A data analysis and construction method and system for color painting
By analyzing the recessive and dominant gene characteristic values of color paintings, generating reconstructed pictures, and evaluating the accuracy of gene analysis of basic paintings, the subjective problem of art and culture data analysis in the existing technology is solved, and the characteristics of color paintings are realized Comprehensive and accurate understanding improves the accuracy and efficiency of data analysis.
Patent Information
- Application Number
- CN202411139644.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-20
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2044-08-20
AI Technical Summary
The existing technology relies too much on intuitive perception in art and culture data analysis, which leads to the subjective color of data collection and is difficult to provide rigorous scientific basis, resulting in one-sidedness and deviation of feature analysis.
By collecting texture information of colored paintings, analyzing recessive gene eigenvalues and dominant gene eigenvalues, generating base painting reconstructed pictures, evaluating the gene analysis accuracy of base paintings, and realizing the data analysis process.
It realizes objective analysis of color painting structure information, accurately grasps physical characteristics and style characteristics, provides a rich data research platform, improves the accuracy and processing efficiency of data analysis, and reduces the influence of subjective factors.
Smart Images

Figure CN118658013B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of color painting data analysis and processing, and in particular to a data analysis and construction method and system for color paintings. Background Art
[0002] With the continuous development of digital technology, the data analysis and construction technology of color paintings are also constantly improving. However, the existing color painting data analysis mostly focuses on color painting storage and simple retrieval, which makes it difficult to store the comprehensive characteristics and value of color paintings. As an important part of human cultural heritage, the comprehensive characteristics and values behind color paintings are crucial to cultural communication.
[0003] For example, the invention patent with the announcement number CN101887366B announces the non-realistic rendering in computer graphics, especially the digital simulation and synthesis technology of the artistic style of Yunnan heavy color painting, which belongs to the field of non-realistic graphics rendering technology; its features include the establishment and management module of the basic graphic element library of Yunnan heavy color painting, the Yunnan heavy color painting line drawing module based on the graphic element library and the coloring rendering module based on the Yunnan heavy color painting line drawing; the basic graphic element library is used to provide the graphic elements required for the foreground line drawing, such as hair, face shape, torso, etc.; the Yunnan heavy color painting line drawing module can select suitable graphic elements from the graphic element library for layer combination to produce the Yunnan heavy color painting line drawing; in order to coordinate the various graphic elements, the system provides various operations on the graphic elements, such as translation, selection, scaling, etc., and can also modify the shape of the graphic elements; the coloring rendering module based on the Yunnan heavy color painting line drawing is responsible for coloring the line drawing, and the coloring effect can reflect the texture characteristics and color characteristics of the original heavy color painting. In addition, it also realizes some unique feature simulations of Yunnan heavy color painting, such as the hollow feeling of the edge of the figure and the background fusion.
[0004] For example, the invention patent with announcement number CN109901905B announces a method, device, equipment and computer-readable storage medium for picture color modulation, the method comprising: obtaining picture illumination information, current time information and color correspondence information, wherein the color correspondence information comprises a correspondence between preset colors and illumination information and time information; adjusting the color of the picture according to the picture illumination information, current time information and the correspondence; the provided method, device, equipment and computer-readable storage medium can modulate the picture color according to the current time, so that the color of the picture changes with time, is more consistent with the external environment, and can improve the user experience.
[0005] However, in the process of implementing the embodiments of the present application, it was found that the above-mentioned technology has at least the following technical problems: currently, when analyzing data related to art and culture, people often rely too much on intuitive perception and only obtain superficial information "at first glance". At the same time, the collection of data in art and culture is highly subjective. The limitations of this data collection make it difficult to provide a more rigorous scientific basis for the subsequent evaluation of the characteristics of art and culture data, resulting in one-sidedness and bias in the characteristics of art and culture data. Summary of the invention
[0006] In view of the deficiencies in the prior art, the present invention provides a method and system for constructing data analysis for color paintings, which can effectively solve the problems involved in the above-mentioned background technology.
[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: In the first aspect, the present invention provides a data analysis and construction method for colored paintings, including: collecting texture information of colored paintings, analyzing recessive gene characteristic values of colored paintings, analyzing and obtaining recessive gene element sets of colored paintings, analyzing dominant gene characteristic values of colored paintings, and analyzing and obtaining dominant gene element sets of colored paintings; S2. Integrating the recessive gene element sets and dominant gene element sets of colored paintings to generate a colored painting reconstruction map, thereby evaluating the accuracy of gene analysis of colored paintings; S3. Comparing the accuracy of gene analysis of colored paintings with the accurate threshold of gene analysis, and providing feedback on the data analysis of colored paintings, thereby completing the data analysis process of colored paintings.
[0008] As a further method, the recessive gene element set of the colorful painting is obtained by analysis. The specific analysis process is: matching the recessive gene characteristic value of the colorful painting with the recessive gene element set corresponding to each recessive gene characteristic value interval stored in the colorful painting information library, thereby obtaining the recessive gene element set of the colorful painting by analysis.
[0009] As a further method, the analysis obtains the dominant gene element set of the colorful painting. The specific analysis process is: matching the dominant gene characteristic value of the colorful painting with the dominant gene element set corresponding to each dominant gene characteristic value interval stored in the colorful painting information library, thereby analyzing and obtaining the dominant gene element set of the colorful painting.
[0010] As a further method, the data analysis process of the color painting is completed, and the specific analysis process is: comparing the gene analysis accuracy of the color painting with the gene analysis accuracy threshold. If the gene analysis accuracy of the color painting is greater than or equal to the gene analysis accuracy threshold, feedback is given to the data analysis of the color painting, thereby completing the data analysis process of the color painting.
[0011] The second aspect of the present invention provides a system for constructing a data analysis method for colored paintings, including: a colored painting gene element set analysis module, which is used to collect texture information of colored paintings, analyze recessive gene characteristic values of colored paintings, and analyze to obtain recessive gene element sets of colored paintings, and analyze dominant gene characteristic values of colored paintings to obtain dominant gene element sets of colored paintings; a gene analysis accuracy evaluation module, which is used to integrate recessive gene element sets and dominant gene element sets of colored paintings, generate a colored painting reconstruction map, and thereby evaluate the gene analysis accuracy of colored paintings; a colored painting data analysis feedback module, which is used to compare the gene analysis accuracy of colored paintings with the gene analysis accuracy threshold, and provide feedback on the data analysis of colored paintings, thereby completing the data analysis process of colored paintings.
[0012] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:
[0013] (1) The present invention provides a method and system for constructing data analysis of colored paintings. By collecting texture information of colored paintings and analyzing recessive gene characteristic values and dominant gene characteristic values of colored paintings, the structural information of colored paintings can be objectively analyzed, and the physical characteristics and style features of colored paintings can be more accurately grasped. Thus, the recessive gene element set and the dominant gene element set of colored paintings can be analyzed respectively, which is helpful to fully and deeply understand the obvious element characteristics of colored paintings. At the same time, a rich data research platform is provided for subsequent research and comparison of colored paintings. The recessive gene element set and the dominant gene element set of colored paintings are integrated to evaluate the accuracy of gene analysis of colored paintings. The accuracy of gene analysis of colored paintings is compared with the accurate threshold of gene analysis, and feedback is given to the data analysis of colored paintings, thereby completing the data analysis process of colored paintings, making the data analysis process of colored paintings interactive. The color painting analysis parameters can be adjusted in time according to the color painting data analysis results to improve the accuracy and processing efficiency of color painting data analysis. At the same time, the feedback mechanism also provides users with intuitive information about the color painting analysis results.
[0014] (2) By analyzing the texture complexity, network connection complexity and texture roughness of the paintings, the present invention can capture the high-resolution details of the paintings in real time and accurately, thereby greatly improving the accuracy and efficiency of external expression feature extraction, which is conducive to comprehensively and accurately reflecting the physical properties and visual effects of the paintings, thereby making the dominant gene characteristic values obtained by analysis highly representative and comparable.
[0015] (3) By analyzing the maximum difference in visual weight, fading index, and usage frequency of various patterns of the paintings, the present invention can greatly improve the ability to process painting image data while ensuring the accuracy of analysis, and quickly and accurately reveal the visual focus, hierarchical relationship, and color delicacy of the paintings, thereby refining and summarizing the inherent artistic style and cultural connotation of the paintings as recessive gene characteristic values. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.
[0017] Figure 1 The figure is a schematic flow chart of the method steps of the present invention.
[0018] Figure 2 It is a schematic diagram of system module connection of the present invention.
[0019] Figure 3 It is a schematic diagram of the directional gradient histogram involved in the present invention.
[0020] Figure numerals: 1. Directional gradient curve. DETAILED DESCRIPTION
[0021] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0022] Reference Figure 1 As shown, the first aspect of the present invention provides a data analysis and construction method for color paintings, including: S1. Collecting texture information of color paintings, analyzing recessive gene characteristic values of color paintings, analyzing and obtaining recessive gene element sets of color paintings, analyzing dominant gene characteristic values of color paintings, and analyzing and obtaining dominant gene element sets of color paintings; wherein the recessive genes and dominant genes of color paintings borrow the concepts of gene dominance and recessiveness in biology to analogize certain characteristics in color painting art, and this analogy helps to more deeply understand and analyze the artistic style and cultural connotation of color paintings; dominant genes can be analogized to direct, obvious and easily perceived artistic characteristics by observers, such as the use of colors; recessive genes in color painting art represent those artistic characteristics that are not directly revealed and require in-depth observation and analysis to be perceived, such as artistic concepts.
[0023] The above-mentioned collection of texture information of the painting is specifically to quickly extract information of each point on the surface of the painting, including reflectivity and color, through a three-dimensional laser scanner, and to capture high-definition images of the painting with a high-resolution camera to obtain the color, texture and detail information of the painting. The texture information of other paintings is specifically described in the following instructions.
[0024] Specifically, the analysis obtains the recessive gene element set of the painting, and the specific analysis process is: matching the recessive gene characteristic value of the painting with the recessive gene element set corresponding to each recessive gene characteristic value interval stored in the painting information library, thereby analyzing and obtaining the recessive gene element set of the painting; the specific matching process is: referring to each recessive gene characteristic value interval corresponding to the recessive gene element set, querying the recessive gene characteristic value interval corresponding to the recessive gene characteristic value of the painting for matching, then the recessive gene element set corresponding to the recessive gene characteristic value interval is the recessive gene element set of the painting; the recessive gene element sets corresponding to the above-mentioned recessive gene characteristic value intervals are formulated by painting analysis experts based on historical painting analysis work experience.
[0025] Specifically, the recessive gene characteristic value of the painting is analyzed, and the specific analysis process is as follows: the painting is divided into each painting sub-region, based on the texture information of the painting, the color value of each pixel point of each painting sub-region is extracted, and the visual weight value corresponding to the color value of each pixel point preset in the painting information library is matched, and the visual weight value of each painting sub-region is analyzed and obtained, and the visual weight value ranked first and last is extracted, and the difference processing is performed to obtain the maximum difference of the visual weight of the painting; the color value of each pixel point of each painting sub-region refers to the color value of each painting sub-region obtained by traversing each painting sub-region through the open source computer vision library. The color values of each pixel point, by combining computer vision technology and image processing algorithms, can efficiently extract color information in the image, and provide data support for subsequent image analysis, color recognition or art research; the visual weight values corresponding to the color values of each pixel point are obtained by painting analysis experts based on color psychology, visual perception principles and experience accumulated in design practice; the visual weight values of each painting sub-area are obtained by the above analysis, and the visual weight values corresponding to the color values of each pixel point that matches each painting sub-area are queried from the painting information database, and the visual weight values corresponding to the color values of each pixel point are accumulated to obtain the visual weight values of each painting sub-area.
[0026] It should be explained that the above-mentioned painting information database is a database for storing parameters involved in the painting analysis process.
[0027] The average red, green and blue values of each painting sub-region are obtained, and the brightness of each painting sub-region is obtained through data processing. The brightness of each painting sub-region is subjected to difference processing with the brightness of the corresponding adjacent painting sub-region to obtain the brightness difference of each painting sub-region, and the de-blurring index of the painting is obtained through variance processing; the average red, green and blue values of each painting sub-region are obtained by respectively performing mean processing on the color values of each pixel point in each painting sub-region. The brightness of each painting sub-region is obtained through data processing. The specific data processing procedure is as follows: ,in is the brightness of the a-th painted sub-region, is the average red color level of the a-th painted sub-area, that is, the average red value among the average red, green and blue values. is the average green color level of the a-th color painting sub-area, that is, the average green value among the average red, green and blue values. is the average blue color level of the a-th color painting sub-region, that is, the average blue value in the average red, green and blue values; the above variance processing of the brightness difference of each color painting sub-region is specifically: ,in is the result of processing the average brightness difference of each painted sub-area, is the brightness difference of the a-th painted sub-region, a is the number of each painted sub-region, , x is the total number of painted sub-areas.
[0028] The usage count of various patterns of the painting is extracted from the texture information of the painting, and the ratio is processed with the total number of patterns of the painting to obtain the usage frequency of each pattern of the painting. The specific identification method of the various patterns of the above-mentioned painting is as follows: the painting is converted into a binary image through a high-resolution scanner, that is, the pixels in the image are either black or white, so that it becomes a high-precision digital image file, the contour of each pattern in the painting is extracted through a contour tracking algorithm, the Fourier descriptor of each pattern contour is calculated, the similarity of the Fourier descriptors of each pattern contour is analyzed, and the pattern contours with a similarity greater than 90% are classified into the same category, thereby classifying various patterns, and counting the number of patterns in each pattern, which is the usage count of various patterns of the painting.
[0029] The recessive gene characteristic value of the painting is obtained by comprehensive analysis. In this embodiment, the recessive gene characteristic value of the painting is obtained by comprehensive analysis of the maximum difference in visual weight of the painting, the fading index, and the frequency of use of each type of pattern. It is a quantitative evaluation value of the recessive gene characteristic of the painting. The specific expression is:
[0030] ;
[0031] in, is the recessive gene characteristic value of color painting. The maximum difference in visual weight, as a key indicator to measure the visual impact and sense of balance of the picture, shows the artist's precise control of visual dynamics. The fading index reveals the delicacy of color painting in the use of colors. Through the gradual change and fusion of colors, it creates a rich spatial layering and emotional atmosphere. The frequency of use of various patterns reflects the integration and innovation of tradition and modernity in the creative concept of color painting, and also demonstrates the persistence and pursuit of cultural inheritance and individual development in terms of values. A higher fading index can improve the visual clarity of the color painting, thereby reducing the maximum difference in visual weight. A higher frequency of pattern use also indicates that the color painting has a higher degree of harmony, which increases the fading index and further reduces the maximum difference in visual weight. Together, they constitute a quantitative evaluation system for the intrinsic style strength of the color painting, that is, quantitatively evaluating the recessive gene characteristic value of the color painting. Through the above data processing, we can deeply explore the detailed information in the color painting image, capture the recessive features that are difficult to detect by traditional subjective evaluation methods, and realize the accurate identification and classification of the color painting features, thereby improving the accuracy of the color painting gene characteristics.
[0032] It is the maximum difference in the visual weight of the painting, reflecting the maximum value of the weight difference produced by different colors, textures or composition elements in the painting in visual perception. Visual weight has an important influence on the overall balance, dynamics and visual impact of the painting.
[0033] It is the fading index of the painting. This parameter plays an important role in the overall atmosphere and visual effect of the painting. It is also an important parameter reflecting the artistic depth and expressiveness of the painting. A high fading index may mean that the painting has a stronger spatial depth and sense of layering.
[0034] The frequency of use of the f-type patterns of colored paintings reflects the design choices and style preferences in the creation process of colored paintings, and also reflects the cultural, historical or artistic value carried by the pattern. Different patterns have different symbolic meanings and aesthetic characteristics, and their frequency of use directly affects the overall style and characteristics of the colored paintings.
[0035] It is the influence factor corresponding to the maximum visual weight difference unit value preset in the painting information library, which represents the influence degree of the maximum visual weight difference unit value on the recessive gene characteristic value of the painting. The fitting curve corresponding to the maximum visual weight difference is obtained by fitting the relationship between the historical color brightness, the historical color saturation and the maximum visual weight difference, and the fitting curve corresponding to the maximum visual weight difference is brought into the fitting curve according to the real-time color brightness and color saturation, thereby obtaining the influence factor corresponding to the maximum visual weight difference unit value in this example. The value range of this value in this example is 0 to 1.
[0036] It is the influence factor corresponding to the unit value of the fading index preset in the color painting information database, which represents the influence degree of the unit value of the fading index on the recessive gene characteristic value of the color painting. The fitting curve corresponding to the fading index is obtained by fitting the relationship between the historical number of transition layers, the historical color gradient rate and the fading index, and the fitting curve corresponding to the fading index is brought into the fitting curve according to the real-time number of transition layers and the color gradient rate, thereby obtaining the influence factor corresponding to the unit value of the fading index in this example. The value range of this value in this example is 0 to 1.
[0037] It is the impact factor corresponding to the preset frequency unit value in the painting information database, which represents the degree of influence of the frequency unit value on the recessive gene characteristic value of the painting. The fitting curve corresponding to the frequency of use is obtained by fitting the relationship between the historical pattern usage density, the historical pattern number and the usage frequency, and the fitting curve corresponding to the frequency of use is brought into the fitting curve according to the real-time pattern usage density and the number of patterns, thereby obtaining the impact factor corresponding to the frequency unit value in this example. The value range of this value in this example is 0 to 1.
[0038] f is the number of each pattern. , s is the total number of pattern types, and e is a natural constant.
[0039] In a specific embodiment, the present invention can greatly improve the ability to process painting image data while ensuring analysis accuracy by analyzing the maximum difference in visual weight of the painting, the fading index, and the frequency of use of various patterns of the painting, and quickly and accurately reveal the visual focus, hierarchical relationship, and color delicacy of the painting, thereby refining and summarizing the inherent artistic style and cultural connotation of the painting as recessive gene characteristic values.
[0040] Specifically, the analysis obtains the dominant gene element set of the painting, and the specific analysis process is: matching the dominant gene characteristic value of the painting with the dominant gene element set corresponding to each dominant gene characteristic value interval stored in the painting information library, thereby analyzing and obtaining the dominant gene element set of the painting; the specific matching process is: referring to each dominant gene characteristic value interval corresponding to the dominant gene element set, querying the dominant gene characteristic value interval corresponding to the dominant gene characteristic value of the painting for matching, then the dominant gene element set corresponding to the dominant gene characteristic value interval is the dominant gene element set of the painting; the dominant gene element set corresponding to the above-mentioned dominant gene characteristic value interval is formulated by the painting analysis experts based on the historical painting analysis work experience.
[0041] Furthermore, the specific analysis process of the dominant gene characteristic value of the painting is as follows: extracting the grayscale symbiosis matrix of the painting from the texture information of the painting, obtaining the grayscale symbiosis matrix entropy of the painting through data processing, matching it with the texture complexity corresponding to each grayscale symbiosis matrix entropy interval stored in the painting information library, thereby obtaining the texture complexity of the painting; the grayscale symbiosis matrix of the painting refers to the parameters such as the number of gray levels, distance and angle specified by the painting analysis expert and obtained by analyzing the painting through the matrix laboratory software; the grayscale symbiosis matrix entropy of the painting obtained through data processing is , refers to the matrix laboratory software calculating the grayscale symbiosis matrix of the painting according to the definition of entropy value, thereby obtaining the grayscale symbiosis matrix entropy of the painting; the texture complexity of the above-mentioned painting, the specific matching process is: each grayscale symbiosis matrix entropy interval corresponds to the texture complexity, query the grayscale symbiosis matrix entropy interval corresponding to the grayscale symbiosis matrix entropy of the painting, then the texture complexity corresponding to the grayscale symbiosis matrix entropy interval is the texture complexity of the painting; the texture complexity corresponding to the above-mentioned grayscale symbiosis matrix entropy interval is formulated by painting analysis experts based on the experience of historical painting analysis.
[0042] The shape elements of the painting are located from the texture information of the painting, and the correlation between the shape elements of the painting and the corresponding adjacent shape elements is extracted. The shape elements with a correlation greater than a correlation threshold are connected to the corresponding adjacent shape elements to obtain the number of connections of the shape elements of the painting, which is matched with the network connection complexity corresponding to the interval of the number of connections of each shape element stored in the painting information library, thereby obtaining the network connection complexity of the painting; the above-mentioned correlation threshold represents the minimum value of a reasonable range of correlation between the shape elements of the painting and the corresponding adjacent shape elements, and is formulated by painting analysis experts based on the experience of historical painting analysis; the shape elements of the painting are segmented through the encoding-decoding structure and jump connection design in the U-Net model in the deep learning model, Each independent shape element in the painting is separated out, so that more detail information is retained, high-precision segmentation can be achieved, and each shape element can be accurately identified and separated; the correlation between each shape element of the above-mentioned extracted painting and the corresponding adjacent shape element is obtained by painting analysis experts based on historical experience marking; the network connection complexity of the above-mentioned painting, the specific matching process is: each shape element connection number interval corresponds to the network connection complexity, and the shape element connection number interval corresponding to the shape element connection number of the painting is queried, then the network connection complexity corresponding to the shape element connection number interval is the network connection complexity of the painting; the network connection complexity corresponding to the above-mentioned shape element connection number interval is formulated by painting analysis experts based on historical painting analysis work experience.
[0043] Obtain the directional gradient histogram of the painting, locate and smoothly connect the top center position points of the square columns in the directional gradient histogram, obtain the directional gradient curve of the painting, obtain the rate of change function of the directional gradient curve through data processing, and use the matrix laboratory software to write a script to read the painting image and convert it into a grayscale image. Apply the directional gradient histogram algorithm to extract image features, thereby generating the directional gradient histogram of the painting, such as Figure 3 As shown, the horizontal axis is the gradient direction, the unit is degree, and the vertical axis is the gradient intensity, which is usually dimensionless. In this embodiment, the unit is marked as percentage. The directional gradient histogram of the painting can well capture the shape changes of the painting elements, and at the same time has a certain invariance to the overall translation, rotation and other geometric transformations of the painting; the directional gradient curve of the above-mentioned painting, such as Figure 3 As shown in the mark 1 in FIG. 1 , the directional gradient curve function is obtained by data processing using the matrix laboratory software, and the function is derived to obtain the rate of change function of the directional gradient curve.
[0044] The change rate functions corresponding to the center positions of the tops of adjacent square columns are subjected to difference processing and mean analysis in turn to obtain the average change rate of the directional gradient curve of the painting, which is matched with the texture roughness corresponding to each average change rate interval stored in the painting information database, and finally the texture roughness of the painting is matched. The average change rate of the directional gradient curve of the painting is obtained by averaging the results of difference processing of the change rate functions corresponding to the center positions of the tops of adjacent square columns; the specific matching process of the texture roughness of the painting is as follows: each average change rate interval corresponds to the texture roughness, and the average change rate interval corresponding to the average change rate of the directional gradient curve of the painting is queried, then the texture roughness corresponding to the average change rate interval is the texture roughness of the painting; the texture roughness corresponding to each average change rate interval is formulated by painting analysis experts based on the experience of historical painting analysis.
[0045] The dominant gene characteristic value of the painting is obtained through comprehensive analysis. In this embodiment, the dominant gene characteristic value of the painting is obtained through comprehensive analysis of the texture complexity, network connection complexity and texture roughness of the painting, which is a quantitative evaluation value of the dominant gene characteristic of the painting. The specific expression is:
[0046] ;
[0047] in, is the dominant gene characteristic value of the painting. The painting with high texture complexity may be more delicate and complex in composition, creating a rich visual effect through multi-level texture superposition and interlacing. Texture roughness is one of the important components of the dominant characteristics of the painting. Rough texture may increase the texture and depth of the painting, while fine texture may make the painting look more delicate and exquisite. The network connection complexity of the painting reflects the interaction and correlation between the elements of the painting. If the network connection complexity is high, the overall visual effect of the painting may be more harmonious and unified. A high texture complexity indicates that the composition of the painting is More complex may increase the texture roughness, and higher texture roughness may indicate that the painting elements are more complex, which increases the network connection complexity of the painting. The three parameters are used together for the composition method, drawing technology and visual effects of the painting, giving the painting a unique artistic charm and aesthetic value. Therefore, through the quantitative evaluation of the characteristics of the painting, the artistic information of the painting can be more accurately recorded and preserved, which can reduce the influence of subjective factors on the evaluation results of the painting, making the gene analysis of the painting more objective and fair. At the same time, the standardized evaluation system also provides a unified standard and basis for the classification, archiving and display of the painting.
[0048] The texture complexity of a painting refers to the richness of the texture elements in the painting, which reflects the richness and complexity of the visual details of the painting.
[0049] It is the network connection complexity of the painting, which reflects the degree of association between the shape elements in the painting.
[0050] It is the roughness of the texture of the painting, reflecting the visual texture and tactile effect of the painting.
[0051] It is the influence factor corresponding to the texture complexity unit value preset in the painting information library, which represents the influence degree of the texture complexity unit value on the dominant gene characteristic value of the painting. The fitting curve corresponding to the texture complexity is obtained by fitting the relationship between the historical grayscale level, the historical window size and the texture complexity, and the fitting curve corresponding to the texture complexity is brought into the fitting curve according to the real-time grayscale level and the window size, thereby obtaining the influence factor corresponding to the texture complexity unit value in this example. The value range of this value in this example is 0 to 1.
[0052] It is the impact factor corresponding to the unit value of network connection complexity preset in the painting information library, which represents the degree of influence of the unit value of network connection complexity on the dominant gene characteristic value of the painting. The fitting curve corresponding to the network connection complexity is obtained by fitting the relationship between the number of historical nodes, the size of historical paintings and the network connection complexity, and the fitting curve corresponding to the network connection complexity is brought into the fitting curve according to the real-time number of nodes and the size of the paintings, thereby obtaining the impact factor corresponding to the unit value of network connection complexity in this example. The value range of this value in this example is 0 to 1.
[0053] It is the influence factor corresponding to the texture roughness unit value preset in the painting information library, which represents the influence degree of the texture roughness unit value on the dominant gene characteristic value of the painting. The fitting curve corresponding to the texture roughness is obtained by fitting the relationship between the historical measurement direction, the historical pigment type and the texture roughness, and the fitting curve corresponding to the texture roughness is brought into the fitting curve according to the real-time measurement direction and the pigment type, thereby obtaining the influence factor corresponding to the texture roughness unit value in this example. The value range of this value in this example is 0 to 1.
[0054] In this exemplary embodiment, the variation table of the dominant gene characteristic value of the above-mentioned color painting and its corresponding parameter is shown in Table 1:
[0055] Table 1 The changes of dominant gene characteristic values and corresponding parameters of colored painting
[0056]
[0057] In this example embodiment, the value of the influence factor corresponding to the unit value of texture complexity is set to 0.25, the value of the influence factor corresponding to the unit value of network connection complexity is set to 0.32, and the value of the influence factor corresponding to the unit value of texture roughness is set to 0.28. It can be seen from Table 1 that the texture complexity, network connection complexity and texture roughness are not simply linear changes with the dominant gene characteristic value. Even a slight change in the input parameters may lead to a significant change in the dominant gene characteristic value. Therefore, the dominant gene characteristic value of the painting can be flexibly judged to match the best dominant gene element set.
[0058] In a specific embodiment, the present invention can capture the high-resolution details of the painting in real time and accurately by analyzing the texture complexity, network connection complexity and texture roughness of the painting, thereby greatly improving the accuracy and efficiency of external expression feature extraction, which is conducive to comprehensively and accurately reflecting the physical properties and visual effects of the painting, thereby making the dominant gene characteristic values obtained by analysis highly representative and comparable.
[0059] S2. The recessive gene element set and dominant gene element set of the painting are integrated to generate a painting reconstruction map, thereby evaluating the accuracy of the genetic analysis of the painting.
[0060] Specifically, the generation of the color painting reconstruction map, the specific generation process is: the recessive gene element set of the color painting and the dominant gene element set of the color painting are input into the programming program, thereby generating the color painting reconstruction map; the specific programming program is: the Python programming program reads the characteristic parameters in the recessive gene element set and the dominant gene element set, and normalizes them to make them suitable for subsequent processing, and uses the image reconstruction algorithm based on feature matching in the Python programming program, inputs the read characteristic parameters in the recessive gene element set and the dominant gene element set, and passes them to the image reconstruction algorithm based on feature matching, so as to generate the color painting reconstruction map.
[0061] In an example embodiment, assuming that the feature parameters in the recessive gene element set include texture complexity of 5 and brushstroke style of smooth, the feature parameters in the dominant gene element set include main color of blue, background color of white, width of 800 pixels, height of 600 pixels, and the theme feature list includes "mountain", "water" and "tree", the following Python programming language can be used to generate a color painting reconstruction map. The following pseudo code is given in this example embodiment:
[0062] GenerateImageFromGenes(implicitGenes,explicitGenes):
[0063] #Assume that implicitGenes and explicitGenes are the characteristic parameters of the recessive gene element set and the dominant gene element set respectively
[0064] #1. Extract image parameters
[0065] width = explicitGenes['width']
[0066] height = explicitGenes['height']
[0067] Main color = explicitGenes['main_color']
[0068] Background color=explicitGenes['background_color']
[0069] Texture complexity = implicitGenes['texture_complexity']
[0070] Brush stroke style = implicitGenes['brush_stroke_style']
[0071] Theme features = explicitGenes['theme_features'] #This is a feature parameter related to the theme
[0072] #2. Initialize an empty image (pseudo operation)
[0073] Image = Create a new Image (width, height, background-color)
[0074] #3. Simulate the image reconstruction algorithm based on feature matching (highly simplified)
[0075] Reconstruct image content = reconstruct image content using feature matching (dominant color, texture complexity, brushstroke style, theme features)
[0076] Define functions to reconstruct image content (dominant color, texture complexity, brush stroke style, theme features) using feature matching
[0077] Returns the generated image content (based on the provided features)
[0078] #4. Apply the reconstructed image content to the empty image (pseudo operation)
[0079] DrawContentsToImage(image, reconstructImageContents)
[0080] #5. Save or display the image (pseudo operation)
[0081] saveimage(image,'reconstructedimage.png')
[0082] #Call the function and pass in the recessive gene and dominant gene
[0083] implicitGenes={
[0084] 'texture_complexity': 5,
[0085] 'brush_stroke_style': 'Smooth'
[0086] }
[0087] explicitGenes={
[0088] 'width':800,
[0089] 'height':600,
[0090] 'main_color':'#0000FF',
[0091] 'background_color':'#FFFFFF',
[0092] 'theme_features':['mountain', 'water', 'trees']#hypothetical theme feature list
[0093] }
[0094] GenerateImageFromGenes(implicitGenes,explicitGenes)
[0095] Specifically, the accuracy of the gene analysis of the painting, in this embodiment, is obtained by comprehensive analysis of the overlap of the line contours of the painting and the structural similarity index between the reconstructed image of the painting and the painting, and is used to determine the accuracy of the gene analysis of the painting. The specific expression is:
[0096] ;
[0097] in, The accuracy of genetic analysis of the painting is that the line contour overlap is high, indicating that the painting and the reconstructed painting are highly similar, further indicating that the reconstructed painting retains most of the original structural features of the painting, making the structural similarity index of the two relatively high. The degree of closeness between the reconstructed painting and the original painting is measured from the perspective of line contour and structural similarity, thereby reflecting the accuracy and effectiveness of the genetic analysis method, providing important reference indicators for the data analysis method of the painting, and providing data support for the improvement of the data analysis method of the painting.
[0098] It is the coincidence degree of the line contour of the painting, which is used to evaluate the similarity between the generated painting reconstruction image and the original painting in terms of line contour.
[0099] It is the structural similarity index between the reconstructed painting and the painting, which is used to measure the structural similarity between the reconstructed painting and the original painting.
[0100] It is the influence factor corresponding to the unit value of line outline coincidence preset in the painting information database, which represents the influence degree of the unit value of line outline coincidence on the accuracy of genetic analysis of the painting. The fitting curve corresponding to the line outline coincidence is obtained by fitting the relationship between the historical painting line complexity, the historical painting line overlap rate and the line outline coincidence. The fitting curve corresponding to the line outline coincidence is brought into the fitting curve according to the real-time painting line complexity and the painting line overlap rate, thereby obtaining the influence factor corresponding to the unit value of line outline coincidence in this example. The value range of this value in this example is 0 to 1.
[0101] It is the impact factor corresponding to the unit value of the structural similarity index preset in the painting information database, which represents the influence degree of the structural similarity index on the accuracy of genetic analysis of the painting. The fitting curve corresponding to the structural similarity index is obtained by fitting the relationship between the painting structure complexity, the number of historical painting elements and the structural similarity index. The real-time painting structure complexity and the number of painting elements are brought into the fitting curve corresponding to the structural similarity index to obtain the impact factor corresponding to the unit value of the structural similarity index in this example. The value range of this value in this example is 0 to 1.
[0102] Furthermore, the accuracy of genetic analysis of the painting is evaluated, and the specific evaluation process is: obtaining the line contour map of the painting and the line contour map of the reconstructed painting, and performing overlap comparison to obtain the overlapped line contour length, and performing ratio analysis with the total length of the line contour of the painting to obtain the overlap of the line contour of the painting; the above-mentioned obtaining the line contour map of the painting and the line contour map of the reconstructed painting refers to converting the painting and the reconstructed painting to the same size through the GNU image processing program, and then using the built-in find edge function of the GNU image processing program to extract the line contour map of the painting and the line contour map of the reconstructed painting; the above-mentioned overlapped line contour length refers to performing overlap comparison on the corresponding lines on the line contour map of the painting and the line contour map of the reconstructed painting, and calculating the overlapped line contour length of the two.
[0103] The structural similarity index between the reconstructed image of the painting and the painting is obtained, thereby comprehensively evaluating the accuracy of the genetic analysis of the painting; the above-mentioned obtaining the structural similarity index between the reconstructed image of the painting and the painting refers to converting the painting and the reconstructed image of the painting to the same size through the GNU image processing program, and using the built-in structural similarity index algorithm of the GNU image processing program to calculate the structural similarity index between the painting and the reconstructed image of the painting.
[0104] S3. Compare the accuracy of the gene analysis of the painting with the gene analysis accuracy threshold, and provide feedback on the data analysis of the painting, thereby completing the data analysis process of the painting.
[0105] Specifically, the data analysis process of the painting is completed, and the specific analysis process is: comparing the gene analysis accuracy of the painting with the gene analysis accuracy threshold. The above gene analysis accuracy threshold represents the minimum value of the reasonable range of gene analysis accuracy, which is determined by the painting analysis experts based on the historical painting analysis work experience.
[0106] If the accuracy of the gene analysis of the painting is greater than or equal to the gene analysis accuracy threshold, the data analysis of the painting is fed back, thereby completing the data analysis process of the painting; the above-mentioned feedback on the data analysis of the painting refers to that the accuracy of the gene analysis of the painting is greater than or equal to the gene analysis accuracy threshold. This result shows that in the process of data analysis of the painting, the effect of the data analysis of the painting has reached the expected effect, and then feedback is given that the accuracy of the painting data analysis method is high and has generalizability. If the accuracy of the gene analysis of the painting is less than the gene analysis accuracy threshold, the painting analysis system pops up the generated painting data analysis abnormality log, and the painting analysis expert deeply analyzes the problems existing in the painting data analysis process and formulates targeted improvement measures accordingly, including but not limited to adjusting the parameters in the painting analysis process, optimizing the selection and processing flow of the parameter source, introducing more advanced analysis technology or tools, etc. The above content is the entire content of the data analysis process of the painting; the painting analysis system is a system for users to interact with paintings, and the painting information database can be called.
[0107] In this exemplary embodiment, the variation table of the gene analysis accuracy of the above-mentioned colored paintings and its corresponding parameters is shown in Table 2:
[0108] Table 2 Genetic analysis accuracy of colored paintings and changes in their corresponding parameters
[0109] Line outline overlap / percentage Structural similarity index / percentage Gene analysis accuracy / percentage 76 80 215 80 85 223 68 60 191
[0110] In this example embodiment, the value of the influence factor corresponding to the unit value of line contour overlap is set to 0.25, and the value of the influence factor corresponding to the unit value of the structural similarity index is set to 0.4. Since the gene analysis accuracy formula contains an exponential function, the gene analysis accuracy is highly sensitive to the input parameters. At the same time, Table 2 quantifies the parameters to better interact with the user and clearly display the gene parameters of the painting.
[0111] In a specific embodiment, the present invention provides a data analysis construction method and system for colored paintings. By collecting texture information of colored paintings and analyzing recessive gene characteristic values and dominant gene characteristic values of colored paintings, the structural information of colored paintings can be objectively analyzed, and the physical characteristics and style characteristics of colored paintings can be grasped more accurately. The recessive gene element set and the dominant gene element set of colored paintings are respectively analyzed, which is helpful to fully and deeply understand the obvious element characteristics of colored paintings. At the same time, a rich data research platform is provided for subsequent research and comparison of colored paintings. The recessive gene element set and the dominant gene element set of colored paintings are integrated to evaluate the accuracy of gene analysis of colored paintings, compare the accuracy of gene analysis of colored paintings with the accurate threshold of gene analysis, and provide feedback on the data analysis of colored paintings, thereby completing the data analysis process of colored paintings, making the data analysis process of colored paintings interactive, and timely adjusting the color painting analysis parameters according to the color painting data analysis results to improve the accuracy and processing efficiency of color painting data analysis. At the same time, the feedback mechanism also provides users with intuitive information about the color painting analysis results.
[0112] Reference Figure 2 As shown, the second aspect of the present invention provides a system for constructing a data analysis method for color paintings, comprising: a color painting gene element set analysis module, a gene analysis accuracy assessment module, a color painting data analysis feedback module and a color painting information library.
[0113] The color painting gene element set analysis module is connected to the gene analysis accuracy evaluation module and the color painting information library respectively; the gene analysis accuracy evaluation module is connected to the color painting data analysis feedback module and the color painting information library respectively; the color painting data analysis feedback module is connected to the color painting information library.
[0114] The color painting gene element set analysis module is used to collect texture information of color paintings, analyze recessive gene characteristic values of color paintings, and obtain recessive gene element sets of color paintings by analysis, and analyze dominant gene characteristic values of color paintings by analysis to obtain dominant gene element sets of color paintings.
[0115] The gene analysis accuracy assessment module is used to integrate the recessive gene element set and the dominant gene element set of the colorful painting to generate a colorful painting reconstruction map, thereby assessing the gene analysis accuracy of the colorful painting.
[0116] The color painting data analysis feedback module is used to compare the gene analysis accuracy of the color painting with the gene analysis accuracy threshold, and to provide feedback on the data analysis of the color painting, thereby completing the data analysis process of the color painting.
[0117] The color painting information database is used to store the visual weight value corresponding to the color value of each pixel point, the influencing factor corresponding to the unit value of the maximum difference in visual weight, the influencing factor corresponding to the unit value of the fading index, the influencing factor corresponding to the unit value of the usage frequency, the texture complexity corresponding to each grayscale co-occurrence matrix entropy interval, the network connection complexity corresponding to each shape element connection number interval, the texture roughness corresponding to each average change rate interval, the influencing factor corresponding to the unit value of texture complexity, the influencing factor corresponding to the unit value of network connection complexity, the influencing factor corresponding to the unit value of texture roughness, the recessive gene element set corresponding to each recessive gene characteristic value interval, the dominant gene element set corresponding to each dominant gene characteristic value interval, the influencing factor corresponding to the unit value of line contour overlap and the influencing factor corresponding to the unit value of the structural similarity index.
[0118] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.
Claims
1. A data analysis and construction method for color painting, characterized in that: include: S1. Collecting texture information of the painting, analyzing the recessive gene characteristic value of the painting, parsing to obtain the recessive gene element set of the painting, analyzing the dominant gene characteristic value of the painting, parsing to obtain the dominant gene element set of the painting; S2. Comprehensively analyze the recessive gene element set and dominant gene element set of the colorful painting to generate a colorful painting reconstruction map, thereby evaluating the accuracy of the genetic analysis of the colorful painting; S3. Compare the accuracy of the gene analysis of the painting with the accurate threshold of the gene analysis, and provide feedback on the data analysis of the painting, thereby completing the data analysis process of the painting; The specific analysis process of analyzing the recessive gene characteristic value of the color painting is as follows: The painting is divided into sub-regions, and based on the texture information of the painting, the color value of each pixel in each sub-region is extracted, and the color value is matched with the visual weight value corresponding to the color value of each pixel preset in the painting information library, and the visual weight value of each sub-region is obtained by analysis, and the visual weight values are sorted from large to small, and the visual weight values ranked first and last are extracted, and the difference processing is performed to obtain the maximum difference of the visual weight of the painting; The average red, green and blue values of each painting sub-region are obtained, and the brightness of each painting sub-region is obtained through data processing. The brightness of each painting sub-region is subjected to difference processing with the brightness of the corresponding adjacent painting sub-region to obtain the brightness difference of each painting sub-region, and the de-blooming index of the painting is obtained through variance processing; The number of patterns used in the paintings is extracted from the texture information of the paintings, and the ratio is processed with the total number of patterns in the paintings to obtain the usage frequency of each pattern in the paintings. Based on this comprehensive analysis, the recessive gene characteristic value of the paintings is obtained. The specific analysis process of the dominant gene characteristic value of the color painting is as follows: Extracting the gray-level co-occurrence matrix of the painting from the texture information of the painting, obtaining the gray-level co-occurrence matrix entropy of the painting through data processing, and matching it with the texture complexity corresponding to each gray-level co-occurrence matrix entropy interval stored in the painting information database, thereby obtaining the texture complexity of the painting; Locate each shape element of the painting from the texture information of the painting, extract the correlation between each shape element of the painting and the corresponding adjacent shape element, connect each shape element with a correlation greater than a correlation threshold with the corresponding adjacent shape element, obtain the number of shape element connections of the painting, match it with the network connection complexity corresponding to the connection number interval of each shape element stored in the painting information library, and thus obtain the network connection complexity of the painting; The directional gradient histogram of the painting is obtained, and the center position points of the top of each square column in the directional gradient histogram are located and smoothly connected to obtain the directional gradient curve of the painting. The change rate function of the directional gradient curve is obtained through data processing. The change rate functions corresponding to the center position points of the tops of adjacent square columns are subjected to difference processing and mean analysis in turn to obtain the average change rate of the directional gradient curve of the painting. The average change rate is matched with the texture roughness corresponding to each average change rate interval stored in the painting information database, and finally the texture roughness of the painting is matched, and the dominant gene characteristic value of the painting is obtained by comprehensive analysis.
2. A method for data analysis and construction of color paintings according to claim 1, characterized in that: The analysis obtains the recessive gene element set of the color painting, and the specific analysis process is: The recessive gene characteristic value of the color painting is matched with the recessive gene element set corresponding to each recessive gene characteristic value interval stored in the color painting information database, thereby analyzing and obtaining the recessive gene element set of the color painting.
3. A data analysis and construction method for color painting according to claim 1, characterized in that: The analysis obtains the dominant gene element set of the color painting, and the specific analysis process is: The dominant gene characteristic value of the color painting is matched with the dominant gene element set corresponding to each dominant gene characteristic value interval stored in the color painting information database, thereby parsing and obtaining the dominant gene element set of the color painting.
4. A data analysis and construction method for color painting according to claim 1, characterized in that: The specific generation process of generating the color painting reconstruction image is as follows: The recessive gene element set of the painted color and the dominant gene element set of the painted color are input into a programming program, thereby generating a painted color reconstruction map.
5. The data analysis and construction method for color painting according to claim 1, characterized in that: The evaluation of the accuracy of the gene analysis of the painting is carried out in the following specific process: Obtaining the line contour map of the painting and the line contour map of the reconstructed painting, and performing overlap comparison to obtain the length of the overlapped line contour, and performing ratio analysis with the total length of the line contour of the painting to obtain the overlap degree of the line contour of the painting; At the same time, the structural similarity index between the reconstructed image of the painting and the painting is obtained to comprehensively evaluate the accuracy of the genetic analysis of the painting.
6. A method for analyzing and constructing data for color painting according to claim 5, characterized in that: The gene analysis accuracy of the painting is specifically expressed as: ; in, To determine the accuracy of the genetic analysis of the painting, is the overlap of the line contours of the painting, is the structural similarity index between the reconstructed image and the painting, is the influence factor corresponding to the unit value of line contour overlap preset in the painting information database, It is the impact factor corresponding to the unit value of the structural similarity index preset in the painting information database.
7. The data analysis and construction method for color painting according to claim 1, characterized in that: The data analysis process of the completed painting is as follows: The gene analysis accuracy of the painting is compared with the gene analysis accuracy threshold. If the gene analysis accuracy of the painting is greater than or equal to the gene analysis accuracy threshold, feedback is given to the data analysis of the painting, thereby completing the data analysis process of the painting.
8. A system using the data analysis and construction method for color paintings as claimed in any one of claims 1 to 7, characterized in that: include: The color painting gene element set parsing module is used to collect the texture information of the color painting, analyze the recessive gene characteristic value of the color painting, and obtain the recessive gene element set of the color painting through parsing, and analyze the dominant gene characteristic value of the color painting through parsing to obtain the dominant gene element set of the color painting; The gene analysis accuracy evaluation module is used to integrate the recessive gene element set and the dominant gene element set of the colorful painting to generate a colorful painting reconstruction map, thereby evaluating the accuracy of the gene analysis of the colorful painting; The color painting data analysis feedback module is used to compare the gene analysis accuracy of the color painting with the gene analysis accuracy threshold, and to provide feedback on the data analysis of the color painting, thereby completing the data analysis process of the color painting.
