基于连通域内模板匹配的靶机与弹迹检测

Detection of Target Drone and Cannonball based on Image Matching

针对靶机与射弹识别以及靶机与射弹分离提取的图像处理的难点问题,提出了一种基于Sobel算子和连通域内模板匹配的靶机与射弹的提取与识别方法。该方法不仅适用于不同种类的射弹与靶机类型,同时对不同天气条件下的观测图像也具有较好的鲁棒性,能有效满足防空射击评价系统的需求。

Aiming at the difficult problems of identification drone and ball as well as image processing after separating drone and ball ,this paper presents the way for extracting and identifying drone and hall on the basis of Sobel operator and template matching. which is not only adapted to the different kinds of balls and drones ,but also offers a better robotic characteristics to observe images under the different weather condition, meeting the requirement of air defence shooting evaluation system.