As a carrier of power transmission, high-voltage overhead transmission lines are an important guarantee for the stable operation of the power system. Therefore, it is particularly important to detect abnormal phenomena for transmission lines as early as possible (such as foreign objects, broken strands, etc.) and implement maintenance in a timely manner. In this paper, the vigorous promotion of intelligent operation robots will help ensure the safe, efficient, and economical operation of the power grid system. A pose estimation method of unmanned aerial vehicles is proposed based on point-line feature fusion to address the weak pattern recognition and automatic tracking capabilities in existing power operation drones, as well as the poor robustness of visual odometry. A complete target detection and tracking system of unmanned aerial vehiclesis developed. Compared to other similar unmanned aerial vehicles environment perception schemes, the proposed target tracking algorithm can efficiently extract point and line features from infrared images, and achieve target detection and automatic tracking functions of small multi rotor drones in transmission line scenes with extremely low power consumption.