Abstract: In order to improve the safe operation level of power system,aiming at the problem of bird’s nest identification on transmission lines,a bird’s nest detection network based on YOLO is proposed. Firstly,the backbone network is built by constructing GhostNet module,and the feature layer extraction method is optimized. Then,by improving the feature pyramid connection layer and combining PANet structure,the feature pyramid of bottleneck network is constructed,and finally the YOLO-NEST network is built. A data set is constructed and expanded for training. The proposed network is compared with other target detection algorithms,it is concluded that the proposed network is more efficient in bird’s nest detection of transmission lines.