Practical Method of Low-Light-Level Binocular Ranging Based on Triangulation and Error Correction

doi:https://doi.org/10.1145/3168390.3168417

CASIA OpenIR > 综合信息系统研究中心

	Practical Method of Low-Light-Level Binocular Ranging Based on Triangulation and Error Correction
	Qi Shi; Lei Ma; Yiping Yang
	2017-12
会议名称	2017 International Conference on Computer Science and Artificial Intelligence (CSAI 2017)
会议日期	December 5-7, 2017
会议地点	Kalbis Institute, Jakarta, Indonesia
所属项目编号	CXJJ-16M109
摘要	Reliable target distance measurement at long range during the night is crucial for security surveillance at night. We present a practical method for low-light-level ranging using a motionless stereo vision system. The approach can measure the target distance within 200 meters. Our range finding algorithm uses triangulation to reconstruct target point coordinates, instead of precomputing disparity maps. Trilateration method is adopted to calculate median distance of the target. Based on statistical tests, a biexponential-function-based error correction model is proposed. The distance estimation is corrected to final distance after fitting the model. Ranging error can be dramatically reduced by error correction. Extensive experiments on datasets show that our method achieves excellent ranging performance and is stable in repeatability, the mean relative accuracy is below 1%.
关键词	Low-light-level Image Binocular Stereo Vision Triangulation Trilateration Error Correction
DOI	https://doi.org/10.1145/3168390.3168417
URL	查看原文
收录类别	EI
资助项目	National Defense Science and Technology Innovation Fund of the Chinese Academy of Sciences
引用统计
文献类型	会议论文
条目标识符	http://ir.ia.ac.cn/handle/173211/20406
专题	综合信息系统研究中心
通讯作者	Yiping Yang
作者单位	Institution of Automation, Chinese Academy of Sciences
推荐引用方式 GB/T 7714	Qi Shi,Lei Ma,Yiping Yang. Practical Method of Low-Light-Level Binocular Ranging Based on Triangulation and Error Correction[C],2017.

条目包含的文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
Practical Method of （397KB）	会议论文		开放获取	CC BY-NC-SA	浏览请求全文