红外技术能有效地检测电力设备过热缺陷，具有远距离、不接触、不取样、准确、快速、直观等特点。传统的电力设备故障红外人工诊断耗时、耗力，而针对人工诊断不足提出的智能诊断其难点之一在于能否较好的获得感兴趣区域（ROI，Region of interest）。红外图像具有强度集中、对比度低等性质，常用的分割算法用于电力设备红外图像ROI 获取，其结果往往是过分割。针对过分割难点，本文提出一种基于FAsT-Match 算法的电力设备红外图像分割方法。首先，运用FAsT-Match 算法在可见光图像中近似模板匹配，然后在红外与可见光图像之间通过近似仿射变换找到目标在红外图像中的近似区域，最后用分割算法对近似区域分割。实验结果表明，提出的方法能够较好地解决电力设备红外图像过分割问题。

Infrared Thermography(IRT) plays a very important role in monitoring and inspecting thermal defects of electrical equipment without shutting down. It has many advantages such as non-contact detection, free from electromagnetic interference, safety, reliability and providing large inspection coverage. The traditional manual analysis of infrared images may take a lot of time and efforts. To avoid the lack of manual analysis, many intelligent fault diagnosis methods for electrical equipment are proposed, but one of the greatest difficulties is to find the accurate ROI. The result of infrared image segmentation is often over-segmented when using traditional segmentation algorithms due to its over-centralized distributions and low intensity contrasts. In this paper, a novel approach based on FAsT-Match algorithm is proposed for infrared images segmentation of electrical equipment. Firstly, FAsT-Match algorithm is used for target template matching in visible image. Secondly, rough target region in infrared image is got by approximate affine transformation between infrared image and visible image. Finally, several segmentation algorithms are applied. The experiment shows the effectiveness of the method.