| 基于眼动跟踪特征的汽车造型评价方法 |
| 卢兆麟1,2, 李升波1, 徐少兵1, 成波1 |
| 1. 清华大学 汽车工程系, 汽车节能与安全国家重点实验室, 北京 100084, 中国; 2. 慕尼黑工业大学 工业设计系, 慕尼黑 80805, 德国 |
| Automobile style evaluation based on eye tracking |
| LU Zhaolin1,2, LI Shengbo1, XU Shaobing1, CHENG Bo1 |
| 1. State Key Laboratory of Automotive Safety and Energy, Department of Automotive Engineering, Tsinghua University, Beijing 100084, China; 2. Chair of Industrial Design, Technical University of Munich, Munich 80805, Germany |
摘要:
| ||||||
| 摘要为了弥补主观性评价方法的不足, 以获得用户对于汽车造型设计的客观性评价, 提出了一种基于眼动跟踪特征的汽车造型评价方法。以眼动热点图为基础构建车型评价指标体系, 包括对热点图进行背景剥离, 通过“色度-饱和度-亮度(HSV)”模型提取其颜色特征, 并计算其灰度直方图的各项统计值;进一步将汽车造型的评价定义为二分类问题, 运用Fisher判别建立函数模型。通过实例验证了该方法的可靠性与实用性。结果表明: 该方法总体上具有较高准确度, 为实现汽车造型评价提供了一种可行的技术途径。 | ||||||
| 关键词 ：汽车造型,评价方法,眼动跟踪技术,特征提取,判别函数 | ||||||
| Abstract：Traditional subjective evaluation methods have limited success for automotive designs. An automobile style evaluation method based on eye tracking is developed to more accurately recognize the users' objective assessment of the automotive form design. An index system is structured based on a heat map including striping the background, extracting the color characteristics using a hue-saturation-value (HSV) model and calculating the statistics of the grey level histogram. The automobile form evaluation is defined as a binary classification model with the function model built from a Fisher discriminate. The reliability and practicality of this method are validated using a test case. The results show that this method is accurate and a feasible technical approach for rapid evaluation of automotive styling. | ||||||
| Key words：auto formevaluation methodeye trackingfeature extractiondiscriminate function | ||||||
| 收稿日期: 2014-10-27 出版日期: 2015-09-18 | ||||||
| ||||||
| 通讯作者:成波,教授,E-mail:chengbo@tsinghua.edu.cnE-mail: chengbo@tsinghua.edu.cn | ||||||
| 引用本文: |
| 卢兆麟, 李升波, 徐少兵, 成波. 基于眼动跟踪特征的汽车造型评价方法[J]. 清华大学学报（自然科学版）, 2015, 55(7): 775-781. LU Zhaolin, LI Shengbo, XU Shaobing, CHENG Bo. Automobile style evaluation based on eye tracking. Journal of Tsinghua University(Science and Technology), 2015, 55(7): 775-781. |
| 链接本文: |
| http://jst.tsinghuajournals.com/CN/或 http://jst.tsinghuajournals.com/CN/Y2015/V55/I7/775 |
图表:
 |
| 表1 基于灰度直方图的统计指标 |
 |
| 图1 基于眼动跟踪特征的汽车造型评价方法流程 |
 |
| 图2 训练样本 |
 |
| 图3 实验场景 |
 |
| 图4 训练样本的眼动热点图 |
 |
| 图5 眼动热点图处理 |
 |
| 表1 样本No．1的分析结果 |
 |
| 表4 样本No．2的分析结果 |
 |
| 图6 其中一个验证样本 |
 |
| 表5 眼动热点图与Likert量表结果对比 |
参考文献:
| [1] 中国汽车工业协会, 大众汽车集团(中国). 中国汽车产业发展报告: 2013 [M]. 北京: 社会科学文献出版社, 2013. China Association of Automobile Manufactures, Volkswagen China. Annual Report on Automotive Industry in China: 2013 [M]. Beijing: Social Science Academic Press, 2013. (in Chinese) [2] 黄琦, 孙守迁. 基于意象认知模型的汽车草图设计技术研究 [J]. 浙江大学学报: 工学版, 2006, 40(4): 553-559. HUANG Qi, SUN Shouqian. Research on automobile sketch design based on image cognition model [J]. Journal of Zhejiang University: Engineering Science, 2006, 40(4): 553-559. (in Chinese) [3] Yadav H C, Jain R, Sandarbh S, et al. Prioritization of aesthetic attributes of car profile [J]. International Journal of Industrial Ergonomics, 2013, 43: 297-303. [4] 胡伟峰, 赵江洪. 用户期望意象驱动的汽车造型基因进化 [J]. 机械工程学报, 2011, 47(16): 176-181. HU Weifeng, ZHAO Jianghong. Automobile styling gene evolution driven by users expectation image [J]. Journal of Mechanical Engineering, 2011, 47(16): 176-181. (in Chinese) [5] 胡伟峰, 赵江洪, 赵丹华. 基于造型特征线的汽车造型意象研究 [J]. 中国机械工程, 2009, 20(4): 496-500. HU Weifeng, ZHAO Jianghong, ZHAO Danhua. Study on styling image of vehicle based on form feature lines [J]. China Mechanical Engineering, 2009, 20(4): 496-500. (in Chinese) [6] Hsiao S W. Fuzzy set theory applied to car style design [J]. International Journal of Vehicle Design, 1994, 15(3): 255-278. [7] Hsiao S W. Fuzzy set theory on car-color design [J]. Color Research and Application, 1994, 19(3): 202-213. [8] Hsiao S W, Chan G M S. A semantic recognition based approach for car's concept design [J]. International Journal of Vehicle Design, 1997, 18(1): 53-82. [9] McCormack J P, Cagan J, Vogel C M. Speaking the Buick language: Capturing, understanding, and exploring brand identity with shape grammars [J]. Design Studies, 2004, 25: 1-29. [10] Lai H H, Chang Y M, Chang H C. A robust design approach for enhancing the feeling quality of a product: A car profile case study [J]. International Journal of Industrial Ergonomics, 2005, 35: 445-460. [11] Coughlan P, Mashman R. Once is not enough: Repeated exposure to and aesthetic evaluation of an automobile design prototype [J]. Design Studies, 1999, 20: 553-563. [12] LUO Shijian, FU Yetao, ZHOU Yuxiao. Perceptual matching of shape design style between wheel hub and car type [J]. International Journal of Industrial Ergonomics, 2012, 42: 90-102. [13] Hyoenae J, Radach R, Deubel H. The Mind's Eye: Cognitive and Applied Aspects of Eye Movement Research [M]. Amsterdam, Netherlands: Elsevier Science BV, 2003: 13-15. [14] Djamasbi S, Siegel M, Tullis T, et al. Efficiency, trust, and visual appeal: Usability testing through eye tracking [C]//Proceedings of the 43rd Hawaii International Conference on System Sciences. Big Island, HI: IEEE Computer Society, 2010: 1-10. [15] Andrienko G, Andrienko N, Burch M, et al. Visual analytics methodology for eye movement studies [J]. IEEE Transactions on Visualization and Computer Graphics, 2012, 18(12): 2889-2898. [16] 杨淑莹. 模式识别与智能计算 [M]. 北京: 电子工业出版社, 2011: 105.YANG Shuying. Pattern Recognition and Intelligent Computing [M]. Beijing: Publishing House of Electric Industry, 2011: 105. (in Chinese) [17] 卢兆麟, 汤文成, 薛澄岐. 一种基于形状文法的产品设计DNA 推理方法 [J]. 东南大学学报: 自然科学版, 2010, 40 (4): 704-711. LU Zhaolin, TANG Wencheng, XUE Chengqi. A method of design DNA reasoning based on shape grammar [J]. Journal of Southeast University: Natural Science Edition, 2010, 40(4): 704-711. (in Chinese) |
相关文章:
|
