光学与光电技术, 2018, 16 (3): 74, 网络出版: 2018-07-31
多目标光斑中心提取算法研究
Research on Multi-Target Spot Center Extraction Algorithm
多目标光斑 棋盘法 特征点 高精度 高效率 multi-target spot checkerboard method feature point high precision high efficiency
摘要
针对重载车车头、车尾相对位置微小变化量高精度高效率的检测,提出一组由CMOS相机、正透镜、多点靶标组成的相对位置微小形变量检测系统,并且针对该系统中的多目标光斑中心提取,提出并使用棋盘算法、以及特征点算法实时提取多目标点中心,实现了多目标高精度高效率的提取,且经过大量实验验证了棋盘算法及特征点算法结果的准确性、高效性、稳定性,在复杂背景的外场条件下,棋盘法的稳定性优于特征点法,而在单一背景条件下,特征点法的提取效率远远优于棋盘法,其中棋盘算法已应用于实际测量中,并取得了良好效果。
Abstract
Aiming at the high precision and high efficiency detection of the small change of the relative position of the heavy truck, a set of small deformation detection system of relative position composed of CMOS camera, positive lens and multi-target is proposed. The board algorithm and the feature point algorithm to extract the multi-target point center in real time to achieve multi-target high-precision and high-efficiency extraction are proposed. After a large number of experiments, the accuracy of the checkerboard algorithm and the feature point algorithm is verified The stability of the checkerboard algorithm is better than that of the feature point method under the background conditions of complex background, while the extraction efficiency of the feature point method is far superior to the checkerboard algorithm under the single background condition, in which the checkerboard algorithm has been applied to the actual measurement, and achieved good results.
陈宇, 何平安, 郑仁杰. 多目标光斑中心提取算法研究[J]. 光学与光电技术, 2018, 16(3): 74. CHEN Yu, HE Ping-an, ZHENG Ren-jie. Research on Multi-Target Spot Center Extraction Algorithm[J]. OPTICS & OPTOELECTRONIC TECHNOLOGY, 2018, 16(3): 74.