摘要
为了有效提高用户对视觉统计任务中颜色编码的决策效率,首先需要探究人类视觉系统对界面颜色编码的感知问题。因此,将显著性差异的概念引入到颜色编码研究中,生成一套针对视觉统计任务的色彩空间,并通过典型的视觉统计任务实验,研究颜色差异阈值对视觉统计任务的感知绩效影响,进而总结颜色编码对于视觉统计任务的感知精度和感知规律。实验结果表明:首先,在相关性识别任务和平均值比较任务中,人类视觉系统对颜色的感知精度比以往色彩模型预测的更精确,说明颜色编码时的差异阈值(JND)选择可以更小;其次,当选择不同颜色轴进行颜色编码时,相关性识别任务用明度编码识别绩效会更高;最后,在不同基础相关系数和点群数量条件下,相关性识别精度和平均值比较精度均符合韦伯线性函数,反映了观察者在进行视觉统计任务时处理信息熵的能力。本文为信息可视化设计者提供设计指导,并从心理物理学角度,对可视化界面信息的颜色编码提供了量化依据。
In order to effectively enhance users′decision-making efficiency for color encoding in visual statistics tasks,it is necessary to first investigate the perceptual problem of the human visual system for color encoding in visualization interfaces.Therefore,the concept of significant difference was introduced into the color encoding research,leading to the creation of a set of new color spaces for visual statistics tasks.The effect of color difference threshold on the perceptual performance of visual statistics tasks was then examined through typical visual statistics task experiments,so as to summarize the perceptual accuracy and laws of color coding for visual statistics tasks.The results revealed that:firstly,in the correlation recognition task and the mean comparison task,the perceptual accuracy of the human visual system was more accurate than that predicted by previous color models,indicating that the difference threshold(JND)selection for color encoding could be smaller.Secondly,when different color axes were selected for color encoding,the recognition performance of the correlation recognition task was higher with lightness encoding.Finally,under different basic correlation coefficients and number of point groups,the accuracy of correlation recognition and mean comparison followed the Weber linear function,which reflected the observers′ability to process information entropy when performing visual statistical tasks.The above findings provide design guidance for information visualization designers,and offer a quantitative basis for color coding of visual interface information from the perspective of psychophysics.
作者
郭琪
伍金春
薛澄岐
GUO Qi;WU Jin-chun;XUE Cheng-qi(School of Art Design and Media,East China University of Science and Technology,Shanghai 200000,China;School of Mechanical Engineering,Southeast University,Nanjing Jiangsu 210000,China)
出处
《图学学报》
CSCD
北大核心
2023年第2期408-414,共7页
Journal of Graphics
基金
国家自然科学基金项目(52205264)
上海市浦江人才计划项目(2021PJC032)。
关键词
可视化界面
视觉统计任务
颜色编码
颜色感知
界面评价
information interface
visual statistical task
color encoding
color perception
interface evaluation
