Abstract:

Significance The clothing market has undergone a paradigm transform from mass production to customization and personalization, while consumers are increasingly emphasizing the aesthetic expectations and emotional identity associated with clothing products. Aligning product design and development with the psychological needs of consumers to enhance the emotional value of the product has become a pivotal goal for enterprises. However, human subjective feelings are usually uncertain and ambiguous, posing challenges in quantifying consumers' perceptual preferences and evaluations. Kansei engineering is a design methodology utilizing engineering techniques to quantify human emotions and perceptions, enabling the acquisition of perceptual measurements and establishing the relationship between perceptual and physical attributes. In order to clarify the development and application of Kansei engineering and to master its frontier and development trends in textile and clothing, this paper comprehensively reviews the research progress in Kansei engineering for textile and clothing applications.

Progress Kansei engineering serves as a widely adopted method for quantifying emotions, finding extensive utility in textiles and clothing. Its applications primarily include clothing and fabric design, consumer psychology analysis, clothing product evaluation, and the development of intelligent systems. Initially, the implementation of Kansei engineering relied mainly on the semantic differential method to capture and quantify subjective feelings, and then the correlation between consumers', Kansei information and the objective physical quantities will be established through regression analysis or other methods. However, Kansei information obtained by the semantic differential method is susceptible to various influencing factors, thereby reducing its accuracy. Further, there has been a turn towards combining bioelectrical signals such as electrocardiogram, electroencephalo-gram, and electromyogram, heart rate, eye tracking and other modalities to capture changes in subjective feelings. Additionally, fuzzy mathematical methods have shown promise in addressing the uncertainty and ambiguity of perceptual evaluations. In recent years, remarkable advancements in deep learning techniques have also been witnessed, significantly enhancing the performance of classification and regression tasks. This approach not only facilitates the prediction and interpretation of consumers', aesthetic perceptions but also enables the correlation of user requirements with design factors, thereby assisting designers in achieving innovative designs. Machine learning-based intelligent systems for clothing are personalized, dynamic and have high predictive accuracy, which has become a research hotspot in textile and clothing. Its realization will significantly enhance the emotional attributes of clothing products and the convenience of the clothing market.

Conclusion and Prospect Kansei engineering can effectively facilitate clothing product evaluation and consumer psychology analysis and optimize the clothing product design. Although its application in textiles and clothing is progressively maturing, certain challenges persist. 1) Limited focus on other perceptual modalities. Most research predominantly relies on the semantic differential method to gather Kansei information from vision and tactile. However, integrating diverse Kansei information from multiple modalities (such as olfaction, physical behavior, and physiological perception) can enrich the understanding of user feelings. 2) Homogeneity and small sample size of selected populations. Diversified participants can be representative of a wider consumer group. 3) Overreliance on subjective evaluation. Current perceptual engineering research relies mainly on subjective evaluations, neglecting the potential benefits of incorporating objective methods and engineering technology tools. The combination of subjective and objective data can be better achieved through methods such as physiological measurement, artificial intelligence, and fuzzy mathematics, to deeply explore users'emotional needs and the inherent value of products.

In the future, the multimodal Kansei engineering is expected to be a potential effective means for textile and clothing Kansei study. Complex consumer populations can be categorized by Kansei engineering techniques. Then, emotions, perspectives, behavior, and environmental factors of consumers could be combined to construct a Kansei information database for the consumer group. Additionally, it is crucial to develop Kansei design and recommendation systems for textile and clothing products. The implementation of a Kansei-based virtual fitting system can provide consumers with immersive user collaborative design and personalization recommendation services. Finally, incorporating more advanced artificial intelligence techniques in Kansei engineering research will significantly improve the predictive capabilities of Kansei models on the perceptual preferences of consumers and increase the emotional added value of textile and clothing products.

Key words: Kansei engineering, human factor, Kansei information, textile, clothing