摘要
在指挥大厅场景下,如何让应急指挥人员直接操纵应急GIS的信息展示,是一个亟待解决的问题。本文尝试将动态手势识别技术应用于指挥大厅场景下的GIS人机交互操作,可实现对地理信息的人性化、自然化操纵展示。本文提出了采用卡尔曼滤波优化的动态手势识别改进方法,实验结果表明该改进方法的动态手势算法识别准确率可达94.38%,比未经改进的算法识别准确率提高6.25%。在实时识别方面,利用有限状态自动机的思想进行手势起止检测以提取手势片段,手势识别仅需50 ms,可满足实时交互操作的需求,并实验验证了利用动态手势来控制GIS中地图的放大、缩小、移动等操作。
In the command hall scenario,how to enable emergency commanders to directly manipulate the information display of emergency GIS is an urgent problem to be solved.In this paper,dynamic gesture recognition technology was applied to GIS human-computer interactive operation in command hall scenario,which could realize humanized and naturalized manipulation display of geographic information.This paper proposed an improved dynamic gesture recognition method using Kalman filtering optimization,and the experimental results showed that the recognition accuracy of the improved dynamic gesture algorithm could reach 94.38%,which was 6.25%higher than the recognition accuracy of the unimproved algorithm.In real-time recognition,the gesture start and stop detection was performed by using the idea of finite-state machine to extract gesture fragments,and the gesture recognition only took 50 ms,which could meet the demand of real-time interactive operation.Using dynamic gestures to control operations such as zooming in,zooming out,and moving the map in GIS were implemented.
作者
乔恬
吴铭星
邸国辉
QIAO Tian;WU Mingxing;DI Guohui(School of Geodesy and Geomatics,Wuhan University,Wuhan 430079,China;Hubei Institute of Water Resources Survey and Design,Wuhan 430064,China)
出处
《测绘与空间地理信息》
2025年第2期42-45,49,共5页
Geomatics & Spatial Information Technology
基金
湖北省水利重点科研项目——水利水电工程中倾斜摄影和BIM融合技术的应用研究(HBSLKY202110)资助。
