激光与光电子学进展, 2015, 52 (9): 091501, 网络出版: 2015-08-28   

强视差下的移动相机运动目标检测 下载: 714次

Moving Target Detection on Moving Camera with the Presence of Strong Parallax
作者单位
南京理工大学电子工程与光电技术学院, 江苏 南京 210094
摘要
为了降低强视差下移动相机运动目标检测的虚警率,提出了一种基于深度约束方程的运动目标检测方法。重点研究了二维图像坐标系到三维世界坐标系之间的关系,并根据图像深度信息和摄影几何原理,结合前后两帧图像和相机内部参数,提出了深度约束方程。深度约束方程包含了图像灰度信息和深度信息,利用深度约束方程可以有效地去除因相机移动产生的视差对目标检测的影响。利用该方法对实际的图像序列进行了分析与处理,并与相关算法进行了对比分析,实验结果表明该方法能够消除移动相机下由于视差造成的虚警,提高移动平台运动目标检测的准确性。
Abstract
In order to reduce the strong parallax false alarm rate of the moving target detection on a moving camera, a moving target detection method based on depth constraint equation is proposed. The relationship between twodimensional image coordinate and three-dimensional world coordinate is mainly studied. According to the image depth information and photography geometry theory, with the two frames image coordinates and the camera internal parameters, the depth constraint equation is established. The depth constraint equation includes gray value and depth information of image. Using the depth constraint equation to detect moving target can effective avoid the influence caused by moving camera. Experiments are performed based on actual image sequence, and related algorithm is compared and analyzed. Experimental results indicate that the proposed method can avoid the parallax false alarm and improve the precision of moving target detection on a moving camera is improved.

丁祺, 顾国华, 徐富元, 任侃, 钱惟贤, 陈钱. 强视差下的移动相机运动目标检测[J]. 激光与光电子学进展, 2015, 52(9): 091501. Ding Qi, Gu Guohua, Xu Fuyuan, Ren Kan, Qian Weixian, Chen Qian. Moving Target Detection on Moving Camera with the Presence of Strong Parallax[J]. Laser & Optoelectronics Progress, 2015, 52(9): 091501.

本文已被 7 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!