Identification and location of distribution network cables based on multi-sensor system

In the current trend of domestic distribution network maintenance, the traditional power-off operation has been gradually replaced by power-on operation. To address the challenges of high labor intensity, low work efficiency, and safety risks associated with manual maintenance methods, researches on intelligent operation platforms for distribution network maintenance is developing rapidly. Specifically, quick and accurate identification of target cables for maintenance is essential for completing operations such as cable stripping and live wire connection in distribution network maintenance. However, due to the complex conditions of the environment, traditional methods can hardly specify the target, while the conventional deep neural networks cannot meet real-time requirements. Therefore, this paper proposes a system that utilizes RGB+Lidar to identify and locate the target cables. Besides, a lightweight image segmentation model with a straight-line attention module is developed to further improve accuracy. The verification on the testing set illustrates the satisfied speed and accuracy of the proposed method in practical scenarios.