An Automatic Detection Method of Bird's Nest on Electric Tower Based on Attention Full Convolutional Neural Networks
Wuzhong Dong, Lie Wu, Qi Wang, Sen Cheng · 2021
Automatic detection of the bird's nest on the transmission line tower is of great significance to ensure the stable operation of the power grid. The main challenge is that the diversification and complexity of the environment will have a greater impact on the detection results. In addition, the real-time performance of the detection algorithm is the key to the successful deployment of the algorithm. In order to solve the above challenges, this paper proposes an Attention Complete Convolutional Neural Network (AFCNN) to detect the bird's nest on the electric tower. The specific content is as follows: First, design a shallow feature extraction model to shorten the calculation time of the model. Subsequently, the attention network is added to the feature extraction model to reduce the interference of the surrounding environment on the detection results and improve the detection accuracy. The algorithm uses natural images collected by drones for training and testing. Experimental results show that the algorithm can be applied to bird nest detection in complex environments and has high real-time performance. Our work provides a solution to ensure the safe operation of the grid.