easy-to-read, and easy-to-understand visual layout. Graphic layout is very important in fields such as printed materials, web design, poster production, and magazine layout. It can help convey information, attract readers' attention, and convey certain emotions and styles.

[0003] In children's books, illustrations are an indispensable part. To a certain extent, illustrations can more vividly summarize the meaning of the text for children, increase children's expectations for the book, make them more willing to browse the subsequent content, and can also cultivate children's sense of art, enabling them to have a certain appreciation ability and interest in art. Therefore, in most current children's publications, pictures are basically inserted and occupy a relatively high proportion.

[0004] However, if there are too many patterns inserted in the text, it will cause reading difficulties for children. On the other hand, if there are too few illustrations, children will lose their interest in reading. Moreover, when the emotional expressions of pictures and text are quite different, it is easy to mislead children's emotional expressions. Summary of the Invention

[0005] The present invention provides a graphic layout method based on artificial intelligence, which has the beneficial effects of optimizing the distribution ratio of graphics and text, increasing children's reading interest, and balancing the emotional consistency of graphics and text, and solves the problems mentioned in the above background art, that is, too many patterns will cause reading difficulties for children, too few illustrations will cause children to lose their interest in reading, and when the emotional expressions of pictures and text are quite different, it is easy to mislead children's emotional expressions.

[0006] The present invention provides the following technical solution: A graphic layout method based on artificial intelligence includes the following processes:

[0007] S1. Use a collection device to scan the story text and pictures to obtain text reference data and picture reference data respectively;

[0008] S2. Use a calculation module to combine and calculate the collected text reference data and picture reference data to obtain a layout reference coefficient PBX. The specific calculation method is as follows:

[0009]

[0010] Where GLX is the adaptation value between paragraphs, obtained from the copywriting reference data, representing the relevance between several natural paragraphs in the copywriting. a is its weight value in the calculation, representing the proportion it occupies in the calculation of the typesetting reference coefficient PBX;

[0011] Where TDX is the correlation value between the picture and the paragraph, obtained by combining the copywriting reference data and the picture reference data, representing the relevance between the picture and the natural paragraph in the copywriting. b is its weight value in the calculation, representing the proportion it occupies in the calculation of the typesetting reference coefficient PBX;

[0012] Where TPX is the intervention value of the picture, obtained by combining the copywriting reference data and the picture reference data, representing the degree of fit between the picture and the natural paragraph in the copywriting. c is its weight value in the calculation, representing the proportion it occupies in the calculation of the typesetting reference coefficient PBX;

[0013] Where the values of a, b, and c are selected and set by the customer themselves, and a + b + c = 1, a ≠ b ≠ c;

[0014] Where A1, A2, and A3 are all correction coefficients, and the values of the three are selected and set by the customer themselves;

[0015] S3. Use the typesetting reference coefficient PBX obtained by calculation through the execution module to perform a secondary analysis on the text, perform typesetting, and send the typeset result to the detection port. The staff will detect the typeset file through the detection to provide modifications. And the detection port will record the number of operations for each single modification, and compare the number of operations according to the evaluation criteria to provide a reference for method update;

[0016] S4. After the modification is completed and confirmed to be correct, the execution module sends the modified typesetting method to the terminal module. After receiving the typesetting data, the terminal module performs actual typesetting on the input document and picture and sends it to the detection port for confirmation again;

[0017] S5. After the staff makes the final confirmation, the terminal module sends the finalized version to the relevant website or printing manufacturer for pushing and physical printing operations.

[0018] As an alternative solution of the graphic typesetting method based on artificial intelligence according to the present invention, wherein: the adaptation value GLX of the story segment in the S2 step is obtained by analyzing the relevance between natural paragraphs of the input copywriting, including the context connection LXX between natural paragraphs and the emotional connection QGX between natural paragraphs. The specific calculation formula is as follows:

[0019] GLX = d * LXX + e * QGX

[0020] Wherein, d and e are respectively the weight values of the context connectivity LXX and the emotional connectivity QGX. The values of d and e are selected and set by the customer themselves, and d + e = 1, d ≠ e.

[0021] As an alternative solution of the AI-based graphic and text layout method described in the present invention, wherein: the context connectivity LXX is obtained by analyzing the copywriting by artificial intelligence, and its proportion is determined by comparing its value with the evaluation criteria. The specific analysis method is as follows:

[0022] When the context connectivity LXX is between 0 and 4, it represents that the context connectivity between natural paragraphs is of low level, and the proportion is adjusted synchronously;

[0023] When the context connectivity LXX is between 5 and 7, it represents that the context connectivity between natural paragraphs is of medium level, and the proportion is adjusted synchronously;

[0024] When the context connectivity LXX is between 8 and 10, it represents that the context connectivity between natural paragraphs is of high level, and the proportion is adjusted synchronously.

[0025] As an alternative solution of the AI-based graphic and text layout method described in the present invention, wherein: the emotional connectivity QGX is obtained by analyzing the copywriting by artificial intelligence, and its proportion is adjusted by comparing its value with the evaluation criteria. The specific analysis and adjustment method is as follows:

[0026] When the emotional connectivity QGX is between 0 and 3, it represents that the emotion of the current paragraph is low, and the proportion is adjusted synchronously;

[0027] When the emotional connectivity QGX is between 3 and 5, it represents that the emotion of the current paragraph is stable, and the proportion is adjusted synchronously;

[0028] When the emotional connectivity QGX is between 6 and 10, it represents that the emotion of the current paragraph is high, and the proportion is adjusted synchronously.

[0029] As an alternative solution of the AI-based graphic and text layout method described in the present invention, wherein: the association value TDX between the picture and the paragraph in the S2 step is obtained by analyzing the data between the input copywriting and the picture by artificial intelligence, including the emotional value TQG of the picture and the picture tone value TSD. The specific calculation method is as follows:

[0030] TDX = TQG * TSD.

[0031] As an alternative solution of the AI-based graphic and text layout method described in the present invention, wherein: the emotional value TQG of the picture is consistent with the emotional connection QGX, and the two have the same attributes. The picture tone value TSD is obtained by analyzing the emotional value TQG of the picture.

[0032] As an alternative solution of the AI-based graphic and text layout method described in the present invention, wherein: the picture tone value TSD includes warm tones, cold tones, and neutral tones. The specific selection methods are as follows:

[0033] When the emotional value TQG of the picture is within the downturn range, the picture tone value TSD selects a cold tone, denoted as 2;

[0034] When the emotional value TQG of the picture is within the stable range, the picture tone value TSD selects a neutral tone, denoted as 1;

[0035] When the emotional value TQG of the picture is within the upsurge range, the picture tone value TSD selects a warm tone, denoted as 3.

[0036] As an alternative solution of the AI-based graphic and text layout method described in the present invention, wherein: the intervention value TPX of the picture is obtained by analyzing the copy reference data and the picture reference data through artificial intelligence, and includes the picture size TPC, the length DCD of a single natural paragraph of the copy, and the correlation value TDX between the picture and the paragraph. The specific calculation method is as follows:

[0037] TPX = (TPC * TDX) + (DCD * TDX) + TDX.

[0038] As an alternative solution of the AI-based graphic and text layout method described in the present invention, wherein: the evaluation criteria in the S3 step include an update threshold and an update threshold. The specific evaluation method is as follows:

[0039] When the number of operations ≤ the update threshold, it means that the practical level of this method is high level and no update is required;

[0040] When the update threshold < the number of operations < the update threshold, it means that the practical level of this method is medium level and appropriate update and adjustment are required;

[0041] When the number of operations ≥ the update threshold, it means that the practical level of this method is low level and significant update and adjustment are required.

[0042] The present invention has the following beneficial effects:

[0043] 1. The AI-based graphic and text layout method realizes intelligent optimization of graphic and text layout through the S1-S5 processes. From the collection of copywriting and pictures to the combined calculation of layout reference coefficients by the calculation module, and then to the secondary analysis and actual layout by the execution module, this method highly integrates AI technology, effectively improving the layout quality and efficiency. Especially in step S2, through the weight adjustment based on context connectivity and emotional connectivity, the method can more accurately express the text emotion and optimize the connection between paragraphs, enabling children to be more engaged in reading. In addition, through the comprehensive analysis of the picture emotion value, color tone value, and intervention value, the method is more accurate in the graphic and text relevance, enhancing the layout, making the pictures consistent with the text paragraphs, improving children's empathy during reading, increasing their eagerness for the subsequent content of the reading material, and to a certain extent, balancing the distribution uniformity of text and pictures, making children's reading smoother.

[0044] 2. The AI-based graphic and text layout method introduces an evaluation criterion for update threshold and updated threshold in step S3, effectively enhancing the practicality and continuous optimization ability of the method. Through the relationship between the number of operations and the update threshold, the method can intelligently judge whether update adjustment is needed, keeping the layout method at a high, medium, or low practical level. This adaptive update strategy helps the method always be in the best state to adapt to changing layout requirements. For layout staff, they can conduct detection and modification after layout, improving the work efficiency and accuracy. The loop formed by steps S3 and S4 provides a reliable guarantee for the refinement and self-adjustment of layout, enabling the method to be continuously improved in practice and having greater practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a schematic diagram of the method steps of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0047] Embodiment 1

[0048] Please refer to Figure 1 , an AI-based graphic and text layout method, characterized by including the following processes:

[0049] S1. Use a collection device to scan the story copywriting and pictures to obtain copywriting reference data and picture reference data respectively;

[0050] S2. The calculation module combines the collected copy reference data and picture reference data for calculation to obtain the layout reference coefficient PBX. The specific calculation method is as follows:

[0051]

[0052] In the formula, GLX is the adaptation value between paragraphs, obtained from the copy reference data, representing the relevance of the context between several natural paragraphs in the copy. a is its weight value in the calculation, representing the proportion it occupies in the calculation of the layout reference coefficient PBX;

[0053] In the formula, TDX is the association value between the picture and the paragraph, obtained by combining the copy reference data and the picture reference data, representing the relevance between the picture and the natural paragraph in the copy. b is its weight value in the calculation, representing the proportion it occupies in the calculation of the layout reference coefficient PBX;

[0054] In the formula, TPX is the intervention value of the picture, obtained by combining the copy reference data and the picture reference data, representing the fitting degree between the picture and the natural paragraph in the copy. c is its weight value in the calculation, representing the proportion it occupies in the calculation of the layout reference coefficient PBX;

[0055] In the formula, the values of a, b, and c are selected and set by the customer themselves, and a + b + c = 1, a ≠ b ≠ c;

[0056] In the formula, A1, A2, and A3 are all correction coefficients, and their values are selected and set by the customer themselves;

[0057] S3. The execution module performs a secondary analysis on the text using the obtained layout reference coefficient PBX through calculation, performs typesetting, and sends the typeset result to the detection port. The staff conducts inspections on the typeset file through detection to provide modifications, and the detection port records the number of operations for each single modification and compares the number of operations according to the evaluation criteria to provide a reference for method updates;

[0058] S4. After the modification is completed and confirmed to be correct, the execution module sends the modified typesetting method to the terminal module. After receiving the typesetting data, the terminal module performs actual typesetting on the input document and picture and sends it to the detection port for confirmation again;

[0059] S5. After the staff makes the final confirmation, the terminal module sends the final version to the relevant website or printing manufacturer for pushing and physical printing operations.

[0060] In this embodiment: In children's books, illustrations are an indispensable part. To a certain extent, illustrations can more vividly summarize the meaning of the text for children, increase children's expectations for the book, make them more willing to browse the subsequent content, and can also cultivate a sense of art in children, enabling them to have a certain appreciation ability and interest in art. Therefore, in most current children's published books, pictures are basically inserted and the occupancy rate is relatively high.

[0061] However, if too many patterns are inserted in the text, it will cause reading difficulties for children, while too few illustrations will cause children to lose interest in reading.

[0062] This method calculates the typesetting reference coefficient by integrating the text reference data and the picture reference data. This comprehensive calculation takes into account factors such as the adaptation value between paragraphs, the correlation value between pictures and paragraphs, and the intervention value of pictures, etc., providing a more scientific basis for typesetting, making the typesetting more accurate and reasonable. The typesetting reference coefficient is used for secondary analysis and typesetting of the text, making the typesetting process more intelligent.

[0063] This method can not only improve the typesetting quality, but also maintain the coordination and consistency between the text and the pictures, making the output effect more professional. The typesetting result will be sent to the detection port for staff to detect and modify.

[0064] This interactive process can ensure the accuracy of typesetting and meet the needs of customers. At the same time, by recording the number of operations and comparing and evaluating, it provides valuable feedback for the improvement of the system. After being modified and confirmed by the staff, the typesetting method will be sent to the terminal module for actual typesetting, which ensures the accuracy and consistency of the final draft, providing a reliable basis for the final output of website push or printing operations.

[0065] The method has greatly improved the efficiency of typesetting in aspects such as automated processing, comprehensive calculation, and automatic typesetting. At the same time, through intelligent analysis and comprehensive consideration, it ensures the consistency between the text and the pictures, making the typesetting result more professional and unified. The method provides users with an efficient, accurate, and consistent typesetting experience through an automated and intelligent process. It can not only greatly save time and human resources, but also ensure the typesetting quality and the professionalism of the final output, thus effectively avoiding the problem that the illustrations in children's books interfere with the reading quality and improving the reading experience of children.

[0066] Embodiment 2

[0067] Please refer to Figure 1, the adaptation value GLX of the story segment in step S2 is obtained by analyzing the relevance between natural paragraphs of the input copywriting, including the contextual connectivity LXX and the emotional connectivity QGX between natural paragraphs. The specific calculation formula is as follows:

[0068] GLX = d * LXX + e * QGX

[0069] In the formula, d and e are the weight values of the contextual connectivity LXX and the emotional connectivity QGX respectively. The values of d and e are selected and set by the customer themselves, and d + e = 1, d ≠ e.

[0070] In this embodiment: Focusing on a more in-depth analysis of the relevance between natural paragraphs of the text, not only considering the contextual connectivity, but also incorporating the emotional connectivity. This method introduces more intelligent and user-friendly elements in the typesetting field, thus improving the typesetting effect and user experience.

[0071] Introducing the weight values of the contextual connectivity and the emotional connectivity makes the calculation of the typesetting reference coefficient more comprehensive and accurate, and can better capture the relevance between different natural paragraphs in the copywriting, so as to more accurately handle the arrangement and presentation of paragraphs in the typesetting process, and improve the overall quality of typesetting.

[0072] By incorporating the weight value of the emotional connectivity, the typesetting method can more sensitively capture the emotional changes between text paragraphs, which helps to convey more accurate emotions and tones in typesetting, so that the typesetting result is more in line with the original intention of the author, and enhances the emotional expressiveness of the text.

[0073] The weight values are set by the customer themselves, which means that different customers can customize according to their own needs and preferences. This flexibility makes the typesetting result more able to meet the requirements of different users, thus increasing the scope of application and flexibility of the typesetting method.

[0074] Taking into account the contextual connectivity and the emotional connectivity comprehensively, the typesetting method can more accurately express the information and emotions of the text, which helps to improve the reading experience of readers, makes it easier for them to understand and resonate, and thus improves the satisfaction and participation of users.

[0075] The comprehensive consideration of the contextual connectivity and the emotional connectivity makes the typesetting method more tend to professional typesetting. Whether it is publications, promotional materials or other text forms, this method can better present the content, making the typesetting result more professional and authoritative.

[0076] By considering context connectivity and emotional connectivity, it brings a more accurate, personalized, and emotionally rich effect to typesetting. This method not only improves the typesetting quality but also enhances the user experience, and is expected to promote the development of the typesetting field towards a more intelligent and user-friendly direction.

[0077] Example 3

[0078] Please refer to Figure 1 , the context connectivity LXX is obtained by analyzing the text through artificial intelligence, and its proportion is determined by comparing its value with the evaluation criteria. The specific analysis method is as follows:

[0079] When the context connectivity LXX is between 0 and 4, it represents a low-level connectivity between natural paragraphs, and the proportion is adjusted synchronously;

[0080] When the context connectivity LXX is between 5 and 7, it represents a medium-level connectivity between natural paragraphs, and the proportion is adjusted synchronously;

[0081] When the context connectivity LXX is between 8 and 10, it represents a high-level connectivity between natural paragraphs, and the proportion is adjusted synchronously.

[0082] In this embodiment: This method uses artificial intelligence to analyze the text and divides different levels of connectivity according to the context connectivity between natural paragraphs, which helps to capture the relationship between text paragraphs more accurately, so as to present a more reasonable and coherent arrangement in the typesetting process.

[0083] According to the levels of different context connectivities, the method will correspondingly adjust the proportion of each level. This makes the typesetting result more in line with the characteristics of the text and the author's intention, achieving a personalized typesetting effect and improving the customization of typesetting.

[0084] Through the adjustment of the connectivity levels and proportions, the method can finely control the typesetting between each natural paragraph. This helps to better handle the transitions, associations, and coherences between different paragraphs, making the typesetting result more professional and fluent.

[0085] The method divides the context connectivity into three levels: low, medium, and high, and sets different proportions for each level. This method can maintain the overall consistency of typesetting while taking into account the diversity of paragraph relationships, achieving the best typesetting balance.

[0086] By adjusting the proportion according to the context connectivity, the typesetting method can better present the logical relationship and emotional changes of the text, which helps to enhance the reading experience of readers and makes it easier for them to understand and immerse in the text.

[0087] Through the adjustment of the context connection level and proportion, a more accurate, personalized, meticulous and professional typesetting effect is achieved. This method can meet the typesetting needs of different texts, improve the user experience, and promote the intelligent innovation and development in the field of typesetting.

[0088] Example 4

[0089] Please refer to Figure 1 , the emotional connection QGX is obtained by analyzing the copywriting through artificial intelligence, and its proportion is adjusted by comparing its value with the evaluation criteria. The specific analysis and adjustment methods are as follows:

[0090] When the emotional connection QGX is between 0 and 3, it represents the low emotion of the current paragraph, and the proportion occupied is adjusted synchronously;

[0091] When the emotional connection QGX is between 3 and 5, it represents the stable emotion of the current paragraph, and the proportion occupied is adjusted synchronously;

[0092] When the emotional connection QGX is between 6 and 10, it represents the high emotion of the current paragraph, and the proportion occupied is adjusted synchronously.

[0093] In this embodiment: The method analyzes the emotion of the copywriting through artificial intelligence and divides different levels of emotional states according to the size of the emotional connection. This helps to better capture the emotional changes between text paragraphs and makes the typesetting result more accurately convey emotions.

[0094] According to the emotional connection level, the method adjusts the proportion of each level, so that the typesetting result can better present the emotional changes between text paragraphs. This can maintain the consistency of emotions in the overall typesetting and effectively express the ups and downs of emotions at the same time.

[0095] By adjusting the proportion according to the size of the emotional connection, the typesetting method can more accurately convey the emotional changes of the text, making it easier for readers to resonate. This helps readers to understand and experience the text more deeply and enhances the emotional experience of reading.

[0096] Through the analysis and proportion adjustment of the emotional connection, the method can better control the emotional changes between paragraphs, making the transmission of emotions smoother and more coherent. This helps to effectively guide the emotional experience of readers in typesetting and achieve a better typesetting effect.

[0097] The method adjusts the emotional proportion to make the typesetting result more in line with the emotional color of the text, thereby improving the readability of the text. Readers can grasp the emotional orientation of the text faster and better understand and digest the content.

[0098] Through the adjustment of the emotional connection level and proportion, a more emotionally rich, consistent, and resonant typesetting effect is achieved. This method can better convey the emotional information of the text, enhance the emotional experience of readers, and promote the innovative development of emotional intelligent typesetting in the field of typesetting.

[0099] Example 5

[0100] Please refer to Figure 1 , in the S2 step, the association value TDX between the picture and the paragraph is obtained by analyzing the data between the input copywriting and the picture through artificial intelligence, including the emotional value TQG of the picture and the picture tone value TSD. The specific calculation method is as follows:

[0101] TDX = TQG * TSD.

[0102] The emotional value TQG of the picture is consistent with the emotional connection level QGX, and they have the same attributes. The picture tone value TSD is obtained by analyzing the emotional value TQG of the picture.

[0103] The picture tone value TSD includes warm tones, cold tones, and neutral tones. The specific selection method is as follows:

[0104] When the emotional value TQG of the picture is within the downturn range, the picture tone value TSD selects a cold tone, denoted as 2;

[0105] When the emotional value TQG of the picture is within the stable range, the picture tone value TSD selects a neutral tone, denoted as 1;

[0106] When the emotional value TQG of the picture is within the upsurge range, the picture tone value TSD selects a warm tone, denoted as 3.

[0107] In this embodiment: By analyzing the emotional value of the picture, the typesetting method can better capture the emotions conveyed by the picture, coordinate the emotions of the picture and the copywriting, which helps to make the typesetting result more rich and vivid, and enhances the emotional expressiveness of the typesetting.

[0108] Considering the picture tone value, the method can maintain the harmony and consistency of colors during the typesetting process. By making the colors of the copywriting and the picture match, the typesetting can be more visually appealing, improving the aesthetic feeling and coordination of the typesetting.

[0109] Integrating the emotional value and tone value of the picture, the typesetting method can better handle the relevance between the picture and the copywriting, which helps to make the coordination between the picture and the copywriting higher, making the whole typesetting more integrated and unified.

[0110] By considering the hue values of the pictures, the typesetting method can better convey the emotion and theme of the text visually, which helps readers understand the emotional atmosphere of the text more quickly and enhances the visual communication effect of the typesetting.

[0111] By comprehensively considering the emotion values and hue values, the typesetting method can better integrate the pictures with the copywriting, increasing the visual appeal of the typesetting. This helps attract readers' attention and makes them more willing to read the content in depth.

[0112] By considering the emotion values and hue values of the pictures, this method achieves a more emotional, colorful and visually coordinated typesetting effect. This method not only improves the expressiveness and attractiveness of the typesetting, but also brings more creativity and depth to the typesetting results.

[0113] The emotion value TQG of the said picture is consistent with the emotional connection QGX, and they have the same attributes. The picture hue value TSD is obtained by analyzing the emotion value TQG of the picture.

[0114] By connecting the hue of the picture with the emotion value, the organic integration of emotion and color is achieved, which helps the typesetting better convey the emotion and theme of the text, thus creating a more in-depth and consistent visual effect.

[0115] Selecting the corresponding hue according to the emotion value can make the picture more prominent and eye-catching visually. The selection of warm colors, cool colors and neutral colors brings diverse visual expressions to the picture, thus enhancing the attractiveness of the typesetting.

[0116] Different hues can create different visual atmospheres and emotions in the typesetting, thus adding more layers and depth to the typesetting. This helps readers have a more visually rich experience of the text.

[0117] By selecting the hue that matches the emotion value, the typesetting method can better guide readers to resonate with the emotion of the text. Different hues can arouse different emotional experiences of readers, thus enhancing the emotional expression effect of the typesetting.

[0118] According to different emotion values, selecting the corresponding hue makes the copywriting and pictures more visually consistent, which helps the overall typesetting result be more harmonious and presents a unified visual effect to readers.

[0119] By connecting the emotion value of the picture with different hues, the integration of emotion and color in the typesetting is achieved. This method enriches the visual effect of the typesetting, making the typesetting more emotional, attractive and consistent, and bringing a more in-depth and rich visual experience to readers.

[0120] Example 6

[0121] Please refer toFigure 1 The intervention value TPX of the said picture is obtained through the analysis of the copywriting reference data and the picture reference data by artificial intelligence, and includes the size TPC of the picture, the length DCD of a single natural paragraph of the copywriting, and the association value TDX between the picture and the paragraph. The specific calculation method is as follows:

[0122] TPX = (TPC * TDX) + (DCD * TDX) + TDX.

[0123] In this embodiment: By comprehensively analyzing the picture size, the length of the copywriting paragraph, and the association value between the picture and the paragraph, the typesetting method can better handle the coordination between the picture and the copywriting, which helps to make the typesetting result more integrated and overall coordinated.

[0124] According to the picture size and the length of the copywriting paragraph, it is possible to more accurately judge whether the picture is suitable for placement in a specific copywriting paragraph, which makes the typesetting more adaptable and enables the picture to better blend into the text.

[0125] By comprehensively considering the association value between the picture and the paragraph, the typesetting method can better optimize the layout of the paragraph, which helps to make the typesetting more reasonable and coherent, allowing the reader to have a smoother transition when reading.

[0126] By comprehensively analyzing the picture intervention value, the method can achieve a balanced arrangement of the picture and the copywriting, which helps to maintain the overall balance of the typesetting, so that the picture is neither too obtrusive or prominent nor suppressed by too much text.

[0127] By optimizing the typesetting according to the picture intervention value, this method can better convey information. Considering the association value between the picture and the paragraph can make the information presented more systematically, improving the reader's understanding efficiency of the content.

[0128] This method comprehensively considers the size of the picture, the length of the copywriting paragraph, and the association value between the picture and the paragraph, and realizes the coordination, adaptability, and balance between the picture and the copywriting. This method not only improves the overall effect and readability of the typesetting, but also provides more possibilities for the integrated presentation of the text and the picture.

[0129] Embodiment 7

[0130] Please refer to Figure 1 , in the said S3 step, the evaluation criteria include an update threshold and an update threshold. The specific evaluation method is as follows:

[0131] When the number of operations ≤ the update threshold, it represents that the practical level of this method is at a high level and no update is required;

[0132] When the update threshold < the number of operations < the update threshold, it represents that the practical level of this method is at a medium level and appropriate update and adjustment are required;

[0133] When the number of operations ≥ the update threshold, it represents that the practical level of this method is low, and significant update adjustments are required.

[0134] In this embodiment: By setting different operation - times thresholds, the typesetting method can be automatically updated according to the actual usage situation. This helps to maintain the timeliness and effect of the typesetting method to adapt to the changing requirements. The typesetting method divides different practical levels according to the number of operations, from high, medium to low. This evaluation method of practical levels can intuitively reflect the performance and status of the typesetting method, providing guidance for further updates.

[0135] The method determines the amplitude of update adjustment according to the size of the operation - times update threshold. This makes the update strategy of the typesetting method more flexible, enabling appropriate updates according to the actual situation, improving the adaptability of the method. Through automatic update, the typesetting method can continuously optimize the typesetting effect and make appropriate adjustments, which helps the typesetting method to maintain high efficiency, accuracy and meet user needs in practical applications.

[0136] Automatic update can keep the typesetting method always at a high or medium practical level, thereby improving user satisfaction with the typesetting effect. Users can always enjoy the best typesetting service.

[0137] Example:

[0138] A certain publishing house focuses on publishing creative and educational children's books. To improve the typesetting quality, they adopt an AI - based graphic and text typesetting method:

[0139] First, the publishing house uses high - performance scanning equipment to comprehensively scan the story copy and related pictures of children's books. This process generates a large amount of raw data, including copy reference data and picture reference data.

[0140] Using a specially designed calculation module, the copy reference data and picture reference data obtained from data collection are combined to calculate the typesetting reference coefficient.

[0141] Using AI technology to deeply analyze the copy to obtain the contextual and emotional connectivity between natural paragraphs. The weights of this information are compared with preset evaluation criteria according to their magnitudes to determine their proportions in the typesetting reference coefficient.

[0142] When the contextual connectivity LXX is defined as 9 through analysis, it represents that the connectivity between the upper and lower paragraphs is strong, and it is not appropriate to insert a paragraph between them.

[0143] At the same time, the emotional connection QGX is analyzed and defined as 8, representing that the emotion in the current paragraph is high, including positive emotions such as happiness, pleasure, and excitement. Therefore, the picture emotion value TQG is also synchronously defined as 8, that is, in the high range. The picture color tone value TSD selects warm colors and is synchronously recorded as 3.

[0144] By obtaining the layout reference coefficient PBX as 45.3% through the calculation formula, it represents that this area is not suitable for setting pictures. Therefore, the next paragraph is analyzed in sequence.

[0145] The new paragraph analysis is defined as 2, representing that the relevance between the new paragraph and the previous paragraph is relatively low. At the same time, the emotional connection QGX is analyzed and defined as 3, representing that the emotion in the current paragraph is low, including negative emotions such as anger, sadness, and grief. Therefore, the picture emotion value TQG is also synchronously defined as 2, that is, in the low range. The picture color tone value TSD selects cold colors and is synchronously recorded as 2.

[0146] By obtaining the layout reference coefficient PBX as 82.9% through the calculation formula, it represents that this area is suitable for setting pictures. Therefore, a suitable picture is inserted between the new paragraph and the previous paragraph.

[0147] Using the execution module, the text of the children's book is analyzed and typeset again according to the calculated layout reference coefficient. The typeset result is sent to the detection port, and professional staff will conduct review, modification, and optimization. The detection port will record the number of operation times for each modification, and by comparing with the preset evaluation criteria, it provides a reference for the optimization and update of the method.

[0148] After the final confirmation by the staff, the terminal module sends the final version to the relevant website or printing manufacturer. The final version will be pushed on the network platform and will also be printed physically for children to read and enjoy.

[0149] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise", or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article, or device.

[0150] The above is only the preferred implementation manner of the present invention. It should be pointed out that for those of ordinary skill in the art in this technical field, without departing from the technical principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A graphic typesetting method based on artificial intelligence, characterized in that: The following processes are included: S1. Scan the story text and pictures through the acquisition device to obtain text reference data and picture reference data respectively; S2. The collected copy reference data and image reference data are combined and calculated through the calculation module to obtain the typesetting reference coefficient , the specific calculation method is as follows: In the formula is the adaptation value between paragraphs, obtained from the copy reference data, and is expressed as the contextual relevance between several natural paragraphs in the copy. is its weight value in the calculation, representing its reference coefficient in typesetting The proportion of the calculation; In the formula is the association value between the image and the paragraph, which is obtained by combining the text reference data and the image reference data, and is expressed as the association between the image and the natural paragraph in the text. is its weight value in the calculation, representing its reference coefficient in typesetting The proportion of the calculation; In the formula is the intervention value of the image, which is obtained by combining the text reference data and the image reference data, and is expressed as the degree of fit between the image and the text paragraph. is its weight value in the calculation, representing its reference coefficient in typesetting The proportion of the calculation; In the formula , , The value of is set by the customer, and , ; In the formula , as well as All of them are correction coefficients, and the values ​​of the three are selected and set by the customer; The adaptation value of the story segment in S2 , obtained by analyzing the correlation between the natural paragraphs of the input copy, including the contextual connection between the natural paragraphs and the emotional connection between paragraphs , the specific calculation formula is as follows: In the formula and Contextual Relevance and emotional connection The weight value of and The value of is set by the customer, and , ; The association value between the picture and the paragraph in S2 Through artificial intelligence, the data between the input copy and the picture is analyzed and obtained, including the emotional value of the picture And the image tone value , the specific calculation method is as follows: ； The intervention value of the image The copy reference data and image reference data are obtained through artificial intelligence analysis, including the size of the image and the length of a single paragraph of the copy And the association value between the image and the paragraph , the specific calculation method is as follows: ； S3, the typesetting reference coefficient obtained by calculation through the execution module The text is analyzed again and typeset, and the typeset results are sent to the test port. The staff tests the typeset files to provide modifications, and the test port records the number of single modification operations and the number of modification operations according to the evaluation criteria to provide a reference for subsequent updates of the method; S4. After the modification is completed and confirmed, the execution module sends the modified typesetting method to the terminal module. After receiving the typesetting data, the terminal module actually typesets the input documents and pictures and sends them to the detection port for confirmation again; S5. After the staff makes the final confirmation, the terminal module will send the final draft to the relevant website or printing manufacturer for push and physical printing.

2. The method for typesetting text and images based on artificial intelligence according to claim 1, characterized in that: The contextual connection , obtained through artificial intelligence analysis of copywriting, and compared with the evaluation criteria according to the size of its value, to determine its proportion. The specific analysis method is as follows: When contextual relevance When it is between 0 and 4, it means that the contextual connection between paragraphs is of low level, and the proportion is adjusted synchronously; When contextual relevance When it is between 5 and 7, it means that the contextual connection between paragraphs is of medium level, and the proportion is adjusted synchronously; When contextual relevance When it is between 8 and 10, it means that the contextual connection between paragraphs is at a high level, and the proportion is adjusted synchronously.

3. The method for typesetting text and images based on artificial intelligence according to claim 2, characterized in that: Emotional connection , the copy is analyzed and obtained through artificial intelligence, and its weight is adjusted according to the size of its value and the evaluation criteria. The specific analysis and adjustment methods are as follows: When emotional connection When it is between 0 and 3, it means that the emotion of the current paragraph is low, and the proportion is adjusted synchronously; When emotional connection When it is between 3 and 5, it means that the emotion of the current paragraph is stable, and the proportion is adjusted synchronously; When emotional connection When it is between 6 and 10, it means that the emotion of the current paragraph is high, and the proportion is adjusted synchronously.

4. The method for typesetting text and images based on artificial intelligence according to claim 3, characterized in that: The sentiment value of the image Emotional connection The two have the same attributes, and the image hue value By the emotional value of the picture Analyze acquisition.

5. The method for typesetting text and images based on artificial intelligence according to claim 4, characterized in that: The image hue value Including warm tones, cool tones and neutral tones, the specific selection method is as follows: When the emotional value of the picture When in the low range, the image tone value Select cool colors, record as 2; When the emotional value of the picture When it is within the stable range, the image hue value Select neutral tones, recorded as 1; When the emotional value of the picture When in the high range, the image hue value Select warm tones and record it as 3.

Images