无人机对船舶目标的改进CamShift跟踪算法研究

针对由于CamShift算法跟踪特征单一引起的对颜色相似目标或背景的干扰和对目标遮挡情况较敏感的问题,提出了一种基于改进CamShift融合局部特征匹配的无人机目标跟踪算法.首先,采用基于 H分量和LBP二维模板的改进CamShift目标跟踪算法以提高对相似目标干扰的鲁棒性;其次,在能保证目标跟踪的实时性要求的前提下,融合局部特征匹配算法中的BRISK匹配算法,可有效改善CamShift对颜色相似目标或背景的干扰的敏感性,同时增强对目标遮挡鲁棒性.实验结果表明,该改进算法通过颜色特征和局部特征的共同定位目标,实现了目标的准确跟踪.

Research on UAV Improved GamShift Tracking Algorithm for Ship Targets

Aiming at the problems of interference to the color similar target or background and the sensitivity to the target occlusion caused by the single tracking characteristic of CamShift algorithm,a new UAV target tracking algorithm based on improved CamShift and local feature matching is proposed.Firstly,the improved CamShift target tracking algorithm based on H-component and LBP 2D template is used to improve the robustness to similar target interference.Secondly,in the premise that the real-time requirements of the target tracking can be guaranteed,the fusion of local feature matching algorithm is the BRISK matching algorithm can effectively im-prove the sensitivity of the CamShift to the interference of color similar targets or backgrounds,and at the same time,enhances the ro-bustness to the target occlusion.The experimental results show that the improved algorithm achieves accurate tracking of the target through the co-localization of color features and local features.