基于补体免疫聚类的电力设备红外图像目标提取算法

Infrared image target extraction method for power equipment based on complement immune clustering

针对电力设备红外检测中,由于背景复杂,痕迹重叠而难以有效提取目标电力设备这一问题,提出一种补体免疫聚类目标提取算法。算法参考免疫系统中补体系统作用机理,建立多个电力设备模板图像库,通过模板与图像的匹配度来确定目标区域的轮廓和位置。对图像进行k-means聚类处理,给每一个像素点赋予新的特征值,重新确定聚类中心,完成对目标区域图像的提取,并证明该算法的收敛性。通过对实验结果的定性定量分析进一步证实该算法的有效性。

In infrared detection of power equipment,due to the complex background and overlapping of traces,it is difficult to extract the target power equipment effectively.This paper proposes an image extraction method based on complement immune clustering.The method refers to the action mechanism of the complement system in the immune system,establishes a plurality of power device template image libraries,and determines the contour and position of target region by matching the template with the image.The image is k-means clustered,and each pixel point is given a new feature value,the cluster center is re-determined,the image of the target area is extracted,and the convergence of the method is proved.The validity of the method is further confirmed by qualitative and quantitative analysis of experimental results.