为了提高电力系统变电设备红外监测的效率, 同时对多个相似目标进行监测, 提出了一种改进的基于 BRISK(Binary Robust Invariant Scalable Keypoints, 二进制鲁棒尺度不变特征)的模板匹配方法。首先在对目标模板图像与被监测图像预处理的基础上, 利用 BRISK算法提取模板与被监测红外图像的特征点, 生成二进制描述子; 利用滑动窗口在被监测图像上等步长滑动, 检测出图像中存在的多个疑似目标区域, 然后通过 Hamming距离来判断正确匹配点对, 从而正确识别变电设备。通过实验平台进行了相关实验, 通过与模板匹配方法比较, 该方法明显减少了运算量, 充分满足系统实时性要求, 提高了变电设备红外监测的效率。

In order to improve the efficiency of infrared monitoring on power system equipment and monitoring multiple targets at the same time, an improved substation equipment infrared image recognition method based on Binary Robust Invariant Scalable Keypoints(BRISK)is proposed. The method first extracts the feature points of the template image and the monitored image using BRISK algorithm, and generates the binary descriptors; then applies a sliding detection window to search the multiple suspected target areas, then determines the correct matching points by Hamming distance; so we can locate the target area on the monitored image. Some experiments are taken to investigate the performance of the method; the result shows that this method can significantly reduces the amount of computation, fully meets the requirements of real-time systems. This method improves the efficiency of infrared monitoring on power system equipment.