基于立体视觉与光流融合的运动目标检测
发布时间：2019-06-01 11:35
【摘要】：针对摄像机与被检测目标同时运动时的目标检测问题,提出一种立体视觉与光流融合的运动目标检测算法。结合立体视觉技术设计了光流与自运动估计模型,运用车辆的运动信息和场景的深度信息估计因摄像机运动产生的自运动光流;采用多分辨率细化的Horn算法估计场景的混合光流;对混合光流和自运动光流进行差分运算,剔除背景中静态目标的运动干扰。经过一系列形态学滤波处理获得运动目标完整区域,依据光流的连通性对运动目标标号,并确定位置信息。以典型的交通场景为对象进行分析,实验结果表明该算法能有效地检测出动态背景下的运动目标。
[Abstract]:In order to solve the problem of target detection when the camera and the detected target move at the same time, a moving target detection algorithm based on stereo vision and optical flow fusion is proposed. Combined with stereo vision technology, the optical flow and self-motion estimation model are designed, and the vehicle motion information and scene depth information are used to estimate the self-moving optical flow caused by camera motion. The multi-resolution thinning Horn algorithm is used to estimate the mixed optical flow of the scene, and the difference operation between the mixed optical flow and the self-moving optical flow is carried out to eliminate the moving interference of the static target in the background. After a series of morphological filtering, the complete region of the moving target is obtained, and the moving target is marked according to the connectivity of the optical flow, and the position information is determined. The typical traffic scene is analyzed. The experimental results show that the algorithm can effectively detect the moving target in dynamic background.
【作者单位】： 上海理工大学光电信息与计算机工程学院;
【基金】：国家自然科学基金(61374197) 上海科委科技创新行动计划资助项目(13510502600)
【分类号】：TP391.41
[Abstract]:In order to solve the problem of target detection when the camera and the detected target move at the same time, a moving target detection algorithm based on stereo vision and optical flow fusion is proposed. Combined with stereo vision technology, the optical flow and self-motion estimation model are designed, and the vehicle motion information and scene depth information are used to estimate the self-moving optical flow caused by camera motion. The multi-resolution thinning Horn algorithm is used to estimate the mixed optical flow of the scene, and the difference operation between the mixed optical flow and the self-moving optical flow is carried out to eliminate the moving interference of the static target in the background. After a series of morphological filtering, the complete region of the moving target is obtained, and the moving target is marked according to the connectivity of the optical flow, and the position information is determined. The typical traffic scene is analyzed. The experimental results show that the algorithm can effectively detect the moving target in dynamic background.
【作者单位】： 上海理工大学光电信息与计算机工程学院;
【基金】：国家自然科学基金(61374197) 上海科委科技创新行动计划资助项目(13510502600)
【分类号】：TP391.41
【相似文献】
相关期刊论文 前10条
1 韩雷;王洪庆;林隐静;;光流法在强对流天气临近预报中的应用[J];北京大学学报(自然科学版);2008年05期
2 龚大墉;汪治华;杨数强;刘岩;;基于目标的局部约束光流分析[J];重庆工学院学报(自然科学版);2009年07期
3 严强;黄增喜;曹丽萍;黄蓉刚;;光流法在机车安全行驶中的应用[J];计算机应用研究;2013年04期
4 马鹏飞;杨金孝;;基于光流法的粒子图像测速[J];科学技术与工程;2012年32期
5 张永亮;卢焕章;贺兴华;谢耀华;;基于光流预测的直线对应算法[J];信号处理;2010年05期
6 孙承志;熊田忠;吉顺平;张家海;;基于差分的光流法在目标检测跟踪中的应用[J];机床与液压;2010年14期
7 王，
本文编号：2490243
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2490243.html
