红外制导在现代战争中占据独特的技术地位, 作为其关键技术之一, 红外小目标检测一直是理论研究的热点, 但是由于多种因素的影响, 红外小目标容易淹没在背景中, 日趋复杂的背景以及隐身技术的应用, 给红外小目标的检测带来极大的困难。结合尺度空间理论, 分析红外小目标图像的能领域特征, 采用多步长能量差异进行目标中心点及尺寸的确定, 最后根据噪声和目标点的特征, 对检测结果中的孤立噪声点进行剔除, 最终检测出小目标。实验结果表明, 该算法计算简单方便, 运行时间短, 能够有效地检测出目标。

IR small target detection is always one of the topic difficulties, and it is easy to be covered by the background due to many factors. The IR small target detection is very difficult for the complicated background and stealthy technology application. With scale-space theory, the characteristics of IR images of the IR small target in the energy field are analyzed. Multi-step energy difference is used to determine the center point and the size of the target. Based on the characteristics of noise and target point, the isolated noise points in the results are rejected to inspect the small target finally. Experimental results show that the algorithm has simple calculation, short running time and can inspect the target effectively.