Lamp product design method based on AIGC and perceptual engineering

Abstract

The invention discloses a lamp product design method based on AIGC and perceptual engineering. According to the technical scheme, the method is characterized by comprising the steps that S1, perceptual vocabularies and samples are collected and screened through big data; s2, making a perceptual questionnaire based on the perceptual vocabularies and the samples, obtaining filling data of the perceptual questionnaire, and importing the filling data into spss for validity analysis; s3, inputting the perceptual vocabularies and the perceptual samples into a Midjouery tool to generate a corresponding virtual finished product graph; the method has the technical effects that the perceptual requirements of users are systematically analyzed by adopting a perceptual engineering method, and new textures are innovatively developed by utilizing an ALGC-assisted industrial design means so as to meet the market requirements and improve the competitiveness of products.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a lamp product design method based on AIGC and Kansei engineering. Background Art

[0002] Artificial Intelligence (AI) is an interdisciplinary and emerging discipline based on computer science, which is an integration of multiple disciplines such as computer science, psychology, and philosophy. It studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence. It attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems.

[0003] The Chinese patent publication number is CN117422002B, an AIGC-based embedded product generation method, system and storage medium, which involves the field of embedded product generation technology. The method includes: according to the embedded product development requirements, the initial product definition is based on the natural language description; the natural language is parsed according to the AI ​​parsing module to obtain the product physical model keywords, and the physical model is parsed to generate a complete product definition; according to the complete product definition, the virtual product generation module is used in combination with the vector database to generate the corresponding hardware, software and appearance data, and a virtual product is generated through a simulator; the virtual product is instantiated, optimized and built to form the final product. Using AIGC for product development, customers can define products according to their own needs, and AIGC will carry out one-stop development and design of the product, so that customers can focus more on the functional realization of the product, thereby improving business competitiveness.

[0004] The design of the products in the above patents lacks systematicity and fails to focus on the user's emotional needs and furniture environment. In addition, existing products are highly homogenized, making it impossible for users to distinguish them based on product features. Summary of the invention

[0005] The purpose of the present invention is to provide a lighting product design method based on AIGC and Kansei Engineering, which has the advantages that the present invention adopts the method of Kansei Engineering to systematically analyze the user's sensory needs, and uses the industrial design means assisted by AIGC to innovatively develop new textures to meet market demand and enhance product competitiveness.

[0006] The above technical objectives of the present invention are achieved through the following technical solutions: A lamp product design method based on AIGC and Kansei Engineering, comprising the following steps: S1: Collect and screen perceptual words and samples through big data; S2: Create a perceptual questionnaire based on perceptual vocabulary and samples, obtain the completed data of the perceptual questionnaire, and import the completed data into SPSS for validity analysis; S3: Input the perceptual vocabulary and perceptual samples into the Midjounery tool to generate the corresponding virtual finished product image.

[0007] The present invention is further configured that step S1 comprises: S11: Collect emotional words from purchase reviews by visiting e-commerce websites; S12: Read literature related to Kansei Engineering and select available vocabulary; S13: Screen the available words according to the similarity of their meanings to obtain the final perceptual words.

[0008] The present invention is further configured that the step S1 further includes: S14: Collect texture samples through design websites; S15: performing size uniformization and decolorization processing on the texture sample photo to obtain a decolorized texture sample; S16: labeling the decolored texture samples to obtain a perceptual sample library.

[0009] The present invention is further configured that step S2 comprises: S21: Verify the authenticity of the filled data through information analysis methods; S22: After validity verification, average value analysis is performed on the filled data; S23: Establish a mapping relationship between perceptual vocabulary and perceptual samples.

[0010] The present invention is further configured that step S3 comprises: S31: Import visual materials such as product prototypes, inspiration references, design sketches, etc. as basic materials into MJ, and submit the images to the MJ robot; S32: Copy the web links of the required images and paste them in the designated area; S33: Add the prompt word of design research and the corresponding suffix after the URL link; S34: Iterate multiple times according to the generated results, optimize the prompt word and suffix weights, and find the most matching prompt word content and weight value.

[0011] The present invention is further configured as follows: the generation of the prompt words in step S33 is based on perceptual vocabulary as the core, and the final prompt words are generated by adding product color, material, product style, camera angle, focal length, and picture detail vocabulary for superposition.

[0012] The present invention is further configured as follows: the suffix in step S33 includes picture weight, output image confusion value and output image stylization.

[0013] The present invention is further configured as follows: the default value of the image weight is 1.5-2, the default value of the output image confusion value is 0-100, and the default value of the output image stylization is 0-1000.

[0014] The present invention is further configured as follows: the step S3 also includes assigning a basic model to the texture sample, inputting it into a three-dimensional software, and generating a three-dimensional rough model.

[0015] In summary, this application starts from the theoretical aspects of sensibility and imagery, and elaborates on the application of sensory engineering in industrial design. The research framework of the present invention is constructed by reading relevant literature. The sensory vocabulary and sensory samples related to ceiling lamps are widely collected through online surveys. Five categories of sensory vocabulary are screened out through the expert group method. The data are recovered through questionnaire surveys. Five categories and ten types of plane texture samples are obtained through analysis and screening of reliability, validity and sample average scores. The draft model of the design is determined through user research. The draft model is combined with the screened texture samples to obtain the base map material of AI raw images. Through AIGC, product generation is effectively realized to meet the different design needs of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of this embodiment. DETAILED DESCRIPTION

[0017] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0018] Example 1: Referring to Figure 1 As shown, a lighting product design method based on AIGC and Kansei Engineering includes the following steps: S1: Collect and screen perceptual words and samples through big data; S2: Create a perceptual questionnaire based on perceptual vocabulary and samples, obtain the completed data of the perceptual questionnaire, and import the completed data into SPSS for validity analysis; S3: Input the perceptual vocabulary and perceptual samples into the Midjounery tool to generate the corresponding virtual finished product image.

[0019] Specifically: Starting from the theoretical aspects of sensibility and imagery, this application elaborates on the application of sensibility engineering in industrial design, constructs the research framework of the present invention by reading relevant literature, collects a wide range of sensibility words and sensibility samples related to ceiling lamps through online surveys, screens out five categories of sensibility words through an expert group method, collects data through questionnaire surveys, obtains five categories and ten kinds of plane texture samples through analysis and screening of reliability, validity and sample average scores, determines the draft model of the design through user research, combines the draft model with the screened texture samples to obtain the base map material of AI raw images, and effectively realizes product generation through AIGC to meet the different design needs of users.

[0020] Further, step S1 includes: S11: Collect emotional words from purchase reviews by visiting e-commerce websites; S12: Read literature related to Kansei Engineering and select available vocabulary; S13: Screen the available words according to the similarity of their meanings to obtain the final perceptual words; S14: Collect texture samples through design websites; S15: performing size uniformization and decolorization processing on the texture sample photo to obtain a decolorized texture sample; S16: labeling the decolored texture samples to obtain a perceptual sample library.

[0021] Further, step S2 includes: S21: Verify the authenticity of the filled data through information analysis methods; S22: After validity verification, average value analysis is performed on the filled data; S23: Establish a mapping relationship between perceptual vocabulary and perceptual samples.

[0022] Further, step S3 includes: S31: Import visual materials such as product prototypes, inspiration references, design sketches, etc. as basic materials into MJ, and submit the images to the MJ robot; S32: Copy the web links of the required images and paste them in the designated area; S33: Add the prompt word of design research and the corresponding suffix after the URL link; S34: Iterate multiple times according to the generated results, optimize the prompt word and suffix weights, and find the most matching prompt word content and weight value.

[0023] Furthermore, the generation of prompt words in step S33 is based on perceptual vocabulary, and the final prompt words are generated by adding product color, material, product style, camera angle, focal length, and picture detail vocabulary for superposition. The suffix in step S33 includes picture weight, output confusion value, and output stylization.

[0024] Furthermore, the default value of the image weight is 1.5-2, the default value of the output chaos value is 0-100, and the default value of the output stylization is 0-1000.

[0025] Furthermore, step S3 also includes assigning a basic model to the texture sample, inputting it into a three-dimensional software, and generating a three-dimensional rough model.

[0026] Embodiment 2: This embodiment takes the generation of an indoor ceiling lamp as an example to further illustrate the technical solution of the present invention.

[0027] 1. Screen the image diagram, including the following: Ten categories of perceptual samples were screened out through the Likert scale method, a large number of design references were obtained through the AI-assisted industrial design method, and 50 design samples in 10 categories were screened through the expert group method. Next, the author further screened each category of samples. Taking the "trendy" sample as an example, samples A1, A2, A3, A4, and A5 were selected, and the samples were scored for their aesthetics and degree of differentiation respectively. The applicant used the evaluation criteria of aesthetics and degree of differentiation to screen the remaining nine groups of texture samples, and determined that the final samples corresponding to the "trendy" perceptual vocabulary were A1 and a2, the final samples corresponding to the "soft" perceptual vocabulary were B2 and b3, the final samples corresponding to the "simple" perceptual vocabulary were C3 and c1, the final samples corresponding to the "elegant" perceptual vocabulary were D2 and d5, and the final samples corresponding to the "modern" perceptual vocabulary were E1 and e5.

[0028] 2. 3D model production, including the following: The wall thickness used in this design study is 2mm. After production, a total of ten types of ceiling lamp white films were obtained. The product will provide three color temperatures of light: 2700K low color temperature, 3500K medium color temperature, and 5000K high color temperature. When the light is on, the product mainly shows its lighting function, and the product texture is not very obvious at this time. In order to more effectively display the decorativeness of the product, this rendering is mainly based on the display in the off-light state. Taking the A1 sample as an example, when the light is off, the product will be used as a decorative object in the room to enhance the level of interior decoration; User research shows that users prefer wood as a decorative material. The author provides two types of wood, walnut and poplar. Each material has its own characteristics and visual effects. Walnut is suitable for creating an elegant and modern atmosphere with its unique color tone. Poplar has a lighter color tone, which conveys a warm and inclusive atmosphere and is suitable for a warm family environment. Both walnut and poplar can enhance the layering and overall beauty of interior decoration. The diameter of the carrier in this design study is 350mm and the height is 55mm. In order to adapt to different indoor spaces (the ceiling lamp in the living room is larger, and the ceiling lamp in the corridor is smaller), the author adjusted the product size, that is, while keeping the height unchanged, the horizontal and vertical dimensions of the product were expanded and reduced. The size of larger spaces such as living rooms was expanded to 500mm, and the size of smaller spaces such as corridors was reduced to 300mm. Through user research, the author found that users prefer regular-shaped products. In addition to the circle discussed above, squares and rectangles are typical regular-shaped graphics. Rectangular products are generally used in living rooms. The length is set to 600mm and the width is set to 350 according to the golden ratio principle. Square products are usually used in medium-sized spaces. The author sets the size to 350mm*350mm and applies the texture of circular products to the new product shape. This design study uses LED light sources. In terms of light source lamp beads, two different design schemes are provided. The first type of lamp beads are evenly distributed, which can produce uniform and soft light effects. It is suitable for scenes that require consistent brightness and no obvious light and shadow changes, such as offices and study areas. The second type of lamp beads has a gradually increasing working area, which can create a gradual light effect from inside to outside, suitable for creating a warm and romantic atmosphere, such as family living rooms, bedrooms and other scenes. Users can choose the appropriate light source according to their preferences and needs to create an ideal light atmosphere for different scenes.

[0029] This specific embodiment is merely an explanation of the present invention and is not a limitation of the present invention. After reading this specification, those skilled in the art may make non-creative modifications to the present embodiment as needed. However, as long as they are within the scope of the claims of the present invention, they are protected by the patent law.

Claims

1. A lighting product design method based on AIGC and Kansei Engineering, characterized in that: The steps include: S1: Collect and screen perceptual words and samples through big data; S2: Create a perceptual questionnaire based on perceptual vocabulary and samples, obtain the completed data of the perceptual questionnaire, and import the completed data into SPSS for validity analysis; S3: Input the perceptual vocabulary and perceptual samples into the Midjounery tool to generate the corresponding virtual finished product image.

2. The lighting product design method based on AIGC and Kansei Engineering according to claim 1, characterized in that: The step S1 comprises: S11: Collect emotional words from purchase reviews by visiting e-commerce websites; S12: Read literature related to Kansei Engineering and select available vocabulary; S13: Screen the available words according to the similarity of their meanings to obtain the final perceptual words.

3. The lighting product design method based on AIGC and Kansei Engineering according to claim 1, characterized in that: The step S1 further comprises: S14: Collect texture samples through design websites; S15: performing size uniformization and decolorization processing on the texture sample photo to obtain a decolorized texture sample; S16: labeling the decolored texture samples to obtain a perceptual sample library.

4. The lighting product design method based on AIGC and Kansei Engineering according to claim 1, characterized in that: The step S2 comprises: S21: Verify the authenticity of the filled data through information analysis methods; S22: After validity verification, average value analysis is performed on the filled data; S23: Establish a mapping relationship between perceptual vocabulary and perceptual samples.

5. The lighting product design method based on AIGC and Kansei Engineering according to claim 1, characterized in that: The step S3 comprises: S31: Import visual materials such as product prototypes, inspiration references, design sketches, etc. as basic materials into MJ, and submit the images to the MJ robot; S32: Copy the web links of the required images and paste them in the designated area; S33: Add the prompt word of design research and the corresponding suffix after the URL link; S34: Iterate multiple times according to the generated results, optimize the prompt word and suffix weights, and find the most matching prompt word content and weight value.

6. The lighting product design method based on AIGC and Kansei Engineering according to claim 5, characterized in that: The generation of prompt words in step S33 is centered on perceptual words, and the final prompt words are generated by adding product color, material, product style, camera angle, focal length, and picture detail words for superposition.

7. The lighting product design method based on AIGC and Kansei Engineering according to claim 5, characterized in that: The suffix in step S33 includes image weight, output image confusion value and output image stylization.

8. The lighting product design method based on AIGC and Kansei Engineering according to claim 7, characterized in that: The default value of the image weight is 1.5-2, the default value of the output image chaos value is 0-100, and the default value of the output image stylization is 0-1000.

9. The lighting product design method based on AIGC and Kansei Engineering according to claim 5, characterized in that: The step S3 also includes assigning a basic model to the texture sample, inputting it into the three-dimensional software, and generating a three-dimensional rough model.

Images