基于眼动赋权及脑电意象认知的产品形态感性工学模型研究

KE Model of Product form Based on Eye-tracking Weighting and Image Cognition by EEG

中文关键词: 感性工学  产品形态  眼动追踪  脑电  感性工学模型  岭回归

英文关键词: Kansei Engineering  product form  eye-tracking  electroencephalogram  KE model  ridge regression

中文摘要:

      目的 为解决传统认知测量主观性强，造成后续构建的感性工学模型可靠性低、客观性弱的问题，提出一种基于眼动赋权及脑电意象认知的产品形态感性工学模型构建方法。方法 确定目标感性意象词及产品对象;基于形态分析法解构产品获取形态特征，通过正交试验法重构产品实验样本;实施意象评价及眼-脑联合的意象认知实验，通过被试者的行为数据分析明确关键意象，根据眼动数据通过变异系数法计算形态特征认知权重，并基于脑电数据的时频分析及ERP的P300成分提取，确定出与意象紧密关联的脑电指标;引入岭回归建立产品形态的感性工学模型。结果 以壁挂式充电桩机身形态进行实例研究，经验证分析，该模型预测结果与意象评估结果一致性高。结论 该模型可客观准确地预测产品形态意象，从而更有效地协助设计师进行感性创新设计。

英文摘要:

      This paper aims to solve the problem that the traditional cognitive measurement has strong subjectivity, which leads to the low reliability and weak objectivity of the subsequent Kansei Engineering (KE) model, thereby proposing a construction method that KE model of product form based on eye-tracking weighting and image cognition by electroencephalogram (EEG)..First, the target perceptual image word and product object were identified. Going forward, based on morphological analysis to deconstruct the product to acquire morphological features, the product experimental sample was reconstructed by orthogonal test methods. Next, the image cognition experiment of image evaluation and eye- tracking combined with EEG was performed, the key image was clear by analyzing testee behavioral data, the coefficient of variation method was used to calculate the cognitive weight of morphological features based on eye movement data, and EEG metrics that correlate closely with imagery were identified by time-frequency analysis of EEG data and P300 component extraction of event-related potential (ERP). Finally, ridge regression was introduced to establish the KE models of product form. The charging piles (wall-mounted) were taken as an example, after verification and analysis, the prediction results of the model are highly consistent with the image evaluation results. In conclusion, the model can objectively and accurately predict the image of product forms to assist designers more effectively in perceptual innovative design.