基于感性需求的安全服装智能设计模型研究

With the rapid development of social economy, people are paying more attention to the importance of safety clothes to protect them in their daily life. There has been a trend to design a safety garment by taking into account both its function and its aesthetic appeal. In the current project, taking the traffic safety school uniform as a case study, the authors aim to develop a consumer oriented intelligent model for the design of safety clothes by using methods and techniques related to sensory analysis, intelligent modelling, database and interactive technologies. In the first place, the design knowledge (including various design elements and rules) about the safety clothes is collected, sorted and denoted in a systematic way to constitute the so-called design knowledge base (DKB). And based on it, a set of safety clothes drawings with variations in every design element and abiding by all the design rules are prepared to constitute the so-called style base (SB). Then, according to the standard methodology of sensory analysis, a perceptual descriptive space and an aesthetic evaluation system are established to study the aesthetic (perceptual) properties of the safety clothes of interest. To be specific, a series of sensory evaluation experiments are designed to obtain, and then to quantify the panelists’ aesthetic perceptions on the safety clothes samples. Due to the inherent uncertainty and imprecision of perceptual data, the current study explores to apply an intelligent pattern recognition method integrating the use of fuzzy inference system (FIS) and ant colony optimization algorithm (ACO) to model the relations between the design elements and the perceptual (aesthetic) properties of the safety clothes samples. A set of experimental data are used to verify the efficacy and robustness of the proposed model. Finally, database and interactive techniques are employed to set up an application system to fulfill all the functions mentioned in our study so as to realize the intelligent and interactive design of safety clothes based on consumers’ personalized aesthetic needs.

随着市场经济的发展，大众安全意识不断提高。兼顾功能与美感成为当前安全服装设计的热点。本项目以中小学交通安全校服为例，探索通过感官分析、智能建模、数据库及计算机交互等技术建立以消费者感性需求为导向的安全服装智能设计模型。本研究首先对安全服装设计相关知识的获取与表达进行深入探讨，建立安全服装设计知识库（包括安全服装设计元素及设计规则），并据此建立安全服装外观数据库。而后，基于感官分析的研究体系，建立安全服装的感性描述空间和感性评价体系。通过一系列感官评价实验获得描述安全服装感性特征的量化数据。由于感性数据所固有的不确定性及不稳定性，探索使用模糊推理系统与蚁群优化算法结合的智能模式识别方法建立安全服装外观设计元素与感性特征间的关系模型，并通过实验验证该模型的效力及鲁棒性。最后，采用数据库及计算机交互技术开发用以实现本模型各项功能的应用系统，实现由感性需求驱动的安全服装智能交互设计。

随着市场经济的发展，大众安全意识不断提高。兼顾功能与美感成为当前安全服装设计的热点。本项目以中小学交通安全校服为例，探索通过感官分析、智能建模、数据库及计算机交互等技术建立以消费者感性需求为导向的安全服装智能设计模型。本研究首先对安全服装设计相关知识的获取与表达进行深入探讨，建立安全服装设计知识库（包括安全服装设计元素及设计规则），并据此建立安全服装外观数据库。而后，基于感官分析的研究体系，建立安全服装的感性描述空间和感性评价体系。通过一系列感官评价实验获得描述安全服装感性特征的量化数据。由于感性数据所固有的不确定性及不稳定性，探索使用模糊推理系统与蚁群优化算法结合的智能模式识别方法建立安全服装外观设计元素与感性特征间的关系模型，并通过实验验证该模型的效力及鲁棒性。最后，采用数据库及计算机交互技术开发用以实现本模型各项功能的应用系统，实现由感性需求驱动的安全服装智能交互设计。