基于多特征融合的空对地目标检测和追踪方法

An Air-to-Ground Target Detecting and Tracking Method Based on Multi-feature Fusion

随着无人机的广泛应用,无人机对地面目标检测和追踪技术成为研究的热点。针对光流法对运动目标检测和追踪出现误检、丢失的问题,应用运动目标纵向的边缘特征配合光流法的方法完成目标检测,应用SIFT与光流法实现目标追踪,并提出恒速模型与光流块匹配的方法解决追踪过程中建筑物等对目标的遮挡问题。实验结果表明,所提方法能够对目标进行有效的识别与追踪,并具有较强的鲁棒性。

With the wide application of Unmanned Aerial Vehicles(UAVs),using UAVs for detecting and tracking the ground targets has become a research hotspot.Aiming at the problem of false detection and loss of target during the detection and tracking of moving objects by the optical flow method,the target detection is performed by using the longitudinal edge features of the moving object in conjunction with the optical flow method.Simultaneously,Scale-Invariant Feature Transform(SIFT)and optical flow method are used to realize target tracking.A method of matching the constant-speed model with the optical flow block is proposed to solve the problem of occlusion of the target by buildings during tracking.The experimental results show that the method can implement effective target identification and tracking,and has strong robustness.