目的 解决现有产品设计评价方法对用户情感偏好分析不够充分、时效性和客观性不足的问题,从感性工学出发,以叉车产品的感性评价为例,提出基于计算机视觉的产品情感设计评价方法。方法 首先,通过网络爬虫获取产品样本图库,基于感性工学（KE）,建立产品感性意象调研问卷并展开实验研究,得到产品感性偏好评价数据集;其次,基于深度残差学习网络（Resnet）构建产品情感设计智能评价系统KE-Resnet101,将设计评价问题转化为设计方案偏好评价问题;最后,通过实例研究深入分析和验证模型的有效性,并分别建立Resnet50与Resnet152模型进行对比验证。结论 KE-Resnet101智能评价系统有较好的准确性与鲁棒性,能够快速优选出用户情感偏好度最高的产品设计方案,提升产品评价的效率、客观性和准确度。
Abstract
To address the limitations in current product design evaluation methods, such as insufficient consideration of users' emotional preferences, lack of timeliness, and objectivity, the work aims to propose a computer vision-based evaluation method for emotional design from the perspective of Kansei Engineering (KE) by taking forklift product design as an example. Firstly, a product sample library was created with web scraping techniques. Based on KE principles, a perceptual image survey was designed and experimental research was conducted to establish a dataset for evaluating users' perceptual preferences. Next, based on the deep residual network (ResNet), an emotional preference evaluation model, KE-Resnet101, was constructed, converting the design evaluation task into an assessment of emotional preferences for design solutions. Finally, a case study was carried out to rigorously examine and validate the effectiveness of the model, with additional comparative analyses by ResNet50 and ResNet152 models. The KE-Resnet101 model achieved high accuracy and robustness, enabling rapid identification of the product design that aligned most closely with users' emotional preferences, thereby enhancing evaluation efficiency, objectivity, and accuracy.
关键词
感性工学 /
情感设计 /
设计评价 /
深度学习 /
计算机视觉
Key words
Kansei engineering /
emotional design /
design evaluation /
deep learning /
computer vision
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] GANN D M, SALTER A J.Innovation in Project-Based, Service-Enhanced Firms: The Construction of Complex Products and Systems[J]. Research Policy, 2000, 29(7/8): 955-972.
[2] LIN H, LUO S J, YING F T, et al.A Study on the Perception of Wireless Headphone Form Design Based on Kansei Engineering[M]//Advances in Affective and Pleasurable Design. Cham: Springer International Publishing, 2019: 369-380.
[3] PARK J, GUNN F, LEE Y, et al.Consumer Acceptance of a Revolutionary Technology-Driven Product: The Role of Adoption in the Industrial Design Development[J]. Journal of Retailing and Consumer Services, 2015, 26: 115-124.
[4] 李彦龙, 蔡谦, 孙久康, 等. 基于BP神经网络的汽车外观设计评价方法[J]. 同济大学学报(自然科学版), 2021, 49(1): 116-123.
LI Y L, CAI Q, SUN J K, et al.An Evaluation Method of Vehicle Exterior Design Based on BP Neural Network[J]. Journal of Tongji University (Natural Science), 2021, 49(1): 116-123.
[5] NAGAMACHI M.Kansei Engineering: A New Ergonomic Consumer-Oriented Technology for Product Development[J]. International Journal of Industrial Ergonomics, 1995, 15(1): 3-11.
[6] 吴天宇, 赵祎乾, 李亚军, 等. 基于混合感性工学模型的产品色彩设计方法[J]. 机械设计, 2021, 38(7): 140-144.
WU T Y, ZHAO Y Q, LI Y J, et al.Design Method of Product Color Based on Hybrid Kansei Engineering Model[J]. Journal of Machine Design, 2021, 38(7): 140-144.
[7] DUAN J J, LUO P S, LIU Q, et al.A Modeling Design Method for Complex Products Based on LSTM Neural Network and Kansei Engineering[J]. Applied Sciences, 2023, 13(2): 710.
[8] CASTILLA N, LLINARES C, BISEGNA F, et al.Emotional Evaluation of Lighting in University Classrooms: A Preliminary Study[J]. Frontiers of Architectural Research, 2018, 7(4): 600-609.
[9] CASTILLA N, LLINARES C, BISEGNA F, et al.Affective Evaluation of the Luminous Environment in University Classrooms[J]. Journal of Environmental Psychology, 2018, 58: 52-62.
[10] ZUO Y X, WANG Z Y.Subjective Product Evaluation System Based on Kansei Engineering and Analytic Hierarchy Process[J]. Symmetry, 2020, 12(8): 1340.
[11] 杨智渊, 杨文波, 杨光, 等. 人工智能赋能的设计评价方法研究与应用[J]. 包装工程, 2021, 42(18): 24-34.
YANG Z Y, YANG W B, YANG G, et al.Research and Application of Artificial Intelligent Empowered Design Evaluation Method[J]. Packaging Engineering, 2021, 42(18): 24-34.
[12] CHEN H Y, YANG C C, KO Y T, et al.Product Form Feature Selection Methodology Based on Numerical Definition-Based Design[J]. Concurrent Engineering, 2014, 22(3): 183-196.
[13] LU P, HSIAO S W, WU F.A Product Shape Design and Evaluation Model Based on Morphology Preference and Macroscopic Shape Information[J]. Entropy, 2021, 23(6): 639.
[14] 段金娟, 宣艾祺, 袁博, 等. 基于感性意象的并条机造型设计[J]. 纺织学报, 2022, 43(4): 160-166.
DUAN J J, XUAN A Q, YUAN B, et al.Drawing Frame Modeling Design Based on Kansei Image[J]. Journal of Textile Research, 2022, 43(4): 160-166.
[15] LAI H H, LIN Y C, YEH C H.Form Design of Product Image Using Grey Relational Analysis and Neural Network Models[J]. Computers & Operations Research, 2005, 32(10): 2689-2711.
[16] ZHAO W G, LI Q, WANG L, et al.Research on Evaluation Method of Product Style Semantics Based on Neural Network[J]. Research Journal of Applied Sciences, Engineering and Technology, 2013, 6(23): 4330-4335.
[17] LI Z R, SHI K, DEY N, et al.Rule-Based back Propagation Neural Networks for Various Precision Rough Set Presented KANSEI Knowledge Prediction: A Case Study on Shoe Product Form Features Extraction[J]. Neural Computing and Applications, 2017, 28(3): 613-630.
[18] 林丽, 张云鹍. 感性工学中的参数化设计方法关键技术研究现状与展望[J]. 图学学报, 2019, 40(5): 936-944.
LIN L, ZHANG Y K.Key Technologies in Parametric Design Methods in Kansei Engineering: State of Art and Progress[J]. Journal of Graphics, 2019, 40(5): 936-944.
[19] TANG C Y, FUNG K Y, LEE E W M, et al. Product Form Design Using Customer Perception Evaluation by a Combined Superellipse Fitting and ANN Approach[J]. Advanced Engineering Informatics, 2013, 27(3): 386-394.
[20] LI X, SU J N, ZHANG Z P, et al.Product Innovation Concept Generation Based on Deep Learning and Kansei Engineering[J]. Journal of Engineering Design, 2021, 32(10): 559-589.
[21] DENG L, YU D.Deep Learning: Methods and Applications[J]. Foundations and Trends® in Signal Processing, 2014, 7(3-4): 197-387.
[22] LIN S L.Application Combining VMD and ResNet101 in Intelligent Diagnosis of Motor Faults[J]. Sensors, 2021, 21(18): 6065.
[23] PRABHAKAR M, PURUSHOTHAMAN R, AWASTHI D P.Deep Learning Based Assessment of Disease Severity for Early Blight in Tomato Crop[J]. Multimedia Tools and Applications, 2020, 79(39): 28773-28784.
[24] OSGOOD C E .The Measurement of Meaning[J]. Audio-Visual Communication Review, 1957, 2(7): 503-4.
[25] HE K M, ZHANG X Y, REN S Q, et al.Deep Residual Learning for Image Recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE, 2016: 770-778.
[26] 于童安. 基于影像反求与人机尺寸分析的叉车造型设计研究[D]. 西安: 西安科技大学, 2021.
YU T A.Research on Forklift Modeling Design Based on Image Reverse Engineering and Man-Machine Dimension Analysis[D]. Xi'an: Xi'an University of Science and Technology, 2021.
[27] SZEGEDY C, LIU W, JIA Y Q, et al.Going Deeper with Convolutions[C]// 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston: IEEE, 2015: 1-9.
[28] ZHANG C L, BAO N, SUN H, et al.A Deep Learning Image Data Augmentation Method for Single Tumor Segmentation[J]. Frontiers in Oncology, 2022, 12: 782988.
[29] OKAFOR E, SCHOMAKER L, WIERING M A.An Analysis of Rotation Matrix and Colour Constancy Data Augmentation in Classifying Images of Animals[J]. Journal of Information and Telecommunication, 2018, 2(4): 465-491.
基金
2024年中央高校基本科研业务费资助项目（JBKYKJCX2024-9）; 中国残联课题残疾人辅助器具专项（2024CDPFAT-05）; 重庆市职业教育教学改革研究项目（Z231015）
PDF(2254 KB)
渝公网安备 50010702501717号