针对红外图像中背景与目标的复杂性和多样性给变电站绝缘套管状态监测带来的困难，提出基于红外弱目标提取的检测方法。首先进行二维小波分解，将低频分量进行Retinex 理论增强，其余高频分量进行局部方差映射，完成小波重构后进行对比度增强，最后实现基于分步双阈值的目标提取。实验结果表明，该方法不仅改善了绝缘套管局部过热区域的红外图像的处理效果，还实现了过热点及其连带过热区域的提取和区分。

The complexity and diversity of backgrounds and targets in the infrared images bring difficulties in monitoring of the insulation sleeve in the substation. In order to overcome this deficiency, a method based on infrared weak target extraction is proposed. First, the image is decomposed by using the two-dimensional wavelet. Then, the low-frequency components are enhanced by retinex theory, and local variance mapping for the other high-frequency components is executed. After wavelet reconstruction, contrast enhancement is carried out. Finally, the target is extracted on the basis of the step-and-double threshold. Experimental results show that the method not only improves infrared image processing effect of the local superheated region of the insulating sleeve, but also realizes the extraction and division of the hot spot and its associated overheated region.