对地面弱小目标的检测算法的研究

摘要

Due to the environment, human error and imaging system's influence, more serious picture jitter and poor imaging quality will appear in camera video during research on dim and small targets on the ground. Therefore, it's necessary to make a motion compensation for image sequence. In order to realize the detection of dim and small targets on the ground, this paper proposed an algorithm to detect the dim and small targets on the ground, which is based on the Harris corner detection. The algorithm presented in this paper, uses the gray information of residuals plot and residual plot sequence. It has some advantages such as the continuity in the aspect of time, high stability, high precision and good real-time performance. It can remove the false alarm points in the residual plots very well, and greatly increase the level of dim and small targets detection on the ground under the circumstances of higher accuracy.%　　在研究地面弱小目标时，由于环境、人为误差及成像系统本身的影响，摄像机拍摄的视频会出现画面抖动比较厉害、成像质量差的现象。因此，需要对图像序列进行运动补偿。为实现对地面弱小目标的检测，提出了一种基于Harris角点检测的对地面弱小目标的检测算法。该算法利用了残差图中的灰度信息及残差图序列，在时间上具有连续性，算法的稳定性高，精度较高，实时性好。能够很好地剔除残差图中的虚警点，在较高精度下大大提高了对于地面弱小目标检测的水平。