基于改进FAsT-Match算法的电力设备红外图像多目标定位

Multi-target Localization for Infrared Images of Electrical Equipment Based on Improved FAs T-Match Algorithm

红外检测技术具有远距离、不接触、不取样、不解体、准确、快速、直观等特点,广泛用于电力设备过热缺陷监测和诊断,对提高电力系统的稳定性具有重要意义。FAs TMatch(fast affine template matching)算法是一种基于灰度值的快速模板匹配算法,可在一幅图像中找到一个近似全局最优目标。文中综合利用可见光和红外图像,提出一种多目标定位方法。首先,通过改进的FAsT-Match算法在电力设备可见光图像中实现多目标定位;其次,利用红外图像和可见光图像之间存在近似仿射变换,求出目标在红外图像中的位置。实验结果表明文中方法的有效性。

Infrared detection technology can effectively monitor and diagnose overheating of electrical equipment online, which has important significance for improving the stability of power system, since it has many advantages such as remote sensing, non-contact, non-sampling, non-disassemble detection, and it is very accurate and fast. FAs T-Match（fast affine template matching） is a fast template matching algorithm based on gray-scale values to find an approximate global optimal target in an image which has photometric invariance and can handle arbitrary 2D affine transformations. In this paper, a multi-targeting localization algorithm was presented for static electrical equipment images by using visual light images and infrared images together. Firstly, multi-targets in visible images of electrical equipment were found based on improved FAs T-Match algorithm; Secondly, approximate targets were located in infrared images through approximate affine transformation between infrared images and visible images. Experimental results showed the effectiveness of this method.