论文部分内容阅读
根据某型号多传感器光电成像设备实际应用的特点,提出了一种基于小波多分辨率分析(MRA)的多算子自适应目标识别跟踪算法,可实现昼夜型光电设备对陆地/海面全动态场景不确定目标的识别跟踪。建立了一种基于帧内、帧间置信度评价的双阈值自适应跟踪预测策略和相似目标辩识准则,有效地提高了目标跟踪的稳定性和鲁棒性。地面仿真及空中实际飞行实验结果表明:该算法能够很好地实现在各种复杂场景条件下的动态场景目标的稳定跟踪。
A multi-operator adaptive target recognition and tracking algorithm based on wavelet multi-resolution analysis (MRA) is proposed according to the practical application of a multi-sensor photoelectric imaging device of a certain type. The algorithm can realize the full dynamic scene Uncertain target recognition tracking. A two-threshold adaptive tracking and prediction strategy and similar target identification criterion based on intra-frame and inter-frame confidence evaluation are established to effectively improve the stability and robustness of target tracking. Experimental results of ground simulation and airborne actual flight show that the proposed algorithm can achieve stable tracking of dynamic scene targets under various complex scene conditions.