基于雷达图像的水面无人艇目标检测技术

摘要

To solve the radar target detection problem of an unmanned surface vehicle ( USV) , this paper proposed an embedded radar image acquisition and processing system. First, the architecture of the embedded radar image acquisition and processing system was presented. Based on the target location information, this system completed the target matching in the image sequence and accomplished target tracking by selecting the location, area, and moment invariant of the target. Then a Kalman filter was used for prediction of the target location information to solve the problem of lost targets, establishing a complete chain of the target in an image sequence. Finally, the feasibility of the embedded radar image acquisition and processing system was verified by sea trial. The experimental results demonstrate that the proposed image processing algorithm, target matching, and Kalman filter prediction algorithm can efficiently complete the target detection and tracking.%针对水面无人艇雷达目标探测问题,进行了嵌入式雷达图像采集处理系统的研究.提出了嵌入式雷达图像采集处理系统的体系结构.在提取目标位置信息的基础上,选择目标的位置、面积、不变矩特征,完成目标在雷达图像序列中的匹配,实现目标跟踪.针对雷达图像序列中偶有目标丢失的问题,利用卡尔曼滤波预测目标位置信息,帮助建立完整的目标链.最后通过海试验证了所设计的嵌入式雷达图像采集处理系统的可行性,试验表明所提出的图像处理算法、目标的特征匹配和卡尔曼滤波预测算法可以有效的完成目标的检测和跟踪.