In view of the status quo of current insulator icing monitoring research

an approach to icing identification for insulator in transmission line images with complex background is proposed. Firstly

position detection of insulators from acquired images is conducted by use of template matching. Then in this area

images will be handled with preprocessing

segmentation and edge extraction. Secondly

icing status is identified based on the pixels of insulator region before and after icing

once it is estimated to be iced

the ice thickness is calculated. Finally

the images captured from the artificial icing laboratory and on-site power transmission lines are used as the analysis objects of the algorithm

the insulator icing thickness is compared between the algorithm results and manual measurements under artificial icing laboratory condition

the thickness error is less than 1. 5 mm. It can be seen that the algorithm can identify icing status and calculate the ice thickness of power transmission line. And analysis results truly reflect the on-site icing conditions of power transmission lines. This work is supported by National Natural Science Foundation of China(No. 51177115)and Key Science and Technology Innovation Team of Shaanxi(No. 2014KCT-16).