Automobile form design method and related equipment

Abstract

The invention discloses an automobile form design method and related equipment, and relates to the technical field of automobile design, and the method comprises the steps: obtaining a perceptual image correlation model of an automobile form; determining an automobile form according to a mapping relation between a user visual perception image and an automobile form characteristic factor in the perception image association model, and obtaining a plurality of design schemes; screening the design scheme through the perceptual vocabularies expected by the user group to obtain a screened design scheme; performing detail divergence and convergence on the screened design schemes to determine a final design scheme; the design efficiency and accuracy are improved, and the scientificity and systematicness of the design process are improved; design resources and time are reasonably distributed by determining the importance sequence of each morphological element.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile design, and in particular to an automobile form design method and related equipment. Background Art

[0002] Automobile design is a complex and systematic project, covering multiple aspects such as form design, planning, engineering and general layout design. Among the many elements of automobile design, visual perception experience plays a vital role, especially when users first come into contact with the car and develop emotional identification, the visual perception stage often plays a decisive role and directly affects the user's purchasing decision.

[0003] In recent years, with the rapid development of the automobile industry, the production and sales of new cars have increased year by year, and the demand for design creativity has also increased. Faced with the dual pressures of industrial upgrading and user needs, automobile appearance design work faces the requirements of higher quality and higher efficiency. However, although relevant research and practice have formed relatively stable design processes and technologies, the research on creative design methods of automobile form is still relatively lagging behind. Existing research is mainly limited to using computer technology to predict design solutions, and how to integrate designers' personal experience and intuitive creativity to enhance the value of design solutions is still a problem that needs to be solved urgently.

[0004] From the perspective of user visual perception, the depth and comprehensiveness of the research on the design rules of automobile form also needs to be strengthened. Most previous studies focused on the morphological characteristics of a certain angle, lacking a comprehensive review and systematic study of automobile form. Especially for SUV form, although there are relevant perceptual studies, there is still a lack of qualitative and quantitative comprehensive research. In addition, there is still a lack of research on the importance ranking of automobile form feature factors, resulting in a lack of scientific guidance for design creativity.

[0005] Therefore, in order to improve the creativity and quality of automobile form design and meet the dual challenges of industrial upgrading and user needs, it is necessary to strengthen the research on automobile form design rules, especially from the perspective of user visual perception, to conduct a comprehensive and systematic study of automobile form. At the same time, it is also necessary to explore how to apply the design process and integrate the designer's personal experience and intuitive creativity to enhance the value of the design solution. Summary of the invention

[0006] The purpose of the present invention is to provide a method for designing an automobile form and related equipment, which improves the efficiency and accuracy of the design and enhances the scientificity and systematicness of the design process.

[0007] The purpose of the present invention is achieved by the following technical solutions:

[0008] In a first aspect, the present application provides a method for designing an automobile form, the method comprising:

[0009] Obtain a perceptual image correlation model for the vehicle form;

[0010] Determine the vehicle form through the mapping relationship between the user's visual perceptual image and the vehicle form characteristic factors in the perceptual image correlation model, and obtain multiple design schemes;

[0011] Screen the design schemes through the emotional vocabulary expected by the user group to obtain the screened design schemes;

[0012] Diverge and converge the details of the screened design schemes to determine the final design scheme.

[0013] Preferably, the obtaining of the perceptual image correlation model for the vehicle form includes:

[0014] Obtain a set of vehicle form types through the form characteristic factors of each vehicle form element;

[0015] Obtain the semantic evaluation of the forms of each form element of different vehicle models by users to determine the user's visual perceptual image;

[0016] Construct a correlation model between the vehicle form and the user's visual perceptual image vocabulary through the user's visual perceptual image and the set of vehicle form types.

[0017] Preferably, the obtaining of the set of vehicle form types through the form characteristic factors of each vehicle form element includes:

[0018] Obtain the form characteristics of each form element of different vehicle models from different perspectives;

[0019] Process the form characteristics to obtain multiple form characteristic indexes of each form element, and establish a form characteristic index system through the multiple form characteristic indexes;

[0020] Analyze the form characteristics of each form element to obtain the importance ranking of form characteristics;

[0021] Obtain the form elements that need to be focused on according to the importance ranking of form characteristics;

[0022] Use the vector space model to process the form characteristics of the form elements that need to be focused on, and form a set of vehicle form types based on form elements through cluster analysis.

[0023] Preferably, the analyzing of the form characteristics of each form element to obtain the importance ranking of form characteristics includes:

[0024] Obtain the importance ranking of the form characteristics of each form element based on the eye movement tracking experiment as the first ranking by dividing the area of interest;

[0025] Obtain the subjective comprehensive weight ranking of the morphological characteristics of each morphological element;

[0026] Fuse the first ranking and the subjective comprehensive weight ranking to obtain the importance ranking of each final morphological element.

[0027] Preferably, the importance ranking based on the eye movement tracking experiment for obtaining the morphological characteristics of each morphological element is used as the first ranking; it includes:

[0028] Divide the regions of interest for the automotive morphology under each perspective respectively; obtain the eye movement experiment data of the subjects staying in each region of interest;

[0029] Calculate the mean value of each item of eye movement experiment data in each region of interest respectively when the subjects observe multiple vehicle models; and sort the mean values to obtain the first sub-ranking of this item of eye movement data;

[0030] Obtain the comprehensive ranking of each region of interest through multiple first sub-rankings based on eye movement data items;

[0031] Determine the first ranking according to the comprehensive ranking of each region of interest and the morphological elements of this region of interest.

[0032] Preferably, the subjective comprehensive weight ranking for obtaining the morphological characteristics of each morphological element includes:

[0033] Obtain each hierarchical index of the automotive morphology through the analytic hierarchy process; among them, the hierarchy includes the target layer, the criterion layer and the sub-criterion layer; the criterion layer includes the front morphological characteristics, the side morphological characteristics and the rear morphological characteristics; the sub-criterion layer corresponds to the region of interest one by one;

[0034] Sort the comprehensive weights of each sub-criterion layer to obtain the comprehensive weight ranking of each morphological element.

[0035] Preferably, using the vector space model to formalize the morphological characteristics of the key morphological elements to be focused on, and through cluster analysis, form a set of automotive morphology types based on morphological elements, including:

[0036] For a certain key morphological element to be focused on, obtain multiple characteristics of this morphological element in multiple vehicle models;

[0037] If the vehicle model frequency of a certain characteristic appears within a preset range, then use this characteristic as the input characteristic option of the space model;

[0038] Obtain the weight value of this characteristic through the vehicle model frequency of each input characteristic and the reverse frequency of the overall vehicle models;

[0039] Construct the space model of the morphological characteristics of this morphological element based on the weight value of the characteristic;

[0040] Based on the spatial model, perform clustering analysis on the forms of multiple vehicle models from various perspectives to obtain a clustering model of vehicle forms from each perspective.

[0041] Preferably, constructing an association model between vehicle form and the user's visual perception image vocabulary through the set of the user's visual perception image and vehicle form types includes:

[0042] Establish the association relationship between each form element and the perception image phrase from multiple angles respectively; obtain an evaluation structure relationship diagram based on the upper-layer image and lower-layer specific form of the vehicle form.

[0043] In a second aspect, the present application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of any method of the present application are implemented.

[0044] In a third aspect, the present application provides a computer-readable storage medium, which stores computer instructions. When a computer reads the computer instructions, the computer executes the steps of any method of the present application.

[0045] Compared with the prior art, the beneficial effects of the present invention at least include: by constructing a perception image association model of vehicle form, designers can quickly determine the vehicle form based on the mapping relationship between the user's visual perception image and vehicle form characteristic factors, and generate multiple design schemes. This method avoids the limitations of relying on personal experience or intuition in traditional design, and improves the scientificity and accuracy of design. Screening design schemes through the emotional vocabulary expected by the vehicle user group can ensure that the final design scheme better meets the aesthetic preferences and needs of users; it helps to improve user satisfaction and acceptance of vehicle form. Conducting detail divergence and convergence on the selected design schemes helps designers further optimize and improve details on the basis of maintaining the overall design direction, thereby enhancing the overall quality and competitiveness of the design; by deeply analyzing the form characteristic factors of each form element of the vehicle and the semantic evaluation of the form elements of different vehicle models by users, designers can quickly and accurately discover more innovative design elements and inspirations, and inject more personality and differential elements into vehicle form design. This method combines scientific methods and tools such as eye movement tracking experiments, analytic hierarchy process, and vector space model, making the design process more scientific, systematic, and controllable; it helps to improve design efficiency, reduce design costs, and enhance the predictability and replicability of design results. Description of the Drawings

[0046] Figure 1 It is a schematic diagram of the vehicle form design method according to an embodiment of the present invention;

[0047] Figure 2 It is the region of interest map of the SUV front view angle (0 degrees) in the embodiment of the present invention;

[0048] Figure 3 It is the region of interest map of the SUV at a 45-degree front angle in the embodiment of the present invention;

[0049] Figure 4 It is the region of interest map of the SUV side view angle in the embodiment of the present invention;

[0050] Figure 5 It is the region of interest map of the SUV at a 45-degree rear view angle in the embodiment of the present invention;

[0051] Figure 6 It is the bar chart of eye movement data analysis for each region of interest of E1 - E5 in the SUV front view in the embodiment of the present invention;

[0052] Figure 7 It is the bar chart of eye movement data analysis for each region of interest of E6' - E11 in the SUV side view in the embodiment of the present invention;

[0053] Figure 8 It is the bar chart of eye movement data analysis for each region of interest of E12' - E16' in the SUV rear view in the embodiment of the present invention;

[0054] Figure 9 It is the structural diagram for evaluating the rear view form of the SUV in the embodiment of the present invention. Detailed implementation manners

[0055] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this invention will be more complete and comprehensive, and the concept of the example embodiments will be fully conveyed to those skilled in the art. Like reference numerals in the figures denote the same or similar structures, and thus their repeated description will be omitted.

[0056] The words expressing positions and directions described in the present invention are all illustrated by taking the accompanying drawings as examples, but can be changed according to needs, and all the changes made are included in the protection scope of the present invention.

[0057] This application takes an SUV as an example to illustrate the steps of the vehicle form design method:

[0058] As a typical product design process, the Double Diamond Model involves two divergence and convergence processes of finding the right questions and finding the right answers. Divergence is the process of exploring various possibilities and increasing options; convergence is the process of evaluation and selection, reducing options. Specifically, it can be summarized as follows: in the first divergence and convergence process, starting from the current situation of the design, the research focuses on the design problems; in the second divergence process, for the discovered problems, multiple solutions are proposed in the divergence process, and in the convergence process, the internal and external limiting factors of the design are fully considered to determine the best design solution.

[0059] An SUV is a sporty vehicle with characteristics such as large wheel sizes, high ground clearance, and good sports performance. Most SUVs have four-wheel drive powertrains. The specific automotive form elements mainly include engine covers, front bumpers, front headlights, fog lamp combinations, tire combinations, front fenders, rearview mirrors, front doors, rear doors, side windows, rear hatches, rear bumpers, rear taillight combinations, roofs, frames, door combinations, rear windshields, and front windshields, etc.

[0060] Refer to the appendix Figure 1 Based on the Double Diamond Model, an embodiment of this application provides a method for automotive form design, and the method includes:

[0061] Obtain the perceptual image correlation model of the automotive form;

[0062] Determine the automotive form through the mapping relationship between the user's visual perceptual image and the automotive form feature factors in the perceptual image correlation model, and obtain multiple design solutions;

[0063] Screen the design solutions through the emotional vocabulary expected by the automotive user group to obtain the screened design solutions;

[0064] Perform divergence and convergence on the details of the screened design solutions to determine the final design solution.

[0065] In some specific applications, for the automotive form design task, the existing form features (local form features and overall form features) are analyzed through the perceptual image correlation model of the automotive form. This usually involves the mining and analysis of a large amount of vehicle models and user evaluation data to reveal users' preferences and cognitions for different automotive form features. According to the SIC population classification system, automotive users are divided into ten typical groups, and different groups have unique user characteristics, which constitute the group basis for user research. After obtaining the perceptual image correlation model, using the mapping relationship in the model and according to the design goal, the automotive form is determined based on the users' visual perceptual images. This usually involves the adjustment and optimization of automotive form feature factors to generate multiple design schemes that meet users' expectations. These schemes will vary in terms of form, lines, proportions, etc. to meet the aesthetic and functional requirements of different users. After obtaining multiple design schemes, further screening and optimization are required. This usually involves the collection and analysis of emotional words expected by the automotive user group, such as "fashionable", "dynamic", "steady", etc. Then, based on these emotional words, the design schemes are screened to retain those that better meet users' expectations and aesthetic preferences. On the basis of the screened schemes, the divergence of local features is carried out, and after design convergence again, the optimal scheme is obtained.

[0066] By constructing the perceptual image correlation model of the automotive form, the mapping relationship between users' visual perceptual images and automotive form feature factors can be accurately captured; it helps designers clarify users' needs and preferences at the initial stage of design, thus designing automotive forms that better meet market demands; at the same time, this method makes the design process more targeted, reducing design mistakes and waste.

[0067] In some embodiments, obtaining the perceptual image correlation model of the automotive form includes:

[0068] Obtain a set of automotive form types through the form feature factors of each form element of the vehicle;

[0069] Obtain the semantic evaluation of the forms of each form element of different vehicle models by users to determine the users' visual perceptual images;

[0070] Construct an association model between the automotive form and the vocabulary of the users' visual perceptual images through the users' visual perceptual images and the set of automotive form types.

[0071] The working principle of the above technical solution is as follows: Automotive form elements refer to various components and features that make up the appearance of a vehicle, such as the engine cover, front bumper, headlights, tires, etc. These elements together determine the appearance and style of the vehicle. For each form element, its unique form feature factors are extracted; these factors can be the size, shape, lines, proportions, etc. of the form, used to describe and distinguish different automotive forms. According to the extracted form feature factors, automotive forms are classified into different types or sets.

[0072] According to the SIC population classification system, automotive users are divided into ten typical groups. Different groups have unique user characteristics, forming the group basis for user research; a certain number of subjects can be selected. The subjects have certain automotive knowledge and aesthetic preferences and can accurately express their views and feelings on different automotive forms. Show the form elements of different models to the subjects and ask the subjects to evaluate the forms of these elements using specific semantic words or phrases. For example, words such as "fashionable", "dynamic", "steady" can be used to describe automotive forms. Collect the evaluation data of the subjects and organize and analyze it. Through statistics and induction, determine the visual perception images of users for different automotive form elements; it is also possible to collect Chinese adjective groups related to SUVs from websites, magazines, and other media and screen them in combination with the following criteria: These criteria include: (1) Select adjectives associated with the form of SUVs. In order to better study the form of SUVs, eliminate adjectives that have nothing to do with the form as much as possible. (2) Try to select adjectives with universality and eliminate adjectives that are not often used or are prone to ambiguity.

[0073] Using the 5-point Likert scale method, the matching degree of adjectives describing product forms is divided into five levels, and the 31 groups of words collected are scored, ranging from "strongly disagree (1)" to "strongly agree (5)", to evaluate the degree of emotional response of each adjective describing SUVs. The effectiveness of the word groups is evaluated through an independent samples t-test, and the related words are simplified through principal component analysis. Vocabulary sorting is carried out from three aspects: form aesthetics, trend, and functionality. The words focusing on the user's visual perception that conform to the form of SUVs include "individual", "innovative", "elegant", "simple", and "technological". The related word groups are used as word groups for describing the user's visual perception image in later research to guide the research on the form of SUVs.

[0074] Based on the set of user visual perception images and automotive form types, establish the mapping relationship between them. It is used to describe the preferences and cognitions of users for different automotive form types. Using technical means such as statistics, data mining, or machine learning, construct an association model between automotive forms and user visual perception image words; this model can reflect the perception and evaluation of users for different automotive form types and provide valuable reference information for designers.

[0075] The effects of the above technical solution are as follows: By obtaining the semantic evaluations of the forms of various morphological elements of different models of vehicles from the subjects, it is possible to deeply understand the users' visual perception images of the vehicle form, thereby accurately capturing the preferences and needs of users for vehicle exterior design; it provides a clear design direction for vehicle designers and helps to design vehicle products that better meet the market demand. Constructing an association model between vehicle form and users' visual perception image vocabulary can transform the abstract perception of vehicle form by users into specific design parameters and indicators; designers can directly refer to these parameters and indicators during the creation process, quickly generate multiple design solutions, and screen and optimize them through the model, thereby improving the design efficiency.

[0076] In the process of constructing the association model, it is necessary to comprehensively consider the morphological characteristic factors of various morphological elements of the vehicle and the semantic evaluations of these elements by users; this helps designers to integrate new design concepts and elements on the basis of retaining traditional design elements, creating vehicle forms with uniqueness and innovation; through the association model, it is possible to predict the reactions and preferences of users to different design solutions, thereby making targeted adjustments and optimizations at the design stage; it helps to enhance the user experience of vehicle products, meet the personalized needs of users, and increase user satisfaction and loyalty.

[0077] In some embodiments, obtaining the set of vehicle form types through the morphological characteristic factors of various morphological elements of the vehicle includes:

[0078] Obtaining the morphological characteristics of various morphological elements of different vehicle models from different perspectives;

[0079] Processing the morphological characteristics to obtain multiple morphological characteristic indicators of each morphological element, and establishing a relevant morphological characteristic index system through the multiple morphological characteristic indicators;

[0080] Analyzing the morphological characteristics of each morphological element to obtain the importance ranking of morphological characteristics;

[0081] Obtaining the morphological elements that need to be focused on according to the importance ranking of morphological characteristics;

[0082] Processing the morphological characteristics of the morphological elements that need to be focused on by using a vector space model, and forming a set of vehicle form types based on morphological elements through cluster analysis.

[0083] The working principle of the above technical solution is as follows: Obtain images or 3D model data of different brand models (SUVs) from multiple perspectives (such as the front, side, top, etc.); The data should include various morphological elements of the car, such as the body contour, front face design, headlight shape, window layout, etc.; Use image processing or 3D modeling technology to extract the morphological features of each morphological element, such as contour lines, dimensional ratios, shape features, etc.; Process the extracted morphological features and convert them into multiple quantifiable morphological feature indicators; Integrate multiple morphological feature indicators to form a complete morphological feature indicator system. This system can reflect the characteristic information of each morphological element of the car.

[0084] According to the ranking of the importance of morphological features, screen out the morphological elements that need to be focused on; Use the vector space model to further process the morphological features of the morphological elements that need to be focused on; Regard each morphological element as a vector, and its characteristic indicators as the components of the vector. Conduct cluster analysis on the processed morphological feature vectors, and group the morphological elements with similar features into the same category.

[0085] The clustering algorithm may include K-means, hierarchical clustering, etc.; According to the clustering results, construct a set of car morphological types based on morphological elements; This set can reflect the commonalities and differences of different models in each morphological element, providing a reference for car design and classification.

[0086] In some specific applications, in order to better display the morphological characteristics of SUV vehicles, a three-dimensional space perspective is selected to study SUVs; The analysis and research of the SUV morphology are based on the vehicle placed in the horizontal plane space, combined with the three-dimensional coordinate positioning of the vehicle itself. The body display is based on the reference points of 0 degrees on the horizontal X-axis, 0 degrees on the Y-axis, and 0 degrees on the Z-axis, and every 45 degrees clockwise rotation around the Z-axis is used as a reference. Rotating from 0 degrees to 180 degrees, a total of five angles (perspectives) are divided, namely 0 degrees, 45 degrees, 90 degrees, 135 degrees, and 180 degrees.

[0087] Analyze the SUV morphology from the engineering perspective and the design perspective. From the engineering perspective, analyze the SUV morphology according to functions and processes; From the design perspective, summarize relevant rules by analyzing the design methods of designers. Integrate the analysis results from the engineering perspective and the design perspective, summarize the methods of SUV morphology research, and determine the scope of relevant research through methods such as expert opinions and user feedback.

[0088] To study the SUV shape more effectively, the morphological characteristics of the SUV are analyzed from five angles (perspectives), and the relevant morphological characteristics of five angles, namely the front view angle (0 degrees), the front 45-degree angle (45 degrees), the side view angle (90 degrees), the rear 45-degree angle (135 degrees), and the rear view angle (180 degrees), are listed. Combining expert opinions, the characteristics of adjacent perspectives among the five typical angles are summarized, the overlapping relevant morphological indicators are merged, and the indicators with weak influence on the SUV shape are simplified. Finally, 19 morphological indicators for the front, side, and rear perspectives are formed, and a three-level indicator system for the characteristics of the relevant shape is established. Each indicator corresponds to a code for easy direct search of the corresponding shape according to the code, as shown in Table 1:

[0089] Table 1

[0090]

[0091] Obtaining the set of automobile shape types through the morphological characteristic factors of each morphological element of the automobile is a complex process involving multiple steps such as data collection, feature extraction, index system construction, feature analysis, vector space model processing, and clustering analysis. This process can comprehensively and accurately reflect the characteristic information of the automobile shape and provide strong support for automobile design and classification.

[0092] In some embodiments, the analysis of the morphological characteristics of each morphological element to obtain the importance ranking of morphological characteristics includes:

[0093] By dividing the region of interest, obtaining the importance ranking of the morphological characteristics of each morphological element based on the eye movement tracking experiment as the first ranking;

[0094] Obtaining the subjective comprehensive weight ranking of the morphological characteristics of each morphological element;

[0095] Fusing the first ranking and the subjective comprehensive weight ranking to obtain the final importance ranking of each morphological element.

[0096] In some embodiments, the obtaining the importance ranking of the morphological characteristics of each morphological element based on the eye movement tracking experiment as the first ranking includes:

[0097] Respectively dividing the region of interest of the automobile shape under each perspective; obtaining the eye movement experiment data and heat map of the subjects staying in each region of interest; the experimental data includes fixation time, first fixation duration, fixation times, number of times before the first fixation, and average pupil diameter;

[0098] Respectively calculating the mean value of each eye movement experiment data in each region of interest when the subjects observe multiple vehicle models; and sorting the mean values to obtain the first sub-ranking of the eye movement data for this item.

[0099] Obtain the comprehensive ranking of each region of interest through multiple first sub - rankings based on eye movement data items;

[0100] Determine the first ranking according to the comprehensive ranking of each region of interest and the morphological elements of this region of interest.

[0101] In some embodiments, the obtaining of the subjective comprehensive weight ranking of the morphological features of each morphological element includes:

[0102] Obtain the hierarchical indicators of the automobile form through the analytic hierarchy process; wherein, the hierarchy includes an objective layer, a criterion layer, and a sub - criterion layer; the criterion layer includes front morphological features, side morphological features, and rear morphological features; the sub - criterion layer corresponds to each region of interest one by one;

[0103] Rank the comprehensive weights of each sub - criterion layer to obtain the comprehensive weight ranking of each morphological element.

[0104] The working principle of the above - mentioned technical solution is as follows:

[0105] Use an eye - movement tracking experiment to quantitatively evaluate the morphological elements of an SUV. Select subjects who meet the experimental requirements, and tidy up the experimental environment to complete the experimental preparation. Select SUV pictures in white or light colors as the experimental objects, process the relevant SUV pictures that meet the experimental angles. To avoid interference from background elements, process the experimental pictures into white - background pictures, perform detail processing on the pictures, remove relevant detail information such as patterns and license plates, and weaken interference factors such as the high reflectivity of vehicle glass and the body. The subjects complete relevant experiments under clear experimental task requirements, and conduct eye - movement experiments on relevant static SUV pictures respectively. After obtaining the eye - movement experiment data, in order to better explore the true feelings of the subjects, after the eye - movement tracking experiment, conduct a follow - up visit to the subjects about the experimental process, record information for the deviated parts, determine the reasons for the anomalies, and finally sort out the experimental results to obtain corresponding conclusions.

[0106] Refer to the appendix Figures 2 - 5, first, assume regions of interest (ROIs) for the morphological regions of the vehicle to form corresponding hypothetical ROIs. Considering the characteristics of the eye movement experiment, to better count the relevant ROIs, three indicators, namely the overall front view morphology, the overall side view morphology, and the overall rear view morphology, are incorporated into the eye movement experiment; relevant data and heatmaps of the eye movement experiment are formed through software processing. The ROIs of the vehicle morphology are divided for each of the three perspectives respectively; an eye movement experiment is conducted on the 45-degree front angle of the SUV, and five ROIs are set, namely E1’ (upper morphology of the 45-degree front angle), E2’ (headlight morphology of the 45-degree front angle), E3’ (styling morphology between the headlights of the 45-degree front angle), E4’ (fog light morphology of the 45-degree front angle), E5’ (middle and lower part of the bumper of the 45-degree front angle), E6 (side window morphology of the 45-degree front angle), E7 (waistline morphology of the 45-degree front angle), E8 (skirt edge morphology of the 45-degree front angle), and E9 (wheel hub morphology of the 45-degree front angle). An eye movement experiment is conducted on the side view of the SUV, and six ROIs are set, namely E6’ (side window morphology), E7’ (waistline morphology), E8’ (skirt edge morphology), E9 (wheel hub morphology), E10 (front face side view morphology), and E11 (rear side view morphology); the region of the 45-degree rear angle of the vehicle is divided; and eight ROIs are set, namely E6” (rear 45-degree side window morphology), E7” (rear 45-degree waistline morphology), E8” (rear 45-degree skirt edge morphology), E12 (upper rear morphology), E13 (45-degree rear tail morphology), E14 (45-degree rear fog light morphology), E15 (45-degree rear license plate position morphology), and E16 (rear bumper morphology).

[0107] During the eye movement tracking experiment, the subjects are required to find the relevant morphologies that are important for this angle from the 45-degree front view pictures within 12 seconds, and the pictures of each vehicle model are adjusted at intervals of 2 seconds.

[0108] Obtain the eye movement experiment data of the subjects staying in each ROI; the experimental data includes fixation time, first fixation duration, fixation count, number of times before the first fixation, and average pupil diameter; after the eye movement experiment, a semi-structured interview is conducted. Combining the trajectory playback of the eye movement experiment, ask the subjects to recall the fixation positions during the experiment

[0109] to understand the situation of the subjects at that time, confirm the correspondence between the eye movement data and the psychology, and further verify the hypothetical vehicle visual perception regions;

[0110] Refer to Appendix Figures 6 - 8 : Calculate the mean values of each eye movement experiment data of the subjects in each ROI when observing multiple vehicle models respectively; and sort the mean values to obtain the first sub-sorting of this eye movement data;

[0111] The data of each region of interest are sorted item by item according to the annotation time, the number of fixations, the duration before the first fixation, etc., and all the data are aggregated to form a comprehensive ranking, as shown in Table 2:

[0112] According to the comprehensive ranking of each region of interest and the morphological elements of that region of interest, a first ranking is determined, and the first ranking is the comprehensive ranking.

[0113] Table 2

[0114]

[0115]

[0116] The subjective evaluation is mainly carried out by using the analytic hierarchy process. According to the criteria of the analytic hierarchy process, the indicators of each level of the SUV form are formed. The relevant indicators are compared pairwise using the nine-level scale method to determine the weights of the relevant indicators and complete the importance ranking of the indicators. As shown in Table 3:

[0117] Table 3

[0118]

[0119] Combining the importance ranking of the SUV morphological features generated by the objective experiment and the results of the analytic hierarchy process, because the methods of subjective and objective research have differences, and there are also certain descriptive differences in some forms. Considering the needs of later design practice, the relevant different forms are discussed in depth, and the focus group method is used to determine the final indicators. In the process of comprehensive ranking, there are some inconsistencies between the indicators of objective evaluation and subjective evaluation. For example, there are deviations in the indicators such as the overall form of the front view, the overall form of the side view, and the overall form of the rear view in the subjective and objective evaluations; the order is set according to the opinions of the focus group experts.

[0120] Combining the ranking of the objective experimental data and the weight ranking of the subjective evaluation, the final ranking of each third-level indicator is shown in Table 4:

[0121] Table 4

[0122]

[0123] Note: * indicates different indicators in subjective evaluation and objective evaluation.

[0124] The effect of the above technical solution is that through the eye movement tracking experiment, the visual behavior and attention distribution of the subjects when observing the car form can be objectively captured, thus providing direct data on which morphological elements are more attention-grabbing and which elements are ignored or less concerned. Combining the analysis of the subjective comprehensive weight, the importance of each morphological element can be evaluated more comprehensively and accurately, thus providing a scientific basis for design decisions.

[0125] By clarifying the importance ranking of each morphological element, designers can prioritize the elements that have the greatest impact on users, thereby reasonably allocating design resources and time. This can not only reduce unnecessary design iterations and modifications, but also speed up the product development process and improve design efficiency.

[0126] Ranking the importance of morphological elements helps designers identify those elements with potential innovative value. Through in-depth research and innovative design of these elements, automotive products with unique selling points and competitive advantages can be developed. Determining the hierarchical indicators and weights of automotive forms through the analytic hierarchy process can ensure the consistency and coordination among various morphological elements during the design process. This helps create automotive products with a strong overall sense and unified style, enhancing the brand's recognition and market competitiveness.

[0127] In summary, by combining eye-tracking experiments and subjective comprehensive weights to determine the importance ranking of automotive morphological elements, the scientific nature and accuracy of design decisions can be significantly improved, the user experience can be optimized, design efficiency can be increased, innovation can be promoted, and design consistency can be enhanced. This method provides a new, more comprehensive, and objective analysis tool for automotive design.

[0128] In some embodiments, the morphological features of the key morphological elements are formalized using the vector space model, and through clustering analysis, a set of automotive form types based on morphological elements is formed, including:

[0129] For a certain key morphological element, multiple features of this morphological element in multiple vehicle models are obtained;

[0130] If the frequency of a certain feature in the vehicle models is within a preset range, then this feature is used as an input feature option for the space model; the weight value of this feature is obtained through the frequency of each input feature in the vehicle models and the inverse frequency of the overall vehicle models;

[0131] w t,d = tf t,d × id t

[0132]

[0133] where w t,d is the weight value of feature t in vehicle model d, tft t,d represents the frequency of feature t appearing in a certain vehicle model d; id t represents the inverse frequency of the overall vehicle models; D represents the total number of all vehicle models; d t is the number of files containing feature t among all vehicle models;

[0134] Construct a spatial model of the morphological features of this morphological element based on the weight values of the features;

[0135] Based on the spatial model, perform clustering analysis on the morphologies of multiple vehicle models from various perspectives to obtain a clustering model of the vehicle morphologies from various perspectives.

[0136] The working principle of the above technical solution is as follows:

[0137] To construct a vectorized space, it is necessary to first define the composition method of each dimension of the vector space. Usually, these dimensions can be a set composed of the basic morphological elements (overall and local features composed of lines, surfaces, and solids, etc.) contained in the vehicle morphology. In the embodiments of this application, it mainly includes the overall and detailed features of SUVs. The composition method of these dimensions is called the feature items of SUVs. Define the SUV morphology as d, and d is represented by the feature item set (t1, t2,..., t m ), where t k represents the kth feature item, and 1 ≤ k ≤ n. Then, assign a certain weight w k to each feature item t k according to its importance in this vehicle model. At this time, the SUV morphology d can be marked as d = (t1 = w1, t2 = w2,..., t n = w n ), briefly recorded as d = (w1, w2,..., w n ). That is to say, if t1, t2,..., t n are regarded as an n-dimensional coordinate system, and w1, w2,..., w n are the corresponding coordinate values, then (w1, w2,..., w n ) is regarded as a vector in the n-dimensional space, and (w1, w2,..., w n ) is called the vector representation of the SUV morphology d.

[0138] The determination of feature items is the key to determining the vector space. The morphological features of the product summarized in the previous step can be used to define the feature vector space according to these features. The number of these features is relatively large. If all of them are regarded as different feature items, it is the simplest approach. However, this approach is likely to result in a vector space with a relatively high complexity in the research. Such a vector space not only involves a large amount of calculation but may also lead to insignificant clustering results due to too many relatively meaningless features. Therefore, it is necessary to minimize the number of features to be processed, thereby reducing the dimension of the vector space, and try to select representative morphological features as vector dimensions to improve the speed of subsequent vehicle model feature comparison. For the selection of feature elements, generally, the morphologies that have an important influence on the vehicle form are retained as the coordinate basis of the vector space. Therefore, the morphologies with high importance and their weights in the SUV form indicators are selected to establish the vector space system. In this step, the vehicle model frequency (SUV Fequency, abbreviated as SF) of the feature appearance is used as the basis for selection, and two thresholds, minSF and maxSF, the minimum and the maximum, are set. Therefore, only the feature frequencies of the feature appearance between (minSF, maxSF) will be selected as feature items.

[0139] In the vector space, each point represents the position of the vehicle model in the space, and the corresponding coordinate values are the weights of different features of the vehicle model. According to the vector space model, first, calculate the importance of each feature for the vehicle model, that is, the weight values of each feature. Appropriate thresholds are set according to the calculated weight values to delete unnecessary interfering features. This process is also called Feature selection. There are different weight calculation methods for the vector space model. The SF-ISF weight is adopted in this embodiment; SF (SUVFequency) is the frequency of the design feature appearance, and ISF (Inverse SUV frequency) is the inverse frequency of the overall vehicle models.

[0140] ISF (Inverse SUV Frequency), the inverse vehicle model frequency, represents the reciprocal of the number of vehicle models containing this feature. This indicator is usually used to measure the particularity of a feature in the overall vehicle model set. That is to say, if a certain feature appears in more vehicle models, then the inverse vehicle model frequency will be relatively small; if a certain feature only appears in some vehicle models, the inverse feature frequency will be high, indicating that this feature has a certain importance. To avoid the situation where a certain feature does not appear at all, resulting in a zero derivative, 1 is added to the denominator.

[0141] Cluster each vehicle model according to the proximity of features to find groups with similar features. Set up a spatial model in corresponding dimensions for the important morphological features of the front part of SUVs. Based on the analysis results of the importance of each feature for the vehicle model; for example, features such as the shape between headlights (grille), the overall shape of the front view, the front face shape, and the headlight shape have relatively high importance in the front view, and a spatial model of the main body can be established according to relevant indicators. Cluster SUV models according to the proximity of features to find groups with similar features. For the analysis of the side shape of SUVs, the importance order of the side view morphological features of SUV vehicles is: the overall shape of the side view, the waistline shape, the front side projection shape, the rear side projection shape, the side window shape, the bottom skirt shape, and the wheel hub shape; preferentially select the morphological indicators with higher importance as the dimensions for establishing the SUV side view shape space.

[0142] Taking the morphological feature of the shape between headlights (grille) as an example, it mainly shows three forms: The first form is relatively independent, with an echo in style. The transition feature between the lamps of traditional energy vehicles is manifested as the grille structure, which has an air intake function and is often used as a style element in design; The second form is a trend of integrated design between the front headlights. The integrated style can be transitioned by decorative parts, and some design forms show that the lamps and the middle transition parts form a whole; The third form is to arrange the headlights and the middle transition parts to form a characteristic combination of parallel up and down, with a sense of hierarchy and beauty. The above three types are used as typical clustering features in the analysis process.

[0143] Cluster other relevant key features in turn. After multi-level clustering, a clustering dendrogram is formed. The features of the front part of SUVs are roughly divided into six types: (1) The first type (FG01): The main components in the front part are independently distributed; (2) The second type (FG02): The transition feature between the front headlights of SUVs shows an integrated trend, but the components remain independent of each other; (3) The third type (FG03): A segmented combination feature is formed between the front headlights and the middle transition feature, emphasizing the through design of the front part; (4) The fourth type (FG04): The front lights and the middle transition parts form an integrated form, with a strong sense of integrity; (5) The fifth type (FG05): There is a transitional connection between the front headlights, and the bumper grille and the front headlight transition parts form an integrated form with a unified style, with a sense of impact; (6) The sixth type (FG06): The upper headlight transition part and the grille of the bumper are integrated into one, forming a relatively large overall grille. Thus, the clustering of the front part shape of relevant SUV models is completed.

[0144] The effects of the above technical solution are as follows: Through the vector space model, the features of morphological elements are transformed into quantifiable numerical representations, which greatly improves the quantization accuracy of morphological features. The weight value of each feature is calculated based on its frequency of occurrence in multiple vehicle models and the inverse frequency of the overall vehicle model, taking into account both the universality of the feature and its uniqueness in specific vehicle models.

[0145] The space model constructed based on the vector space model provides an accurate data basis for cluster analysis. Cluster analysis can group the morphologies of multiple vehicle models from various perspectives according to the similarity of morphological features to form a clustering model, which helps designers identify vehicle models with similar morphological features, thus making it easier to understand and analyze the types and trends of automotive morphologies.

[0146] The set of automotive morphology types obtained through cluster analysis provides designers with rich morphological references and inspiration sources. Designers can make targeted design improvements and innovations based on the morphology types and features in the clustering model. This helps optimize the design decision-making process and improve design efficiency and accuracy.

[0147] Cluster analysis not only reveals the morphological similarities between existing vehicle models but also provides possibilities for the innovation of automotive morphologies. Designers can discover new morphological features and trends by analyzing the blank areas or outliers in the clustering model, thus promoting the innovation and diverse development of automotive morphologies.

[0148] In some embodiments, constructing an association model between automotive morphology and the vocabulary of user visual perception images through the set of user visual perception images and automotive morphology types includes:

[0149] Establishing the association relationships between morphological elements and perception image phrases from multiple angles respectively; obtaining an evaluation structure relationship diagram based on the upper-level images and lower-level specific morphologies of automotive morphologies.

[0150] The working principle of the above technical solution is as follows: The specific implementation process is as follows:

[0151] In the first stage, formulate an interview plan to determine the interviewees, time, location, and interview content.

[0152] In the second stage, according to the plan, select the test subjects (20 males and 20 females, the age of the test subjects is between 25 and 40 years old, and the occupations of the test subjects include: industrial designers, car salespersons, corporate employees, and school teachers). Conduct independent interviews with each test subject, and the interview time is 20 - 30 minutes. To avoid external interference, the interview location is an independent space.

[0153] In the third stage, exterior photos of different angles of different models of SUVs were used in the experiment. To give the test subjects a clearer visual experience, all vehicle photos were taken from the five angles selected previously. The background of the photos was processed to be light-colored using computer software, and the test was conducted in combination with the SUV sample image semantic evaluation form.

[0154] In the fourth stage, based on their intuitive feelings about the photos, the test subjects selected the models that were attractive to them. Regarding the models selected by the test subjects, they were asked about the reasons for their choices. These reasons for choice were the initial evaluations. Among these initial evaluations, the importance of each initial evaluation was determined by the number of times it was selected. Then, further in-depth inquiries were made about the reasons for the test subjects' recognition of the model to confirm the specific form, which was the "lower-level evaluation items". Further questions were asked about the abstract meaning behind the initial evaluations of the test subjects, and the "upper-level evaluation items" (image semantics) would be obtained. In the fifth stage, the relevant results of the evaluation structure diagram were analyzed. The KJ method was used to summarize and combine similar evaluation results, and the overall evaluation structure diagram was sorted out. Combining with the relevant perceptual vocabulary of SUV models, the phrases summarized previously included: "individual", "innovative", "elegant", "simple", and "technological", etc. These words became the upper-level abstract content in the experiment. The KJ method was used to classify the lower-level SUV morphological features, and the specific morphological factors of the product were formed in combination with the feedback from users. Taking the front morphology of SUVs as an example, it included morphological elements such as the overall front morphology, the overall front face morphology, the headlight morphology, the shape morphology between the headlights (grille), the front bumper morphology, and the front fog lamp (combination) morphology, etc. After clustering, 4-6 specific morphological types included in different morphological elements formed the lower-level specific elements. After completing the evaluation structure diagram, starting from the upper-level perceptual image, the specific lower-level features that generated this perception could be found. In the evaluation structure diagram, there were 17 specific morphological features that matched the upper-level word "innovative", indicating that these 17 specific morphological features were consistent with the "innovative" perceptual image in the users' perceptual images. Thus, all the SUV morphological features corresponding to the perceptual image phrases could be listed.

[0155] The effects of the above technical solutions are as follows: By constructing a correlation model between the vehicle morphology and the users' visual perceptual image vocabulary, designers can more accurately understand the preferences and perceptual needs of users for vehicle morphology. This helps to incorporate the real feelings of users into the product design stage, improving the market acceptance and user satisfaction of the product. Through interviews, tests, and analyses, a set of evaluation systems based on user perception was established. This system not only includes the specific evaluations of users on vehicle morphology but also deeply explores the abstract images behind these evaluations, providing a more comprehensive and in-depth perspective for product evaluation.

[0156] By using tools such as the KJ method to summarize and organize the evaluation results, the main needs and preferences of users for the product form can be quickly extracted. This helps designers quickly locate the design direction during the product development process, reduce the number of design iterations, and improve the product development efficiency. Through interactive methods such as interviews and tests, users are directly involved in the product design and evaluation process. This can not only enhance users' sense of identity and belonging to the product, but also provide designers with more real and rich user feedback.

[0157] An embodiment of the present application further provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of any one of the methods in the embodiments of the present application or the functions of the device as described in the embodiments of the present application.

[0158] An embodiment of the present application further provides a computer-readable storage medium for storing a computer program. When the computer program is executed, it implements the steps of any one of the methods in the embodiments of the present application. Its specific implementation manner is consistent with the implementation manner and the achieved technical effects described in the above method embodiments, and some content will not be repeated.

[0159] In the present application, the readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. The program product can adopt any combination of one or more readable media. The readable media can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0160] A computer-readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a readable program code is carried. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium may also be any readable medium that can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber cable, RF, etc., or any suitable combination of the above. The program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the C language or similar programming languages. The program code may be executed entirely on the user computing device, partially on an associated device, executed as a stand-alone software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., by using an Internet service provider to connect through the Internet).

[0161] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention without departing from the principles and spirit of the present invention, and all such changes should fall within the scope of protection of the claims of the present invention.

Claims

1. An automotive form design method, characterized in that, The method includes: Obtaining a perceptual image association model of the vehicle form; Determining the vehicle form through the mapping relationship between the user's visual perceptual image and the vehicle form characteristic factors in the perceptual image association model, and obtaining multiple design schemes; Screening the design schemes through the emotional vocabulary expected by the user group to obtain the screened design schemes; Performing detail divergence and convergence on the screened design schemes to determine the final design scheme.

2. The design method according to claim 1, wherein The obtaining of the perceptual image association model of the vehicle form includes: Obtaining a set of vehicle form types through the form characteristic factors of each form element of the vehicle; Obtaining the semantic evaluation of the form of each form element of different vehicle models by the user to determine the user's visual perceptual image; Constructing an association model between the vehicle form and the user's visual perceptual image vocabulary through the user's visual perceptual image and the set of vehicle form types.

3. The design method according to claim 2, characterized in that, The obtaining of a set of vehicle form types through the form characteristic factors of each form element of the vehicle includes: Obtaining the form characteristics of each form element of different vehicle models from different perspectives; Processing the form characteristics to obtain multiple form characteristic indexes of each form element, and establishing a form characteristic index system through the multiple form characteristic indexes; Analyzing the form characteristics of each form element to obtain the importance ranking of the form characteristics; Obtaining the form elements that need key attention according to the importance ranking of the form characteristics; Processing the form characteristics of the form elements that need key attention by using a vector space model, and forming a set of vehicle form types based on form elements through cluster analysis.

4. The design method according to claim 3, characterized in that The analyzing of the form characteristics of each form element to obtain the importance ranking of the form characteristics includes: Obtaining the importance ranking of the form characteristics of each form element based on an eye movement tracking experiment as the first ranking by dividing the area of interest; Obtaining the subjective comprehensive weight ranking of the form characteristics of each form element; Fusing the first ranking and the subjective comprehensive weight ranking to obtain the final importance ranking of each form element.

5. The design method according to claim 4, wherein The obtaining of the importance ranking of the form characteristics of each form element based on an eye movement tracking experiment as the first ranking includes: Respectively dividing the area of interest of the vehicle form from each perspective; obtaining the eye movement experiment data of the subject staying in each area of interest; Respectively calculating the mean value of each eye movement experiment data of the subject in each area of interest when observing multiple vehicle models; and sorting the mean values to obtain the first sub-ranking of the eye movement data of this item; Obtaining the comprehensive ranking of each area of interest through the first sub-rankings of multiple eye movement data items; Determining the first ranking according to the comprehensive ranking of each area of interest and the form element of this area of interest.

6. The design method according to claim 4, characterized in that The obtaining of the subjective comprehensive weight ranking of the form characteristics of each form element includes: Obtaining each hierarchical index of the vehicle form through the analytic hierarchy process; where the hierarchy includes an objective layer, a criterion layer, and a sub-criterion layer; the criterion layer includes front form characteristics, side form characteristics, and rear form characteristics; the sub-criterion layer corresponds to the area of interest one by one; Sorting the comprehensive weights of each sub-criterion layer to obtain the comprehensive weight ranking of each form element.

7. The design method according to claim 3, characterized in that, The morphological features of the key morphological elements are formalized using the vector space model, and through cluster analysis, a set of automotive morphological types based on morphological elements is formed, including: For a certain key morphological element, multiple features of this morphological element in multiple vehicle models are obtained; If the frequency of a certain feature in the vehicle models is within a preset range, then this feature is used as an input feature option for the space model; The weight value of this feature is obtained through the frequency of the input feature in the vehicle models and the inverse frequency of the overall vehicle models; Based on the weight value of the feature, a space model of the morphological features of this morphological element is constructed; Based on the space model, cluster analysis is performed on the morphologies of multiple vehicle models from each perspective, and a cluster model of automotive morphologies from each perspective is obtained.

8. The design method according to claim 2, characterized in that, The association model between automotive morphology and the user's visual perception image vocabulary is constructed through the user's visual perception image and the set of automotive morphological types, including: The association relationships between morphological elements and perception image phrases are established respectively from multiple angles; an evaluation structure relationship diagram based on the upper-level image and lower-level specific morphology of automotive morphology is obtained.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of any one of claims 1-8 are implemented.

10. A computer-readable storage medium, characterized in that The storage medium stores computer instructions. When the computer reads the computer instructions, the computer executes the steps of any one of claims 1-8.

Images