测绘学报 ›› 2018, Vol. 47 ›› Issue (8): 1114-1122.doi: 10.11947/j.AGCS.2018.20180107

• 虚拟地理环境 • 上一篇    下一篇

虚拟地理试验的地标视觉显著度模型

贾奋励1, 田江鹏1, 智梅霞1, 陈令羽1, 杨清丽2   

  1. 1. 信息工程大学地理空间信息学院, 河南 郑州 450052;
    2. 61287部队, 甘肃 兰州 730020
  • 收稿日期:2018-03-16 修回日期:2018-06-10 出版日期:2018-08-20 发布日期:2018-08-22
  • 通讯作者: 田江鹏 E-mail:tjpeng2011@163.com
  • 作者简介:贾奋励(1976-),女,博士,副教授,研究方向为空间认知与虚拟地理环境、全息地图。E-mail:imjerry76@foxmail.com
  • 基金资助:
    国家自然科学基金(41671407;41701457;41371382);国家重点研发计划(2017YFB0503500)

A Visual Salience Model of Landmark Based on Virtual Geographical Experiments

JIA Fenli1, TIAN Jiangpeng1, ZHI Meixia1, CHEN Lingyu1, YANG Qingli2   

  1. 1. Institute of Surveying and Mapping, Information Engineering University, Zhengzhou 450052, China;
    2. No. 61287 Troops of PLA, Lanzhou 730020, China
  • Received:2018-03-16 Revised:2018-06-10 Online:2018-08-20 Published:2018-08-22
  • Supported by:
    The National Natural Science Foundation of China (Nos. 41671407;41701457;41371382);The National Key Research and Development Program of China (No. 2017YFB0503500)

摘要: 针对地标显著度计算存在的因子不全、赋权主观的问题,提出了一种利用虚拟地理试验构建地标显著度模型的方法。以城市地标显著度模型求解为例,分析了地标视觉显著度影响因子(简称为视觉因子)及计算方法,构建了一种综合眼动追踪、视觉因子可控和可计算的虚拟试验环境。开展了眼动指标与地标显著度之间的关系求解试验,建立起二者的定量关系模型。以该模型为基础,开展了地标视觉显著度与视觉因子间定量关系的求解试验,通过试验数据拟合得到最终的地标视觉显著度模型公式。本文方法的核心是:在视觉因子可控可计算的虚拟环境中,使用眼动指标表征显著度值,通过地标认知试验及数据分析拟合得到显著度模型。该方法对于研究地标判别这类人与环境相互作用机制的问题具有独特的优势,可降低研究的复杂度,克服研究结果的难复现性。

关键词: 虚拟地理试验, 地标, 视觉显著度, 眼动

Abstract: In this paper,a visual salience model of landmark based on virtual geography experiments is proposed for landmark discrimination and salience calculation.Taking the calculation of salience model of city buildings as an example,visual factors of landmarks and their calculation are first analyzed,and then a virtual experiment environment is developed to incorporate eye tracking as well as controllable and computable visual factors.Furthermore,a quantitative model of the relationship between eye movement factors and visual salience of landmark is derived from empirical results.Based on the quantitative model,further experiment is designed to solve the equation of visual salience model of landmark by fitting experimental data,which describes the relationship between a visual salience of landmark and visual factors.The key to the proposed method is to use eye movement factors to represent a salience model in a virtual environment with controllable and computable visual factors,and then the salience model can be calculated through landmark cognition experiments and data analysis.This method can not only reduce the complexity of landmark research,but also overcome the difficulty of repeating the experiments,which makes it a unique advantage to solve problem that tackles interaction mechanism between human and environments such as landmark discrimination.

Key words: virtual geographic experiments, landmark, visual salience model, eye movement

中图分类号: