鱼眼相机导引下的双目跟踪系统

摘要

在智能监控领域,鱼眼相机比传统摄像机拥有更大视野,然而由于其分辨率的限制及鱼眼图像畸变的影响,使得目标物的细节信息难以观察,为克服这个难题,搭建了鱼眼相机导引下的双目跟踪系统.该系统利用鱼眼相机进行全景观察,检测出目标物,采用角度拟合法求出目标物的方位角,通过云台控制双目相机精确定位到目标物,然后对目标物进行距离测量,并使目标在不同的位置进行实验.实验结果表明,系统能准确检测出鱼眼图像中的运动目标,角度标定精度达到系统要求,双目相机能准确测出目标物距离.该智能监控系统能够减小鱼眼相机畸变对检测结果的影响,具有检测视场广和检测精度高的特点,为智能监控技术提供了检测方法.%In the automatic surveillance field,the fish-eye panoramic camera owns larger observation range than traditional camera.However,the details of the objects are not able to be obtained because of low-resolution and distortion images of the fish-eye panoramic camera.To overcome this problem,an automatic surveillance system based on binocular vision guided by the fish-eye camera is set up.Firstly,the fish-eye camera is used to obtain panorama images and detect the target.Secondly,the target azimuth can be calculated by angle-fitting algorithm.Then,the cloud platform is used to control the binocular camera,which observes the target and measures the distance between the target and the system.At last,experiments are done with the target at different distances.Experimental results show that the moving objects can be detected accurately in the fish-eye images,and the distances between the target and the system can be measured accurately by the binocular camera.Also the accuracy of angle calibration can meet the system requirements.The automatic surveillance system reduces the effects of fish-eye image distortion,and has the characteristics of large field of view and high detection accuracy.It provides a detection method for automatic surveillance system.