摘要
为了较好地实现电力设备红外图像故障区域提取,提出了一种基于Canny 算子边界检测的脉冲耦合神经网络(Pulse-coupled Neural Network,PCNN)红外图像区域提取方法。在该方法中,首先以PCNN 模型同步点火特性为基础,通过优化原始PCNN 模型内在的参数,使得模型迭代过程中将图像转换成为时间点火序列,然后引入Canny 边界检测算子并结合区域灰度特性,获取最佳时刻的脉冲输出信息,实现红外图像中热故障区域的有效提取。最后通过真实红外故障图像测试,验证了文中方法的有效性和适用性,同时方便了后续的特征提取与识别。
To implement the extraction of fault regions from infrared images of electronic equipment,in this study,we present a pulse-coupled neural network (PCNN) infrared image region extraction method,which is based on the cooperation of the Canny algorithm.In this method,by using the synchronous pulse characteristics of the original PCNN model,several parameters are simplified to enable the PCNN model to generate time series through iterations.Meanwhile,the canny method is used to improve the ability of the PCNN model to segment infrared images efficiently and extract effective thermal fault regions.Experimental results show that the proposed method has the ability to obtain good segmentation performance and can be suitable for further feature extraction and recognition.
作者
冯振新
许晓路
周东国
江翼
丁国成
FENG Zhengxin;XU Xiaolu;ZHOU Dongguo;JIANG Yi;DING Guocheng(Wuhan NARI Limited Liability Company of State Grid Electric Power Research Institute,Wuhan 430074,China;Wuhan University,School of Power and Mechanical Engineering,Wuhan 430072,China)
出处
《红外技术》
CSCD
北大核心
2019年第7期634-639,共6页
Infrared Technology
基金
国家电网公司总部科技项目资助(524625160017)
