A method based on dual criterion K-means algorithm for infrared(IR) image mutual inductor segmentation is proposed to study on infrared image segmentation of power system. In the process of realizing the segmentation of mutual inductor from IR image
the K-means algorithm is mainly used to extract the colour feature of IR image of the mutual inductor.Then infrared image is converted to grayscale and binary image and after image expansion
image corrosion and deleting small area
according to the dual criterion
mutual inductor will be eventually removed from electric IR image. Through mutual inductor segmentation experiment
fault of the mutual inductor can be identified and detected more effectively from power system.
Classifying and clustering in negative databases [J] . Ran Liu,Wenjian Luo,Lihua Yue. Frontiers of Computer Science . 2013 (6)
Suitable features selection for monitoring thermal condition of electrical equipment using infrared thermography [J] . A.S.N. Huda,S. Taib. Infrared Physics and Technology . 2013