基于质心提取的红外图像小目标检测

低信噪比条件下小目标检测技术的研究是红外信息处理中的关键性问题。由于小目标通常属于高频成分，又易于与噪声混杂在一起，更加增加了检测的难度。文章提出了一种基于质心提取的方法对红外小目标进行检测，在对图像不进行滤波的情况下找到目标可能的质心点集合。对这些置信度高的质心点集合进行分类，以确定各个目标的质心以及目标的个数。最后以确定的目标质心点搜索目标边缘，找出其边界并计算出目标的特征参数。该检测方法对噪声也有一定的抑制作用。

Small Target Detection in Infrared Image Based on Centroid Extraction

Small target detection under conditions of low signal to noise ratio is the key problem in infrared information processing technology. The detection is more difficult because a small target is usually as a result of high frequency components and easy to mix with the noise. In this paper, a center-based method is proposed to detect the infrared small target, by which the possible target center is found without image filtering. The target center and the number of target are determined by classify these high-quality mass points. Finally, the target borders are searched from the center of the target point and the characteristic parameters of the target are calculated. Noise is also decreased in the method.