Environmental protection art design element selection and optimization method based on data analysis

Abstract

The present application relates to the technical field of environmental art intelligent design, and particularly relates to an environmental art design element selection and optimization method based on data analysis. The method comprises the following steps: performing environmental element semantic analysis and environmental design element correlation integration processing according to environmental art design element data, to generate environmental design element correlation data; performing environmental element applicability analysis and environmental element combination selection processing according to the environmental design element correlation data, to generate combination environmental element selection data and perform environmental element implantation effect analysis, perform multi-dimensional collaborative environmental design element optimization on the combination environmental element selection data based on the environmental element implantation effect data, to generate multi-dimensional collaborative environmental design element optimization data, and feed the multi-dimensional collaborative environmental design element optimization data back to a user terminal to perform environmental art design element intelligent rendering work. The present application can realize environmental art design element selection and multi-dimensional collaborative optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of environmental art intelligent design technology, and particularly relates to an environmental art design element selection and optimization method based on data analysis. BACKGROUND

[0002] Under the background of global green and sustainable development, the concept of environmental protection is gradually permeating into design, art, urban construction and other fields. As an innovative application form integrating ecological consciousness and artistic expression, environmental art design is increasingly attracting attention. However, the existing environmental art design method mainly depends on the subjective experience of designers, lacks objective evaluation and optimization of environmental design elements, and is difficult to accurately reflect the propagation effect and public acceptance of the environmental protection concept. In addition, the combination mode and applicable scene of environmental art design elements have high complexity and difference in the in-depth mining and analysis of data such as browsing behavior, visual memory and emotional behavior feedback in actual implantation scene. Traditional environmental art design elements are difficult to realize the selection and multi-dimensional collaborative optimization of environmental art design elements. SUMMARY

[0003] Therefore, it is necessary to provide an environmental art design element selection and optimization method based on data analysis to solve at least one of the above technical problems.

[0004] To achieve the above purpose, an environmental art design element selection and optimization method based on data analysis comprises the following steps:

[0005] Step S1: Obtain environmental art design element data, and perform environmental element semantic analysis according to the environmental art design element data to generate environmental element semantic data;

[0006] Step S2: Perform environmental design element association integration processing on the environmental art design element data based on the environmental element semantic data to generate environmental design element association data;

[0007] Step S3: Perform environmental element applicability analysis according to the environmental design element association data to generate environmental element applicability data; and perform environmental element combination selection processing according to the environmental element applicability data to generate combined environmental element selection data;

[0008] Step S4: Perform environmental element implantation effect analysis according to the combined environmental element selection data to generate environmental element implantation effect data;

[0009] Step S5: Perform multi-dimensional collaborative environmental design element optimization of the combined environmental element selection data based on the environmental element implantation effect data to generate multi-dimensional collaborative environmental design element optimization data, and feed back the multi-dimensional collaborative environmental design element optimization data to the user terminal to perform environmental art design element intelligent rendering work.

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

[0011] Step S11: Obtain environmental art design element data;

[0012] Step S12: Perform environmental design element clustering and division according to the environmental art design element data, to generate divided environmental design element data;

[0013] Step S13: Perform environmental design element feature analysis according to the divided environmental design element data, to generate environmental design element feature data;

[0014] Step S14: Perform environmental element form structure and color extraction processing according to the environmental design element feature data, to generate environmental element form structure data and environmental element color data;

[0015] Step S15: Perform environmental element scene analysis according to the environmental element form structure data and the environmental element color data, to generate environmental element scene data;

[0016] Step S16: Perform environmental element semantic analysis on the environmental art design element data based on the environmental element scene data, to generate environmental element semantic data.

[0017] Further, step S2 comprises the following steps:

[0018] Step S21: Perform environmental element semantic field analysis according to the environmental element semantic data, to generate environmental element semantic field data;

[0019] Step S22: Perform environmental element semantic expansion analysis according to the environmental element semantic field data, to generate environmental element expanded semantic field data;

[0020] Step S23: Perform environmental design element association and integration processing on the environmental art design element data based on the environmental element expanded semantic field data, to generate environmental design element association data.

[0021] Further, step S3 comprises the following steps:

[0022] Step S31: Perform environmental element cultural scene analysis according to the environmental design element association data, to generate environmental element cultural scene data;

[0023] Step S32: Perform environmental element compatibility analysis according to the environmental element cultural scene data, to generate environmental element compatibility data;

[0024] Step S33: Perform heterogeneous environmental element detection on the environmental design element association data based on the environmental element compatibility data, to generate heterogeneous environmental element data;

[0025] Step S34: removing environmental protection ambiguous elements from the environmental protection design element association data based on the heterogeneous environmental protection element data, to generate ambiguous environmental protection element removal data;

[0026] Step S35: performing environmental protection element combination selection processing according to the ambiguous environmental protection element removal data, to generate combined environmental protection element selection data.

[0027] Further, step S4 includes the following steps:

[0028] Step S41: performing environmental protection element propagation path analysis according to the combined environmental protection element selection data, to generate environmental protection element propagation path data;

[0029] Step S42: performing environmental protection element browsing time quantification according to the environmental protection element propagation path data, to generate environmental protection element browsing time data;

[0030] Step S43: performing environmental protection element response analysis according to the environmental protection element browsing time data, to generate environmental protection element response data;

[0031] Step S44: performing environmental protection element implantation effect analysis according to the environmental protection element response data and the environmental protection element browsing time data, to generate environmental protection element implantation effect data.

[0032] Further, step S44 includes the following steps:

[0033] Step S441: performing environmental protection element visual feedback characteristic analysis according to the environmental protection element browsing time data, to generate environmental protection element visual feedback characteristic data;

[0034] Step S442: performing environmental protection element correction processing on the environmental protection element visual feedback characteristic data based on the environmental protection element response data, to generate corrected environmental protection element visual feedback characteristic data;

[0035] Step S443: performing environmental protection element behavior feedback feature analysis according to the corrected environmental protection element visual feedback characteristic data, to generate environmental protection element behavior feedback feature data;

[0036] Step S444: performing environmental protection element implantation effect analysis according to the environmental protection element behavior feedback feature data, to generate environmental protection element implantation effect data.

[0037] Further, step S5 includes the following steps:

[0038] Step S51: performing environmental protection element color three-element analysis according to the environmental protection element implantation effect data, to generate environmental protection element color three-element data;

[0039] Step S52: performing environmental protection element color visual perception timing analysis according to the environmental protection element color three-element data, to generate environmental protection element color visual perception timing data;

[0040] Step S53: Perform environmental element color sensory stimulation analysis based on the environmental element color three-element data and the environmental element color visual perception timing data, and generate environmental element color sensory stimulation data;

[0041] Step S54: Perform environmental element contour characteristic analysis according to the environmental element implantation effect data, and generate environmental element contour characteristic data;

[0042] Step S55: Perform environmental element form tension analysis based on the environmental element color three-element data and the environmental element contour characteristic data, and generate environmental element form tension data;

[0043] Step S56: Perform environmental element interest hue parameter analysis according to the environmental element color sensory stimulation data, and generate environmental element interest hue data;

[0044] Step S57: Perform multi-dimensional collaborative environmental design element optimization of combined environmental element selection data based on the environmental element interest hue data and the environmental element form tension data, generate multi-dimensional collaborative environmental design element optimization data, and feed back the multi-dimensional collaborative environmental design element optimization data to the user terminal to perform environmental art related element intelligent rendering work.

[0045] Further, step S57 includes the following steps:

[0046] Step S571: Perform environmental element focus area analysis according to the environmental element form tension data, and generate environmental element focus area data;

[0047] Step S572: Perform environmental element color sense collaborative optimization processing on the environmental element focus area data based on the environmental element interest hue data, and generate environmental element color sense collaborative data;

[0048] Step S573: Perform multi-dimensional collaborative environmental design element optimization of combined environmental element selection data based on the environmental element color sense collaborative data, and generate multi-dimensional collaborative environmental design element optimization data.

[0049] Further, step S572 includes the following steps:

[0050] Perform environmental element graphic boundary softening processing on the environmental art element focus area, and generate environmental element boundary softening data;

[0051] Perform environmental element area reconstruction according to the environmental art element boundary softening data, and generate environmental element area reconstruction data;

[0052] Perform environmental element color sense collaborative optimization processing on the environmental element area reconstruction data based on the environmental element interest hue data, and generate environmental element color sense collaborative data.

[0053] Further, step S572 includes the following steps:

[0054] According to the combination of environmental protection element selection data, the environmental protection element attribute layer is parsed, and environmental protection element attribute layer data is generated;

[0055] Based on the environmental protection element color feeling coordination data, the multi-dimensional coordinated environmental protection design element optimization is carried out on the environmental protection element attribute layer data, and multi-dimensional coordinated environmental protection design element layer data is generated;

[0056] According to the multi-dimensional coordinated environmental protection design element layer data, the environmental protection element layer is integrated, and multi-dimensional coordinated environmental protection design element optimization data is generated.

[0057] The beneficial effects of the application are:

[0058] The application provides an environmental protection artistic design element selection and optimization method based on data analysis. The method obtains environmental protection artistic design element data, performs environmental protection element semantic analysis based on the environmental protection artistic design element data, converts scattered natural landscapes, renewable material textures, and environmental protection symbols into structured data, and establishes a basic resource library of environmental protection artistic design elements. The environmental protection element semantic analysis is performed based on the basic resource library, natural language processing and image recognition technology are used to accurately interpret the deep meaning, cultural background and environmental protection concept of each design element. The method not only gives the environmental protection design elements clear environmental protection properties, but also constructs a mapping relationship between the elements and the environmental protection concept, avoids the problems of ambiguous expression and conflicting concepts in design, and generates semantic data to make the design elements searchable and classifiable, so that the elements that match the theme can be quickly located, the design material screening efficiency is greatly improved, clear semantic guidance and data support are provided for subsequent design work, and it is ensured that the design is always carried out around the environmental protection core value. The environmental protection design element association and integration processing is performed on the environmental protection artistic design element data based on the environmental protection element semantic data, the limitations of isolated use of elements in traditional design are solved, and the potential relationship between different design elements is mined through semantic association. The method not only enriches the expression level of the design elements, but also can convey more rich environmental protection information in the limited design space. In addition, the association and integration processing can find new application scenarios and combination modes of the design elements, can provide creative inspiration for designers, and avoid design homogenization. The generated environmental protection design element association data provides a modular and systematic element combination scheme for subsequent design, so that designers can quickly call the associated element set according to specific requirements, significantly improves the logicality and coherence of the design, and enhances the overall performance of the environmental protection artistic design. The environmental protection element applicability analysis is performed according to the environmental protection design element association data, the applicability of the associated environmental protection elements is evaluated by comprehensively considering factors such as design carrier, target audience and transmission channel. An evaluation model including physical properties, cultural context and transmission effect is established, and detailed applicability data reports are generated for each element and element combination. The environmental protection element combination selection processing is performed according to the environmental protection element applicability data, and the most suitable environmental protection element combination scheme for a specific scene is intelligently screened. The combination environmental protection element selection data provides a precise execution scheme for subsequent design, reduces the number of design element modifications, reduces the design cost, and improves the landing success rate of the design scheme. The environmental protection element implantation effect analysis is performed according to the combination environmental protection element selection data, the visual presentation and information transmission effect of the selected environmental protection element combination in the actual application scene are simulated and evaluated. Through quantitative analysis of indicators such as harmony of color matching, rationality of element layout and clarity of information transmission, detailed implantation effect data is generated. Potential problems such as visual fatigue caused by too many elements and information level confusion affecting understanding can be found in advance before the design scheme is implemented, and the element combination and design details are adjusted in time according to the effect data, so that the design scheme is optimized.The generated environmental protection element implantation effect data not only provides data basis for the improvement of the design scheme, but also provides objective standard for the quality control of the design result, ensures that the finally presented environmental protection art design work has both aesthetic and practicality, and realizes efficient propagation of environmental protection concept. Based on the multi-dimensional collaborative environmental protection design element optimization of the combination environmental protection element selection data combined with the environmental protection element implantation effect data, multi-dimensional collaborative environmental protection design element optimization data is generated, and the multi-dimensional collaborative environmental protection design element optimization data is fed back to the user terminal to perform environmental protection art design element intelligent rendering work, so as to achieve effect data oriented depth optimization of element combination from multiple dimensions such as color, composition, material and cultural connotation. By using machine learning algorithm, the influence of different optimization strategies on design effect is analyzed, and the optimal element adjustment scheme is found. Multi-dimensional collaborative optimization not only focuses on the visual level of design, but also fully considers the depth of environmental protection concept, so as to ensure that each design element can play the maximum value in the overall scheme, and the generated multi-dimensional collaborative environmental protection design element optimization data is directly connected with the intelligent rendering system of the user terminal, so as to realize automatic rendering and output of the design scheme. The design cycle is greatly shortened, the manual intervention is reduced, the precision and efficiency of design are improved, so as to realize environmental protection art design element selection and multi-dimensional collaborative optimization.

[0059] The environmental protection art design element selection and optimization method based on data analysis of the present application realizes objective evaluation and optimization of environmental protection design elements, so as to realize the propagation effect and public acceptance of accurately reflecting environmental protection concept, and solve the problems that the combination mode and applicable scene of environmental protection art design elements in actual implantation scene browsing behavior, visual memory, emotional behavior feedback and other data have high complexity and difference, and realize environmental protection art design element selection and multi-dimensional collaborative optimization. BRIEF DESCRIPTION OF DRAWINGS

[0060] Other features, objects and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments made with reference to the accompanying drawings:

[0061] Figure 1 The figure is a step flowchart of the environmental protection art design element selection and optimization method based on data analysis of the present application;

[0062] Figure 2 The figure is a detailed step flowchart of step S1 in the method; Figure 1 The figure is a detailed step flowchart of step S2 in the method.

[0063] Figure 3 The figure is a detailed step flowchart of step S2 in the method. Figure 1 The figure is a detailed step flowchart of step S2 in the method. DETAILED DESCRIPTION

[0064] The technical method of the present application will be described clearly and completely below in conjunction with the drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0065] In addition, the drawings are only schematic illustrations of the present application and are not necessarily drawn to scale. Identical reference numerals in the drawings represent identical or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities, which do not necessarily have to correspond to physically or logically independent entities. The functional entities can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0066] It should be understood that although the terms "first", "second" and the like can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of the exemplary embodiments, a first element can be referred to as a second element, and similarly a second element can be referred to as a first element. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0067] To achieve the above-mentioned purpose, please refer to Figures 1 to 3 The present application provides an environmental protection art design element selection and optimization method based on data analysis, which comprises the following steps:

[0068] Step S1: Obtain environmental protection art design element data, and perform environmental protection element semantic analysis according to the environmental protection art design element data to generate environmental protection element semantic data;

[0069] Step S2: Perform environmental protection design element correlation integration processing on the environmental protection art design element data based on the environmental protection element semantic data to generate environmental protection design element correlation data;

[0070] Step S3: Perform environmental protection element applicability analysis according to the environmental protection design element correlation data to generate environmental protection element applicability data; and perform environmental protection element combination selection processing according to the environmental protection element applicability data to generate combined environmental protection element selection data;

[0071] Step S4: Perform environmental protection element implantation effect analysis according to the combined environmental protection element selection data to generate environmental protection element implantation effect data;

[0072] Step S5: Multi-dimensional collaborative environmental protection design element optimization based on the combination of environmental protection element selection data based on environmental protection element implantation effect data, generating multi-dimensional collaborative environmental protection design element optimization data, and feeding back to the user terminal through the multi-dimensional collaborative environmental protection design element optimization data to perform environmental art design element intelligent rendering work.

[0073] In the embodiment of the present application, please refer to Figure 1 The environmental art design element selection and optimization method based on data analysis includes the following steps:

[0074] Step S1: Obtain environmental art design element data, and perform environmental element semantic analysis according to the environmental art design element data to generate environmental element semantic data;

[0075] In the embodiment of the present application, through the network crawler technology, high-definition picture, vector graphics and 3D model and other environmental art design element data are batched from open source gallery (such as Unsplash, Pixabay), environmental protection organization official website and government public data platform and other channels according to the set keywords ("renewable energy" "recycling" "biodiversity" and the like). The image data is subjected to object recognition by using Google CloudVisionAPI, the specific elements (such as solar panels, recycling signs and green plants and the like) in the picture are extracted, and the position and size are marked. The text description data is subjected to named entity recognition and part-of-speech tagging by using Python spaCy library, and the semantic keywords are analyzed. The Word2Vec model is used to train the word vectors of all elements, the element semantic vector space is constructed, the semantic similarity between elements is calculated, and finally the environmental element semantic data is generated.

[0076] Step S2: Based on the environmental element semantic data, the environmental art design element data is subjected to environmental design element association and integration processing to generate environmental design element association data;

[0077] In the embodiment of the present application, a Neo4j graph database is used to build a knowledge graph. Each environmental design element is taken as a node, and semantic association is taken as an edge. The weight of the edge is determined by the similarity calculated by Word2Vec. For example, the "waste tire" node is connected to the "recycling economy" and "handicraft" nodes in both directions. The association integration operation is performed through the Cypher query language of the graph database, and the element set with common semantic labels, such as "turtle", "plastic garbage", and "coral reef" under the "marine protection" label, is retrieved. The D3.js visualization tool is used to optimize the layout of the associated network. The size of the node represents the frequency of element occurrence, and the thickness of the edge represents the association strength. The environmental design element association data containing element combination suggestions and associated paths are generated, and the logical relationship between environmental elements is intuitively displayed.

[0078] Step S3: Perform environmental element applicability analysis according to the environmental design element association data to generate environmental element applicability data; perform environmental element combination selection processing according to the environmental element applicability data to generate combination environmental element selection data;

[0079] In the embodiment of the present application, a multi-dimensional applicability analysis model is established for the environmental design element association data. Blender three-dimensional modeling software is used to simulate the scene of the elements, for example, setting up standardized size templates such as outdoor billboards, mobile phone APP interfaces, and children's picture books, and importing the associated element combinations into the templates for visual testing. ImageJ image processing software is used to calculate quantitative indicators such as color contrast, edge definition, and information density of elements in different scenes, for example, the contrast of text elements in outdoor scenes should not be less than 4.5:1. Combined with the T-test statistical method, the visual effect scores of different element combinations in each scene are compared, and the scheme with the highest comprehensive score is selected. For example, in the children's picture book scene, the cartoonized "garbage classification" element combination has a higher score than the realistic style combination due to its bright colors and simple shapes, and finally forms combination environmental element selection data containing scene type, element list, and layout parameters.

[0080] Step S4: Perform environmental element implantation effect analysis according to the combination environmental element selection data to generate environmental element implantation effect data;

[0081] In the embodiment of the present application, a virtual reality simulation environment is built by means of Unreal Engine, and the combined environmental element selection data is imported into the engine. Parameters such as real light conditions, user visual angle moving path (such as 1.5 meters high, 0.5 meters per second uniform speed movement) are set to simulate the dynamic display effect of the elements in the actual scene. The eye tracking device is used to collect the visual trajectory data of the test user when browsing the design, and the OpenCV library of Python is used to analyze the eye movement data to identify the user's focus area and the length of stay. For example, if the average gaze time of the "endangered animal" element area is less than 2 seconds, it is determined that the information transmission is insufficient. At the same time, a semantic segmentation algorithm (such as DeepLabv3+) is used to divide the design picture into regions, calculate the visual attraction weight of each element area, and finally generate environmental element implantation effect data including element attention, information transmission efficiency and visual comfort.

[0082] Step S5: Multi-dimensional collaborative environmental design element optimization based on the combination of environmental element implantation effect data and combined environmental element selection data, generating multi-dimensional collaborative environmental design element optimization data, and feeding back to the user terminal through the multi-dimensional collaborative environmental design element optimization data to perform environmental art design element intelligent rendering job.

[0083] In the embodiment of the present application, based on the environmental element implantation effect data, a multi-dimensional collaborative optimization model is established using the optimization toolbox of MATLAB. The color saturation, element spacing, text size and other design parameters are used as optimization variables, the visual attraction and information transmission efficiency are used as objective functions, and the constraint conditions are set. Genetic algorithm is used for iterative optimization, 100 groups of design parameter combinations are generated each time, and the real-time rendering comparison effect is realized through Unreal Engine. For example, the first iteration found that the recognition rate of the "recyclable logo" element dropped sharply when it was reduced to 15 pixels, and the algorithm automatically adjusted the lower limit of its size to 20 pixels. After 200 iterations, the parameter combination with the highest comprehensive score is selected, and the multi-dimensional collaborative environmental design element optimization data including element coordinates, color values and scaling ratio are generated. The optimization data is transmitted to the user terminal through HTTP protocol, and the GPU accelerated rendering engine (such as Vulkan) of the terminal device is called to automatically generate the final environmental art design work, completing the whole process closed loop from data driven to intelligent rendering.

[0084] Further, step S1 includes the following steps:

[0085] Step S11: obtaining environmental art design element data;

[0086] Step S12: clustering and dividing environmental design elements according to the environmental art design element data to generate divided environmental design element data;

[0087] Step S13: Perform environmental design element characteristic analysis based on the divided environmental design element data to generate environmental design element characteristic data;

[0088] Step S14: Extract the morphology, structure and color of environmental protection elements based on the characteristic data of environmental protection design elements to generate morphology and structure data and color data of environmental protection elements.

[0089] Step S15: Analyze environmental protection scenarios based on the morphological and structural data and color data of environmental protection elements to generate environmental protection scenario data;

[0090] Step S16: Based on the environmental protection element scenario data, perform environmental protection element semantic analysis on the environmental protection art design element data to generate environmental protection element semantic data.

[0091] As an embodiment of the present invention, reference is made to... Figure 2 As shown, Figure 1 A detailed flowchart of step S1 is shown below. In this embodiment, step S1 includes the following steps:

[0092] Step S11: Obtain environmental art design element data;

[0093] In this embodiment of the invention, the Apache Nutch web crawler framework is used to set seed URLs related to environmental art and design elements, including multiple initial links such as official websites of ecological protection organizations, art and design resource platforms, and environmental forums. Crawler parameters are configured to crawl once per hour, with a crawl depth of 3 levels each time, and the crawled file types are limited to JPEG, PNG, SVG, and PDF formats. Regular expression filtering rules are used to select resources whose filenames or descriptions contain keywords such as "environmentally friendly," "regenerative," and "sustainable." The crawled data is then processed by the Apache Tika parsing tool to extract metadata, including file size, creation time, and color mode information, and stored in the Hadoop Distributed File System (HDFS), forming a raw dataset of environmental art and design elements containing massive amounts of records.

[0094] Step S12: Based on the environmental art design element data, perform environmental design element clustering and classification to generate classified environmental design element data;

[0095] In this embodiment of the invention, the DBSCAN density clustering algorithm is used for processing. Image data is converted into vector form, and SIFT features are extracted using the OpenCV library. Each feature vector contains a multidimensional descriptor. For example, the clustering radius is set to ∈ = 0.5, and the minimum number of samples MinPts = 5. Clustering is then performed on all image vectors. For a point p in the vector space, if the number of samples in its ∈-neighborhood |N ∈(p) | ≥ MinPts, p is a core point, if point q is in the neighborhood of core point p, q is subordinate to p, if point r is neither a core point nor subordinate to a point, it is a noise point. The algorithm aggregates images with similar visual features into the same cluster, and measures the similarity between feature vectors by calculating the Euclidean distance. For example, all images containing tree elements are clustered into the "plant ecology" category, and images containing windmills and solar panels are clustered into the "clean energy" category. Finally, the division of environmental design element data containing cluster labels and image ID lists in the cluster is generated, forming several different environmental design element cluster clusters, clearly presenting the visual similarity grouping of elements.

[0096] Step S13: Perform environmental design element feature analysis according to the division of environmental design element data, and generate environmental design element feature data;

[0097] In the embodiment of the present application, Scikit-learn machine learning library is used for feature analysis. For image data, HOG (Histogram of Oriented Gradients) algorithm is used to extract shape features, for example, the gradient direction distribution of 9 cell units and 4 blocks of each image is calculated to generate a multi-dimensional shape feature vector. The K-means algorithm is used to quantize color features, and the image color space is converted from RGB to Lab, clustered into a variety of dominant colors, and the proportion and average color value of each color tone are recorded. For vector graphics and text description data, NLTK library is used for text segmentation and part-of-speech tagging, and key words such as nouns and verbs are extracted as semantic features. The final generated environmental design element feature data includes shape description vector, color distribution matrix and semantic keyword list of each element, for example, the feature data of a picture depicting garbage classification includes the geometric shape parameters of the garbage can, the color proportion of the contrast, and the semantic labels such as "classification" and "recycling".

[0098] Step S14: Perform environmental element shape structure and color extraction processing according to the environmental design element feature data, and generate environmental element shape structure data and environmental element color data;

[0099] In the embodiment of the present application, the ImageMagick image processing tool is used for morphological structure and color extraction of the image. For morphological structure, the edge detection algorithm (Canny operator) is used to identify the contour lines in the image, and the line coordinates, curvature, etc. are used to generate the morphological structure data of the environmental protection element. Taking an illustration depicting endangered birds as an example, the contour curve of the bird and the texture direction of the wing feathers are extracted as structural information. For color data, the red, green and blue channels are separated from the RGB color space of the image, the pixel value distribution of each channel is counted, the average color value, standard deviation and other statistical quantities are calculated, and the environmental protection element color data including the main color RGB value, color contrast and other indicators are formed. For example, it is determined that the illustration is dominated by blue-green color and the color contrast is 3.2:1.

[0100] Step S15: performing environmental protection element scene analysis according to the environmental protection element morphological structure data and the environmental protection element color data to generate environmental protection element scene data;

[0101] In the embodiment of the present application, the generated environmental protection element morphological structure data and environmental protection element color data are combined with the design scene parameters for scene analysis. Standardized scene templates are constructed, including three typical scenes of a poster (A3 size, 300 dpi), a webpage banner (1920x500 pixels) and product packaging (200x150x50 mm). The morphological structure data is converted into a 3D model using the Blender three-dimensional modeling software, and the material texture is given according to the color data, and the layout simulation is performed in different scene templates. The actual lighting conditions are simulated by ray tracing rendering technology, and the visibility and visual appeal of the elements in each scene are calculated. For example, in the product packaging scene, the display effect of the environmental protection element pattern under different observation angles is analyzed, the best display proportion and position of the element on the front and side of the packaging are recorded, and finally the environmental protection element scene data including the scene type, element layout parameters and visual effect evaluation is generated.

[0102] Step S16: performing environmental protection element semantic analysis on the environmental protection art design element data based on the environmental protection element scene data to generate environmental protection element semantic data.

[0103] In the embodiment of the present application, semantic analysis is performed using the Word2Vec word vector model. The text information such as scene description and element feature keywords in the scene data is converted into a word vector, and the correlation degree with the vocabulary in the pre-trained environmental protection field word vector library is calculated through cosine similarity. For example, when an element is presented in the form of "degradable material" in the product packaging scene, and the color adopts a natural brown tone, the system assigns it semantic labels such as "green packaging" and "ecologically friendly" through word vector matching. At the same time, combined with knowledge graph technology, the form, color and scene information of the element are associated with the concepts in the environmental protection knowledge base to construct an element semantic network. The finally generated environmental protection element semantic data includes element ID, semantic label set and semantic association graph, which clearly defines the deep meaning and application context of each environmental protection design element.

[0104] Further, step S2 includes the following steps:

[0105] Step S21: Perform environmental protection element semantic field analysis according to the environmental protection element semantic data to generate environmental protection element semantic field data;

[0106] Step S22: Perform environmental protection element semantic expansion analysis according to the environmental protection element semantic field data to generate environmental protection element expanded semantic field data;

[0107] Step S23: Perform environmental protection design element association integration processing on the environmental protection art design element data based on the environmental protection element expanded semantic field data to generate environmental protection design element association data.

[0108] As an embodiment of the present application, referring to Figure 3 , it is a detailed step flowchart of step S2 in the embodiment, and step S2 in the embodiment includes the following steps: Figure 1

[0109] Step S21: Perform environmental protection element semantic field analysis according to the environmental protection element semantic data to generate environmental protection element semantic field data;

[0110] ​In the embodiment of the present application, the Stanford CoreNLP natural language processing toolkit is used to process the semantic label set in the environmental protection element semantic data. First, each semantic label is disassembled into independent words by a word segmenter, for example, "green packaging" is split into "green" and "packaging", and then the part-of-speech tagger is used to identify the part of speech of the words, such as "green" is tagged as an adjective and "packaging" is tagged as a noun. Then, the named entity recognition (NER) technology is used to distinguish between ordinary words and entity words in the environmental protection field, and professional terms such as "degradable materials" and "ecological protection" are marked as environmental protection entities. Then, through dependency syntax analysis, the grammatical relationship between words is determined, such as in "marine pollution control", "control" is the core verb and "marine pollution" is the object. Finally, these parsed words, parts of speech, entity categories, and grammatical relationships are organized into environmental protection element semantic field data in a unified format,

[0111] Step S22: Perform environmental protection element semantic expansion analysis based on the environmental protection element semantic field data to generate environmental protection element expanded semantic field data.

[0112] In the embodiment of the present application, the GloVe word vector model is used to convert each word in the environmental protection element semantic field data into a multi-dimensional word vector. Based on the pre-trained environmental protection field corpus, the cosine similarity is calculated to find the expanded words with similar semantics to each word. For example, for "energy saving and emission reduction", the model calculates the words with a similarity higher than 0.8, such as "low-carbon action" and "energy saving", as expanded semantics. At the same time, the ontology knowledge base in the environmental protection field, such as "Environmental Science Ontology", is used to expand the concept hierarchy based on the entity words in the semantic field. When encountering "endangered species", the sub-concepts such as "Northeast Tiger" and "Juhuang" are extracted from the ontology library for semantic extension. In addition, combined with environmental protection policy documents and research literature, the implicit semantic association between words is mined, such as the internal relationship between "garbage classification" and "circular economy" in the policy promotion field, and this association is included in the expanded semantics. Finally, the environmental protection element expanded semantic field data is formed, which includes the original semantic field, direct expanded words, concept hierarchy extension, and semantic association.

[0113] Step S23: Based on the environmental protection element expanded semantic field data, the environmental protection design element association data is generated by performing environmental protection design element association integration processing on the environmental protection art design element data.

[0114] In the embodiment of the present application, Neo4j graph database is used, each element in the environmental protection art design element data is taken as a node, and the semantic relationship in the extended semantic field data is taken as an edge to construct an environmental protection design element semantic association network. For each element node, according to the corresponding extended semantic field, element nodes with the same or related semantics are connected through an edge, and the edge is given a weight according to the semantic similarity. For example, the “wind turbine” element node and the “solar panel” element node are connected through an edge with a weight of 0.85 because they belong to the “clean energy” semantic category. The “plastic recycling logo” node and the “circular economy poster” node are connected through an edge with a weight of 0.7 based on the semantic association relationship. The indirect association path between elements is mined by using the path query algorithm of the graph database, such as the “degradable material packaging bag” node which is associated with the “ecological agriculture” node through intermediate semantic nodes such as “environmentally friendly packaging” and “green product”. Finally, the association network is processed by a community discovery algorithm (such as the Louvain algorithm), the community is divided by continuously optimizing the modularity (Modularity), and the modularity calculation formula is: Where m is the total number of edges in the graph, A ij represents the number of edges between nodes i and j, k i and k j are the degrees of nodes i and j, δ(c I ,c j ) is 1 when nodes i and j belong to the same community, otherwise it is 0. Through the iterative calculation of the Louvain algorithm, closely related elements are divided into different communities to generate environmental protection design element association data containing element node information, edge connection relationship and community classification, and the semantic association system between environmental protection design elements is intuitively presented.

[0115] Further, step S3 includes the following steps:

[0116] Step S31: environmental protection element culture scene analysis is performed according to the environmental protection design element association data to generate environmental protection element culture scene data;

[0117] In the embodiment of the present application, the global cultural database (such as Hofstede cultural dimension database, UNESCO cultural heritage database) is cross-compared with the environmental protection element knowledge base. For each element node in the environmental protection design element association data, the semantic label and associated information are extracted. Taking the "bamboo tableware" element as an example, by querying the cultural database, it is found that it symbolizes "natural harmony" in eastern culture and is regarded as a "representative of sustainable lifestyle" in some western regions. By using text mining technology, the cultural description related to the element is extracted from cultural classics and folklore research reports to build the cultural scene archives of the element. For the "windmill" element, combined with the historical data of Dutch windmill culture, it is determined that it represents "energy utilization tradition" and "regional characteristics" in the Dutch cultural scene. Finally, the information such as cultural region, cultural implication, and cultural symbol carrier corresponding to each element is integrated to generate environmental protection element cultural scene data containing element ID, cultural scene description, and cultural association strength score.

[0118] Step S32: performing environmental protection element compatibility analysis according to the environmental protection element cultural scene data to generate environmental protection element compatibility data;

[0119] In the embodiment of the present application, the cultural scene data of different elements are compared with each other. For the combination of "dragon" element and "environmental protection theme", in eastern culture, "dragon" symbolizes power and auspiciousness, but there may be understanding deviation in some western cultures. By calculating the semantic similarity and conflict keywords (such as cultural taboo words) of the cultural scene description, the compatibility between elements is evaluated. If there are semantic conflict keywords such as "artificial intervention" and "natural purity" in the cultural scene of the "nuclear symbol" element and the "natural original ecology" theme element, it is determined that the compatibility between them is low. At the same time, the compatibility rule base annotated by cultural experts is introduced to correct the algorithm result. Finally, the environmental protection element compatibility data containing element pair ID, compatibility score, conflict point description, and correction suggestion are generated, for example, the compatibility score of the combination of "cherry blossom pattern" and "spring environmental protection activity" element is 0.9, and there is no obvious conflict point.

[0120] Step S33: detecting heterogeneous environmental protection elements in the environmental protection design element association data based on the environmental protection element compatibility data to generate heterogeneous environmental protection element data;

[0121] In the embodiment of the present application, the compatibility data of environmental elements is processed by using cluster analysis and anomaly detection algorithm. All element pairs are K-means clustered according to compatibility scores, and three categories of high compatibility, medium compatibility and low compatibility are divided. For the elements in the low compatibility category, the semantic characteristics and cultural scene differences are further analyzed to identify heterogeneous environmental elements. For example, the "industrial chimney" element and the "ecological protection" theme element are associated in the concept of environmental protection, but the cultural scene and visual style of the two are significantly different, and are judged as heterogeneous elements. At the same time, a heterogeneous element judgment threshold is established, and when the semantic difference degree of the cultural scene between elements exceeds 0.7, it is included in the heterogeneous environmental element data. Finally, the heterogeneous environmental element data containing the ID of the heterogeneous element, the difference characteristic description and the judgment basis is generated, which provides the basis for subsequent screening.

[0122] Step S34: removing environmental ambiguity elements from the environmental design element association data based on the heterogeneous environmental element data to generate ambiguity environmental element removal data;

[0123] In the embodiment of the present application, the BERT pre-training language model is used for deep semantic analysis of the heterogeneous environmental element data. The text description (such as element name, associated semantic label) of each environmental element is input into the BERT model to obtain the word vector representation of each layer in the 12-layer Transformer encoder, and the average vector of the last layer is taken as the semantic feature vector of the element to construct an ambiguity element judgment model. Combined with semantic ambiguity calculation and cultural scene ambiguity analysis, the heterogeneous environmental elements are processed. For example, using a decision tree algorithm, according to the semantic complexity of the element, the number of cultural scene conflicts, the compatibility score and other indicators, and transmitting the indicators to the ambiguity element judgment model, it is determined that the "plastic" element has ambiguity in the "recyclable" and "white pollution" semantic scenes. By analyzing the semantic tendency of the element in different cultural scenes and the compatibility of associated elements, it is determined that the element is an ambiguity element, and it is automatically determined whether to remove. If an element has semantic conflict in more than three cultural scenes and the compatibility score is less than 0.4, it is removed from the environmental design element association data. Finally, the ambiguity environmental element removal data containing the reserved element ID, the removed element ID and the removal reason is generated to ensure that the semantic of the design element is clear and the culture is adapted.

[0124] Step S35: processing the environmental element combination selection according to the ambiguity environmental element removal data to generate combination environmental element selection data.

[0125] In the embodiment of the present application, a multi-objective optimization algorithm is used to select the combined environmental design elements from the processed environmental design element association data, with the compatibility score C between elements, the cultural scene fitting degree S, and the visual style consistency V as the optimization objectives. The size of the element combination is limited (e.g., no more than 5 elements per group), and the genetic algorithm is used for iterative calculation. The initial population contains several groups of random element combinations. The fitness value is calculated according to the optimization objective function in each iteration, and the combinations with high fitness are selected for crossover and mutation operations. The compatibility score between elements is calculated by accumulating the weights of the edges in the association network, and the formula is where n represents the number of elements in the current element combination, and w ij represents the weight value of the environmental elements i and j in the element association network. The cultural scene fitting degree is calculated by the semantic similarity of the elements and the scene keywords, and the formula is where sim(e i ,s k ) is the semantic similarity of the element e i and the scene keyword s k , i represents the ith environmental design element, and k represents the kth keyword in the cultural scene. The visual style consistency is calculated by the matching degree of color, shape, and other features, and the formula is V= where f(v i ,v j ) is the feature matching function, and v i and v j are the visual style feature vectors of the environmental elements i and j, respectively, which can be composed of color histograms, contour shapes, texture directions, and other features. For example, when designing a poster on the theme of "marine protection", the algorithm preferentially selects element combinations with high compatibility and cultural scene fitting, such as "turtle", "coral reef", and "blue gradient". After several iterations, the element combinations with the highest fitness are selected, and the combination environmental element selection data containing the combination ID, element list, and optimization index score are generated, providing accurate element combination solutions for environmental art design.

[0126] Further, step S4 includes the following steps:

[0127] Step S41: Perform environmental element propagation path analysis according to the combined environmental element selection data to generate environmental element propagation path data;

[0128] In this embodiment of the invention, a database is established covering mainstream communication channels such as social media platforms (WeChat, Weibo, Facebook, Instagram), outdoor advertising (bus stop signs, LED screens), and offline event materials (flyers, posters). For each combination of environmental protection elements selected in step S35, exposure data under different channels is collected using network traffic monitoring tools (such as Google Analytics) and outdoor advertising monitoring equipment (people counters, cameras). For example, an advertisement with a combination of "garbage sorting" themed elements is placed on an LED screen in a commercial area of ​​a city. Cameras are used to record the number of times and locations of people stopping, and the exposure per unit time is calculated by combining the people counter data. An element combination poster is published on the Weibo platform, and interaction data such as the number of reposts, comments, and likes are obtained through the API interface. At the same time, user profile data of each channel is analyzed, including age distribution, geographical distribution, and interest preferences. The exposure, interaction, and user profile information of the element combination are integrated to generate environmental protection element dissemination path data that includes the type of communication channel, exposure data, user profile, and channel characteristic score.

[0129] Step S42: Quantify the browsing time of environmental protection elements based on the data on the dissemination paths of environmental protection elements, and generate browsing time data of environmental protection elements;

[0130] In this embodiment of the invention, on social media platforms, front-end tracking technology is used to record the time a user spends from entering a page to leaving it. For outdoor advertising, eye-tracking devices installed near the advertising space collect the time users spend looking at the advertising image. Timing begins when the user's gaze lingers on the advertising area for more than 0.5 seconds and stops when the gaze leaves the advertising area for more than 1 second. For example, element combination images are inserted into environmental protection articles pushed by WeChat official accounts, and tracking code records the duration of user stay in the image area. Eye trackers are installed in front of bus stop advertisements to monitor the time passengers spend viewing the advertisements. The collected raw time data is cleaned to remove outliers (such as data with a stay time of less than 1 second or more than 60 seconds), and then grouped and statistically analyzed according to the dissemination channel and element combination category. The average, median, standard deviation, and other statistical measures of each group are calculated to generate environmental element browsing time data that includes dissemination channel, element combination ID, average browsing time, and time distribution interval.

[0131] Step S43: Analyze the response of environmental elements based on the browsing time data of environmental elements, and generate environmental element response data;

[0132] In the embodiment of the present application, the user behavior is divided into multiple positive response behaviors such as active sharing, comment message, click link and secondary access, and three negative response behaviors such as quickly closing the page and no-stopping browsing. On the social media platform, the response behavior is identified by analyzing the user operation log, such as detecting that the user clicks the "forward" button to record the active sharing behavior. For outdoor advertising, the response is judged in combination with the two-dimensional code scanning data and offline activity participation data. If the user scans the two-dimensional code on the advertisement to participate in the environmental knowledge quiz activity, it is recorded as a positive response. The statistical quantity in the environmental protection element browsing time data and the user portrait data are used as features to train the response behavior prediction model by using the classification algorithm (such as support vector machine SVM) in machine learning. The new browsing time data is input into the model to predict the probability of different response behaviors of the user. Finally, the environmental protection element response data including element combination ID, propagation channel, positive response probability, negative response probability and response behavior type distribution is generated.

[0133] Step S44: Perform environmental protection element implantation effect analysis according to the environmental protection element response data and the environmental protection element browsing time data, and generate environmental protection element implantation effect data.

[0134] In the embodiment of the present application, the propagation influence (exposure amount, forwarding amount) of the environmental protection element, the user attention (browsing time) and the response enthusiasm (positive response probability) are taken as the core evaluation indexes. For example, each index is assigned a weight (the propagation influence accounts for 30%, the user attention accounts for 40%, and the response enthusiasm accounts for 30%), and the implantation effect score of each element combination is calculated by the weighted sum formula Score = w1 x I + w2 x A + w3 x R. Wherein the weight distribution of each index corresponds to w1 = 0.3, w2 = 0.4, and w3 = 0.3, and I, A and R are the standardized scores of the propagation influence, the user attention and the response enthusiasm respectively. At the same time, the Pearson correlation coefficient The correlation between each X element index and Y element index is analyzed, wherein cov(X, Y) is the covariance of variable X and variable Y, which is used to measure the overall error of two variables, and σ X and σ Y are the standard deviations of variable X and variable Y respectively, reflecting the dispersion degree of data. If it is found that the browsing time is longer but the positive response probability is lower, it means that the element may have the problem of unclear information transmission. Finally, the environmental protection element implantation effect data including element combination ID, propagation channel, index score, comprehensive score and problem diagnosis is generated.

[0135] Further, step S44 includes the following steps:

[0136] Step S441: Perform environmental protection element visual feedback characteristic analysis according to the environmental protection element browsing time data, and generate environmental protection element visual feedback characteristic data;

[0137] In the embodiment of the present application, the eye movement trajectory data of the user when browsing the environmental protection elements is processed by using image processing technology and eye movement data analysis tools. The fixation point coordinates recorded by the eye tracker are used to generate a heat map on the element image, and the color depth represents the density of the fixation points, so as to analyze the visual focus distribution of the user. For example, when analyzing the poster of the “marine protection” theme element combination, the heat map shows that the user has the highest fixation point density on the marine pattern involving blue in the picture, and less fixation on the bottom text description area. At the same time, the smoothness and the number of jumps of the eye movement trajectory are calculated to evaluate the fluency of the user's visual browsing. If the eye movement trajectory frequently jumps, it means that the element layout may have visual interference. The heat map data, eye movement index data, visual focus area coordinates and other information are integrated to generate environmental protection element visual feedback characteristic data containing element combination ID, heat map data, eye movement index, and visual focus distribution.

[0138] Step S442: environmental protection element correction processing is performed on the environmental protection element visual feedback characteristic data based on the environmental protection element response data, to generate corrected environmental protection element visual feedback characteristic data;

[0139] In the embodiment of the present application, the visual feedback characteristic data is associated with the response data for analysis. If the visual focus of a certain element combination is concentrated on the decorative pattern, but the positive response probability is low, it means that the user is visually attracted but does not produce effective behavioral response, at this time, the weight distribution in the visual feedback characteristic data is adjusted to reduce the visual importance score of the decorative pattern. Using the decision tree algorithm, taking the response behavior type as the classification target and taking each index in the visual feedback characteristic data as the feature, a correction rule model is constructed. For example, when the user quickly closes the page (negative response) and the number of jumps of the eye movement trajectory exceeds the threshold value, the model determines that the element has a visual information overload problem, and automatically corrects the visual focus distribution data to highlight the core environmental protection information area. Finally, the corrected environmental protection element visual feedback characteristic data containing the corrected heat map data, the corrected eye movement index, and the corrected visual focus distribution is generated.

[0140] Step S443: environmental protection element behavior feedback characteristic analysis is performed according to the corrected environmental protection element visual feedback characteristic data, to generate environmental protection element behavior feedback characteristic data;

[0141] In the embodiment of the present application, the modified visual feedback characteristic data is fused and analyzed with user behavior data (such as click, slide, dwell, etc. Operation). Through the time sequence analysis method, the time correlation between the user visual focus change and the behavior operation is studied, for example, it is found that the probability of user clicking operation after staring at the "environmental protection action registration" button for 3 seconds is relatively high. Using clustering algorithm, the user behavior mode is classified, and the users with similar visual focus moving track and behavior operation sequence are divided into the same class. For example, the users who first stare at the element title, then browse the main pattern, and finally click the link are classified as "information-oriented" behavior mode. Finally, the environmental element behavior feedback characteristic data containing element combination ID, behavior mode classification, behavior association rule, and behavior characteristic index is generated.

[0142] Step S444: Perform environmental element implantation effect analysis according to the environmental element behavior feedback characteristic data, and generate environmental element implantation effect data.

[0143] In the embodiment of the present application, new indexes such as user behavior conversion rate (such as the number of users clicking the link / the total number of users browsing) and behavior path depth (the number of operation steps completed by the user) are introduced, and the original indexes (propagation influence, user attention, and response enthusiasm) are used together to form the evaluation index set. For example, the analytic hierarchy process (AHP) is used to determine the weight of each index, the proportion of behavior conversion rate, and the proportion of behavior path depth, and the weight of the remaining indexes is adjusted accordingly. The comprehensive score of each element combination is calculated, and at the same time, according to the behavior association rules and modes in the behavior feedback characteristic data, the advantages and disadvantages of the element combination in guiding user behavior are analyzed, and the environmental element implantation effect data containing element combination ID, index score, comprehensive score, and behavior optimization suggestion is generated.

[0144] Further, step S5 includes the following steps:

[0145] Step S51: Perform environmental element color three-element analysis according to the environmental element implantation effect data, and generate environmental element color three-element data;

[0146] In the embodiment of the present application, the element images involved in the environmental protection element implantation effect data are color-separated by color extraction tools (such as the convert command of ImageMagick) according to the Adobe RGB color space standard. The color information of each element image is decomposed into hue (H), saturation (S) and value (V) to form a color three-element matrix. For example, for an illustration with the theme of “forest protection”, the hue value of the main color green is 120 degrees, the saturation is 80%, and the value is 60%. The color three-element data of all element images are standardized, the hue value is mapped to the interval of 0-360, the saturation and value are mapped to the interval of 0-100, and finally the environmental protection element color three-element data containing element ID, hue value, saturation value, value and color distribution proportion are generated.

[0147] Step S52: environmental protection element color visual perception timing analysis is performed according to the environmental protection element color three-element data to generate environmental protection element color visual perception timing data;

[0148] In the embodiment of the present application, the eye movement trajectory of the user when browsing the environmental protection element image is recorded by means of an eye tracking device (such as Tobii Pro Glasses3), and the sampling frequency is set to 120 Hz. The eye movement data and the color three-element data are time-stamped matched, and the color three-element change of the user's gaze area in each time slice is analyzed at a time interval of 100 milliseconds. The eye movement coordinates are mapped to the pixel coordinates of the image by means of a bilinear interpolation algorithm, and the formula is: V(x,y)=(1-u)(1-v)V(I,j)+u(1-v)V(i+1,j)+(1-u)vV(I,j+1)+uvV(i+1,j+1), wherein (x,y) is the continuous coordinates of the user's gaze point in the image at the current time, (I,j) is the left upper corner coordinates of the pixel grid where the eye movement coordinates (x,y) are located, u=x-i and v=y-j represent the offset of the gaze point (x,y) relative to the left upper corner of the pixel grid, V(x,y) represents the color three-element value of the gaze point (x,y) obtained by interpolation, V(I,j), (i+1,j), V(I,j+1) and (i+1,j+1) correspond to the color three-element values of the four adjacent pixel points, and V is the color three-element value (H hue, S saturation, V value). For example, within the first 300 milliseconds of the user browsing the “marine ecological protection” poster, the eye movement data shows that the user first gazes at the blue ocean area (hue 210 degrees, saturation 70%, value 50%), and then moves to the orange life buoy area (hue 35 degrees, saturation 90%, value 70%). The color three-element difference value of the adjacent time slice is calculated, and the color three-element difference value of the adjacent time window is calculated: ΔH=∣H t+1 -H t|mod 180, AS = |S t+1 - t |, AV = |V t+1 - t |, wherein H is a hue value, S is a saturation value, V is a lightness value, H t represents the hue value in the t-th time slice, AH is the change amplitude of the hue in adjacent time slices, the color wheel symmetry problem is handled by using the modulo operation of 180 (avoiding the error caused by 360°→0° jump), and the change intensity of the user's gaze area in the hue dimension is reflected, S t represents the saturation value in the t-th time slice, AS represents the change of the perceived color saturation of the user in adjacent gaze points, V t represents the lightness value in the t-th time slice, AV represents the jump degree of the user's visual perception area in the lightness level. The color change curve is constructed, and the environmental element color visual perception time sequence data containing the time stamp, the gaze area coordinates, the color three-element value, and the color change rate are generated.

[0149] Step S53: performing environmental element color sensory stimulation analysis based on the environmental color three-element data and the environmental element color visual perception time sequence data, to generate environmental element color sensory stimulation data;

[0150] In the embodiment of the present application, color contrast intensity, color change frequency, and color emotional tendency are taken as core evaluation indexes. The color contrast intensity is calculated by using the color three-element data, and the difference between the color three-element values of different areas is calculated by using the Euclidean distance formula: wherein H is a hue value, S is a saturation value, V is a lightness value, H1 and H2 represent the hue values of two contrast areas or elements, S1 and S2 represent the saturation degrees of the corresponding areas, and V1 and V2 represent the lightness values, which are used to reflect the lightness and darkness of the color. For example, the color Euclidean distance between the green vegetation area and the gray industrial building area in the poster is 85. The color change frequency is counted according to the time sequence data, for example, if the color changes more than 5 times within 10 seconds of browsing time, it is determined as high frequency change. The emotional tendency corresponding to each color is determined in combination with the color psychology knowledge base (such as the Pantone color emotional guide), for example, red represents “warning” and blue represents “calm”. Each index is assigned a weight (the contrast intensity accounts for 40%, the change frequency accounts for 30%, and the emotional tendency accounts for 30%), the color sensory stimulation score is calculated by weighted summation, and finally the environmental element color sensory stimulation data containing the element ID, the color contrast intensity value, the change frequency value, the emotional tendency label, and the sensory stimulation score is generated.

[0151] Step S54: performing environmental element contour characteristic analysis according to the environmental element implantation effect data, to generate environmental element contour characteristic data;

[0152] In the embodiment of the present application, the Canny edge detection algorithm is used for contour extraction of the element image, and the low threshold and high threshold are set to obtain clear contour lines. The contour characteristics are quantified by calculating the contour perimeter, area, eccentricity and other geometric parameters. The Fourier descriptor is used for frequency domain analysis of the contour shape, and the first several low frequency coefficients are extracted as the shape feature vector to describe the overall morphological characteristics of the contour. At the same time, the smoothness of the contour is analyzed, and the average curvature change of adjacent points on the contour curve is calculated. If the average curvature change is less than a certain interval, it is determined as a smooth contour. Finally, the environmental element contour characteristic data containing element ID, contour perimeter, area, eccentricity, Fourier descriptor coefficient and smoothness is generated.

[0153] Step S55: Perform environmental element form tension analysis based on the environmental element color three-element data and the environmental element contour characteristic data to generate environmental element form tension data;

[0154] In the embodiment of the present application, color guiding force, contour dynamic sense and element proportion coordination are used as evaluation dimensions. According to the color three-element data, the influence of color warm contrast and brightness difference on visual guidance is analyzed. For example, a high-saturation red arrow-shaped element has strong color and strong visual guiding force. The formula Color Guide = α × |ΔH| + β × |ΔS| + γ × |ΔV| is used to evaluate the guiding strength of color to vision, wherein |ΔH|, |ΔS| and |ΔV| are the differences in hue, saturation and brightness between the element color and the surrounding environment color, and α, β and γ are weight coefficients. Combined with the contour characteristic data, the bending degree and direction change rate of the contour are calculated to evaluate the dynamic sense. For example, the direction change rate of a jagged contour is high, and the dynamic sense is strong. The golden section ratio and the three-point method principle are used to analyze the proportion coordination of each part of the element. For example, when the area ratio of the main element and the background element in the poster conforms to the golden section, the coordination score is high. Each dimension is assigned a weight (color guiding force proportion, contour dynamic sense proportion, proportion coordination proportion), and the form tension score is calculated by weighted summation to generate environmental element form tension data containing element ID, color guiding force score, contour dynamic sense score, proportion coordination score and form tension total score.

[0155] Step S56: Perform environmental element interest hue parameter analysis according to the environmental element color sensory stimulation data to generate environmental element interest hue data;

[0156] In the embodiment of the present application, the color three-element values in the color sensory stimulation data of the environmental protection element are clustered by using a clustering analysis algorithm, and similar colors are divided into the same hue category. The average value of the color three elements of each cluster center is calculated as the representative parameter of the hue category. For example, all high-saturation, high-brightness warm colors are clustered into one category, with an average hue value of 30 degrees, a saturation of 85%, and a brightness of 80%. By analyzing the color sensory stimulation scores corresponding to each hue category, the top 3 hue categories with the highest scores are selected as the interest hues. The effectiveness of the interest hues is verified in combination with user behavior data (such as the colors corresponding to the areas where the user stays for a long time). Finally, the environmental protection element interest hue data containing the interest hue ID, the average value of the color three elements, the sensory stimulation score, and the user attention degree is generated.

[0157] Step S57: Multi-dimensional collaborative environmental protection design element optimization based on the environmental protection element interest hue data and the environmental protection element form tension data to select the combined environmental protection element selection data, generate multi-dimensional collaborative environmental protection design element optimization data, and feed back to the user terminal through the multi-dimensional collaborative environmental protection design element optimization data to perform environmental art-related element intelligent rendering work.

[0158] In the embodiment of the present application, the interest hue matching degree, the form tension improvement rate, and the element coordination are used as the optimization target. The Euclidean distance between the element color in each combined environmental protection element selection data and the interest hue is calculated, and the smaller the distance, the higher the interest hue matching degree. The form tension improvement rate is calculated by comparing the form tension scores before and after optimization. The spatial relationship and visual connection between elements are analyzed by using a graph theory algorithm to evaluate the coordination. A particle swarm optimization algorithm (PSO) is used, for example, with a particle number of 50 and an iteration number of 100 to optimize the element combination. The color parameters (hue, saturation, brightness) and the form parameters (size, position, rotation angle) of the elements are adjusted in each iteration, and the target function value of the new combination is calculated. For example, the color of the bicycle pattern in the "environmental travel" theme element combination is adjusted to the blue color in the interest hue, and the outline is enlarged to highlight the dynamic feeling. After iterative optimization, the element combination with the optimal target function value is selected to generate multi-dimensional collaborative environmental protection design element optimization data containing the element ID, the optimized color parameters, the optimized form parameters, and the optimization score. The optimization data is transmitted to the user terminal in real time through the WebSocket protocol, and the GPU accelerated rendering engine (such as DirectX12) of the terminal is called to automatically render the environmental art design elements according to the optimization parameters, completing the whole process optimization from data driving to intelligent rendering.

[0159] Further, step S57 includes the following steps:

[0160] Step S571: Environmental protection element focus area analysis is performed according to the environmental protection element form tension data to generate environmental protection element focus area data;

[0161] In the embodiment of the present application, for the potential focus area, the GrabCut image segmentation algorithm is used for accurate processing. First, the circumscribed rectangle of the potential focus area is taken as the initial contour, the algorithm analyzes the color distribution and texture features of the pixels in the region through iteration, probabilistically models the foreground and background pixels based on Gaussian Mixture Model (GMM), automatically identifies and adjusts the contour boundary, until the main body of the element is completely separated from the background, forming a closed focus area contour. After the contour is determined, the geometric parameters of the focus area are calculated simultaneously: using the pixel counting method, the area S of the region is obtained by the formula: (Where p i is the i-th pixel in the region, and n is the total number of pixels) is obtained. The center coordinates (C x ,C y ) are determined by the formula (Where (x I ,y i ) is the horizontal and vertical coordinates of the i-th pixel) is determined. According to the coordinates of the outermost points of the contour, the length L and the width W of the circumscribed rectangle are calculated using the formula L = max(x i )-min(x i ), W = max(y i )-min(y i ), where max(x i ) and min(x i ) are the maximum and minimum values of the horizontal coordinates of the foreground pixels, and max(y i ) and min(y i ) are the maximum and minimum values of the vertical coordinates of the foreground pixels, and the position of the circumscribed rectangle is determined by the minimum x coordinate and the minimum y coordinate, and finally the focus area data of the environmental protection element is generated.

[0162] Step S572: based on the environmental protection element interest hue data, the environmental protection element focus area data is processed for environmental protection element color feeling coordination optimization, and environmental protection element color feeling coordination data is generated;

[0163] In the embodiment of the present application, the environmental element interest hue data is matched with the environmental element focus area data. For each focus area, the Euclidean distance between the existing color in the area and the interest hue is calculated, and the area with a distance greater than a set threshold is regarded as a color feeling inharmonious area. A color balance adjustment algorithm is used to correct the color of the inharmonious area according to the hue H, saturation S, and lightness V parameters of the interest hue. For example, if the interest hue is a high-saturation blue color, and the current color of an element in the focus area is a low-saturation gray blue color, the color of the element is adjusted to approach the interest hue by increasing the intensity of the blue channel and increasing the saturation. At the same time, the color harmony theory (such as the complementary color and similar color matching principle) is used to analyze the color relationship between the focus area and the surrounding elements. If the focus area is yellow, the surrounding elements can appropriately introduce the complementary color blue for decoration to enhance the color level. When evaluating the optimization effect, the color contrast change value before and after adjustment is calculated using the contrast calculation formula (L max , L min are the maximum and minimum color brightness values, respectively). The color harmony degree evaluation formula (w i is the weight, and h i is the harmony value corresponding to each color relationship) is used to calculate the color harmony score. By calculating the color contrast and color harmony score before and after adjustment, the color feeling coordination optimization effect is evaluated, and the environmental element color feeling coordination data including the focus area ID, the color parameter before optimization, the color parameter after optimization, the color contrast change value, and the color harmony score is finally generated.

[0164] Step S573: Based on the environmental element color feeling coordination data, the combined environmental element selection data is optimized for multi-dimensional coordinated environmental design elements to generate multi-dimensional coordinated environmental design element optimization data.

[0165] In the embodiment of the present application, the color sense synergy degree, the shape tension fit degree, and the spatial coordination between elements are taken as core optimization indexes. The color sense synergy degree is determined according to the color harmony score of step S572, and the shape tension fit degree is calculated by comparing the shape tension score change before and after the optimization of the focus area. The spatial coordination between elements is analyzed by using the topological distance algorithm in graph theory to analyze the relative position relationship of each element in the picture, and the spatial tightness and visual fluency between elements are calculated. Genetic algorithm is used for iterative optimization, for example, the initial population number is set to 80 element combination schemes, the crossover probability is 0.7, and the mutation probability is 0.03. In each iteration, the color parameters and shape parameters (size, rotation angle, position) of the elements are adjusted, for example, the originally small and scattered elements are enlarged and rearranged to form a visual hierarchy around the focus area. According to the optimization objective function, the comprehensive score of each scheme is calculated, and the scheme with the highest score is selected as the final optimization result. The multi-dimensional coordinated environmental design element optimization data including element ID, optimized color parameters, optimized shape parameters, index scores, and comprehensive scores are finally generated, which provides accurate parameter basis for intelligent rendering of environmental art design.

[0166] Further, step S572 includes the following steps:

[0167] The focus area of the environmental art element is subjected to environmental element graph boundary softening processing to generate environmental element boundary softening data.

[0168] In the embodiment of the present application, the contour coordinate information of each focus area is obtained. Gaussian blur algorithm is used for softening processing of the graph boundary, for example, the Gaussian kernel size is set to 5x5, and the elements are calculated by the formula , wherein the standard deviation σ=1.5, and (x, y) represents the neighborhood coordinate offset relative to the current processing pixel. For the pixel points within the range of 10 pixels around the contour, the weighted average is performed by the formula , wherein w i is the Gaussian weight, and RGB i is the neighborhood pixel value. Taking the “Earth” element as an example, the pixel points within a certain range (set to 10 pixels) around the contour are weighted and averaged according to the Gaussian function to calculate the RGB value of each pixel point. For example, for a red pixel point (RGB value is 255, 0, 0) on the contour edge, after the neighborhood pixel points participate in the weighted calculation, the RGB value of the point becomes 230, 10, 10, realizing the smooth transition of the boundary. In the softening process, the original coordinates, the RGB values before and after softening, the softening influence range, and other information of each pixel point are recorded to generate environmental element boundary softening data including focus area ID, boundary pixel coordinates, original RGB value, softening RGB value, and softening parameters (kernel size, standard deviation).

[0169] Preferably, the environmental element region reconstruction data is generated according to the environmental element boundary softening data.

[0170] In the embodiment of the present application, the dilation and erosion operations in morphological image processing are used to reconstruct the focus region. First, the dilated operation is performed on the softened boundary, for example, using a 3*3 square structural element, through the formula The boundary is expanded outward by 2 pixels, where A is the original image, B is the structural element, and the visual existence of the figure is increased, is the symmetry of the structural element about the origin, is the position of the structural element with z as the center. Taking the "leaf" element as an example, after the dilation operation, the edge part of the leaf is extended, and the overall shape is more full. Then, the erosion operation is performed, also using a 3*3 square structural element, through the formula The figure is contracted inward by 1 pixel, B z is the area of the structural element at position z, which eliminates the redundant noise points that may be generated by the dilation operation, making the figure contour more clear. In the reconstruction process, the geometric parameters of the region are recalculated, and the area is calculated through the formula , where p i is the i-th pixel in the region, the perimeter is calculated by the contour tracking algorithm, and the center coordinates are determined by the formula , (x i , y i ) are the image coordinates of the i-th foreground pixel, and the circumscribed rectangle parameters are obtained by calculating the coordinates of the outermost points of the contour. At the same time, the pixel distribution in the reconstructed region is analyzed, and the proportion of different color pixels is counted , where m j is the number of j kinds of color pixels, r j is the proportion of the j-th color pixel, and n is the total number of all pixels in the region. Finally, the environmental element region reconstruction data containing the focus region ID, the reconstructed area, the perimeter, the center coordinates, the circumscribed rectangle parameters, and the pixel color distribution proportion is generated.

[0171] Preferably, the environmental element color feeling coordination data is generated by performing environmental element color feeling coordination optimization processing on the environmental element region reconstruction data based on the environmental element interest color tone data.

[0172] In the embodiment of the present application, the environmental element interest color tone data is matched with the environmental element region reconstruction data. For each reconstructed region, the Euclidean distance formula The distance between the existing color (L1, a1, b1) in the region and the color of interest (L2, a2, b2) in the Lab color space is calculated, where D represents the overall perceived difference between the current region color and the color of interest, (L1, a1, b1) represents the color value of the color mean of the current reconstructed region in the Lab color space, and (L2, a2, b2) represents the reference target Lab color value of the color of interest. For example, the Euclidean distance threshold is set to 25, and the region with a distance greater than the threshold needs to be optimized in color. The color space in the region is converted from RGB to Lab using a color conversion matrix method, and the conversion is realized by the formula where M is the conversion matrix from RGB to Lab space, C is the offset vector, and R, G, and B represent the red, green, and blue three-channel values of the pixels in the original image. According to the coordinates of the color of interest in the Lab space, the color adjustment parameters are calculated. ΔL = L2 - L1, Δa = a2 - a1, and Δb = b2 - b1, where ΔL, Δa, and Δb represent the difference between the current color in the region and the color of interest. At the same time, a color contrast adjustment algorithm is used, and the contrast before and after adjustment is calculated by the formula where C is the original contrast, and C min and C max are the minimum and maximum contrast values, respectively. The color similarity score before and after adjustment is calculated by the formula S = e -αD where S reflects the matching degree of the reconstructed region and the color of interest, and α is the similarity decay coefficient. The color contrast score is calculated by the formula where L max and L min represent the maximum and minimum values of the luminance values in the region.

[0173] Further, step S573 includes the following steps:

[0174] According to the combined environmental element selection data, the environmental element attribute layer is parsed to generate environmental element attribute layer data;

[0175] In the embodiment of the present application, the SVG document parsing technology is used to disassemble the graphic structure of each element by using the combined environmental protection element selection data. The geometric shape (such as rectangle, circle, path), filling attribute (color, gradient), stroke attribute (width, color), transparency and other information of the element are extracted in layers to construct a multi-level attribute layer structure. For example, for an environmental protection theme illustration containing "mountain", "forest" and "river", the contour line of the mountain is taken as the stroke layer, the filled green is taken as the fill layer, and the flowing effect of the river is taken as the dynamic effect layer. At the same time, the superposition order and occlusion relationship between the layers are analyzed, and the Z-axis depth value of the layer is determined by calculating the bounding box and spatial position of the graphic. For elements with gradient effect, the element is further decomposed into a base color layer and a gradient adjustment layer, and the starting point, end point, color distribution and other parameters of the gradient are recorded.

[0176] Preferably, the multi-dimensional collaborative environmental protection design element optimization data is generated by performing multi-dimensional collaborative environmental protection design element optimization on the environmental protection element attribute layer data based on the environmental protection element color feeling synergy data.

[0177] In the embodiment of the present application, the environmental protection element color feeling synergy data is associated with the environmental protection element attribute layer data. For the color attribute of each layer, the optimized color parameters in the color feeling synergy data are adjusted. For example, if the original fill color of a layer is RGB(100, 150, 200) and the target color after color feeling synergy optimization is RGB(120, 180, 220), then the RGB values of all pixels in the layer are transitioned to the target color by a certain proportion through a color interpolation algorithm. For a composite element containing multiple sub-layers, the color harmony theory is used to adjust the color contrast and complementary relationship between the sub-layers. For example, in the "ecosystem" element, the green of the plant layer and the brown of the animal layer are adjusted to complementary colors to enhance the visual hierarchy. At the same time, according to the color similarity score in the color feeling synergy data, the transparency of the layer is optimized, and the transparency of the layer with a lower score is appropriately reduced to make it more integrated with the overall color tone.

[0178] Preferably, the multi-dimensional collaborative environmental protection design element optimization data is generated by performing multi-dimensional collaborative environmental protection design element optimization on the environmental protection element attribute layer data based on the environmental protection element color feeling synergy data.

[0179] In the embodiment of the present application, the layer synthesis algorithm is used for integration processing. Firstly, the stacking order is determined according to the Z-axis depth value of the layer, and all the layers are arranged from the bottom layer to the top layer in turn. For the layer with transparency, the Alpha blending algorithm is used to calculate the final color value of the pixel. For the layer with dynamic effect, such as flowing water, floating leaves, etc., the continuous animation sequence is generated through the key frame interpolation technology. In the integration process, the edge connection between the layers is checked, the anti-aliasing algorithm is used for smoothing the boundary to eliminate the possible jagged edges. At the same time, the overall visual center of the integrated element is calculated to ensure that the element remains balanced in the picture.

[0180] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting, the scope of the present application being defined by the appended claims and not by the above description, therefore all the variations falling within the meaning and scope of the equivalent elements of the application file are intended to be included in the present application.

[0181] The above description is merely a specific implementation of the present application, enabling those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An environmental art design element selection and optimization method based on data analysis, characterized in that, The method comprises the following steps: Step S1: obtaining environmental protection art design element data, and performing environmental protection element semantic analysis on the environmental protection art design element data to generate environmental protection element semantic data; Step S2: performing environmental protection design element correlation integration processing on the environmental protection art design element data based on the environmental protection element semantic data to generate environmental protection design element correlation data; Step S3: performing environmental protection element applicability analysis according to the environmental protection design element correlation data to generate environmental protection element applicability data; Step S4: performing environmental protection element implantation effect analysis according to the environmental protection element applicability data to generate environmental protection element implantation effect data; Step S4 comprises the following steps: Step S41: performing environmental protection element propagation path analysis according to the combined environmental protection element selection data to generate environmental protection element propagation path data; Step S42: performing environmental protection element browsing time quantification according to the environmental protection element propagation path data to generate environmental protection element browsing time data; Step S43: performing environmental protection element response analysis according to the environmental protection element browsing time data to generate environmental protection element response data; Step S44: performing environmental protection element visual feedback characteristic analysis according to the environmental protection element browsing time data to generate environmental protection element visual feedback characteristic data; performing environmental protection element correction processing on the environmental protection element visual feedback characteristic data based on the environmental protection element response data to generate corrected environmental protection element visual feedback characteristic data; performing environmental protection element behavior feedback feature analysis according to the corrected environmental protection element visual feedback characteristic data to generate environmental protection element behavior feedback feature data; and performing environmental protection element implantation effect analysis according to the environmental protection element behavior feedback feature data to generate environmental protection element implantation effect data; Step S5: performing multi-dimensional collaborative environmental protection design element optimization of combined environmental protection element selection based on the environmental protection element implantation effect data to generate multi-dimensional collaborative environmental protection design element optimization data, and feeding back the multi-dimensional collaborative environmental protection design element optimization data to a user terminal to perform environmental protection art design element intelligent rendering work; Step S5 comprises the following steps: Step S51: performing environmental protection element color three-element analysis according to the environmental protection element implantation effect data to generate environmental protection element color three-element data; Step S52: performing environmental protection element color visual perception time sequence analysis according to the environmental protection element color three-element data to generate environmental protection element color visual perception time sequence data; Step S53: performing environmental protection element color sensory stimulation analysis based on the environmental protection color three-element data and the environmental protection element color visual perception time sequence data to generate environmental protection element color sensory stimulation data; Step S54: performing environmental protection element contour characteristic analysis according to the environmental protection element implantation effect data to generate environmental protection element contour characteristic data; Step S55: performing environmental protection element form tension analysis based on the environmental protection element color three-element data and the environmental protection element contour characteristic data to generate environmental protection element form tension data; Step S56: performing environmental protection element interest color tone parameter analysis according to the environmental protection element color sensory stimulation data to generate environmental protection element interest color tone data; ​ Step S57: Perform environmental protection element focus area analysis according to the environmental protection element form tension data to generate environmental protection element focus area data; perform environmental protection element graphic boundary softening processing on the environmental protection art element focus area to generate environmental protection element boundary softening data; perform environmental protection element area reconstruction according to the environmental protection art element boundary softening data to generate environmental protection element area reconstruction data; perform environmental protection element color sense synergistic optimization processing on the environmental protection element area reconstruction data based on the environmental protection element interest tone data to generate environmental protection element color sense synergistic data; perform environmental protection element attribute layer analysis according to the combined environmental protection element selection data to generate environmental protection element attribute layer data; perform multi-dimensional synergistic environmental protection design element optimization on the environmental protection element attribute layer data based on the environmental protection element color sense synergistic data to generate multi-dimensional synergistic environmental protection design element layer data; perform environmental protection element layer integration according to the multi-dimensional synergistic environmental protection design element layer data to generate multi-dimensional synergistic environmental protection design element optimization data; and feedback the multi-dimensional synergistic environmental protection design element optimization data to the user terminal to perform environmental protection art design element intelligent rendering work. 2.The method of claim 1, wherein, Step S1 includes the following steps: Step S11: Obtain environmental protection art design element data; Step S12: Perform environmental protection design element clustering division according to the environmental protection art design element data to generate divided environmental protection design element data; Step S13: Perform environmental protection design element feature analysis according to the divided environmental protection design element data to generate environmental protection design element feature data; Step S14: Perform environmental protection element form structure and color extraction processing according to the environmental protection design element feature data to generate environmental protection element form structure data and environmental protection element color data; Step S15: Perform environmental protection element scene analysis according to the environmental protection element form structure data and the environmental protection element color data to generate environmental protection element scene data; Step S16: Perform environmental protection element semantic analysis on the environmental protection art design element data based on the environmental protection element scene data to generate environmental protection element semantic data. 3.The method of claim 1, wherein, Step S2 includes the following steps: Step S21: Perform environmental protection element semantic field analysis according to the environmental protection element semantic data to generate environmental protection element semantic field data; Step S22: Perform environmental protection element semantic expansion analysis according to the environmental protection element semantic field data to generate environmental protection element expanded semantic field data; Step S23: Perform environmental protection design element association integration processing on the environmental protection art design element data based on the environmental protection element expanded semantic field data to generate environmental protection design element association data. 4.The method of claim 1, wherein, Step S3 includes the following steps: Step S31: Perform environmental protection element cultural scene analysis according to the environmental protection design element association data to generate environmental protection element cultural scene data; Step S32: Perform environmental protection element compatibility analysis according to the environmental protection element cultural scene data to generate environmental protection element compatibility data; Step S33: Perform heterogeneous environmental protection element detection on the environmental protection design element association data based on the environmental protection element compatibility data to generate heterogeneous environmental protection element data; Step S34: Perform environmental protection ambiguity element elimination on the environmental protection design element association data based on the heterogeneous environmental protection element data to generate ambiguity environmental protection element elimination data; Step S35: According to the ambiguous environmental element elimination data, the environmental element combination selection processing is performed to generate the combined environmental element selection data.

Images