Aiming at the fault problem of insulator in transmission line, a set of fault identification algorithm for insulator self-explosion was designed and implemented.The overall framework of the UAV was designing, including navigation ability, obstacle avoidance ability and anti-jamming ability of the work next to the high voltage line; the traditional genetic algorithm was improved, and the optimal entropy threshold method (KSW entropy) was combined to split the insulator image; morphology was used to get pure insulator string; the designed fitting method was used to predict the proper position of the insulator image, and the self-exploding faults of the insulator were identified from the front view and oblique view respectively. The results show that by using the proposed algorithm, the accuracy is high and the speed is about 75% higher than that of the traditional genetic algorithm; the algorithm is suitable for fault identification of insulators taken from multiple angles, which provides guarantee for the insulator fault identification shot by the UAV.