Infrared and Laser Engineering, Volume. 51, Issue 9, 20210596(2022)
Point cloud target recognition algorithm based on stereo vision and feature matching
Figures & Tables(5)
Fig. 3. Comparison before and after using stereo vision optimization
Fig. 4. Comparison of convergence time before and after algorithm optimization
Table 1. Detection and recognition probability
View tableView in ArticleTable 1. Detection and recognition probability
Situation Pa Pc Pt 注：Q1、Q2和Q3分别表示三种测试条件；A、B和C分别表示目标人、I型目标车辆和II型目标车辆；Pa、Pc和Pt分别表示激光雷达内置算法、传统特征匹配算法和文中算法的识别概率 A Q1: B C 92.2% 91.3% 91.6% 98.3% 98.2% 98.6% 98.4% 98.4% 98.7% A Q2: B C 67.5% 66.4% 73.6% 75.6% 74.2% 84.5% 70.7% 71.5% 85.4% A Q3: B C 58.2% 52.3% 69.2% 61.7% 46.2% 73.3% 60.4% 45.7% 71.7%
Tools
Get Citation
Copy Citation Text
Xiaoyan Wang, Gaokui Xu. Point cloud target recognition algorithm based on stereo vision and feature matching[J]. Infrared and Laser Engineering, 2022, 51(9): 20210596
Paper Information
Category: Image processing
Received: Aug. 24, 2021
Accepted: --
Published Online: Jan. 6, 2023
The Author Email: Xu Gaokui (32065203@qq.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20