AUTHOR=Wang Yuanyuan , Tian Haiyang , Yin Tongtong , Song Zhaoyu , Hauwa Abdullahi Suleiman , Zhang Haiyan , Gao Shangbing , Zhou Liguo 

TITLE=The transmission line foreign body detection algorithm based on weighted spatial attention

JOURNAL=Frontiers in Neurorobotics

VOLUME=Volume 18 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/neurorobotics/articles/10.3389/fnbot.2024.1424158

DOI=10.3389/fnbot.2024.1424158

ISSN=1662-5218

ABSTRACT=The secure operation of electric power transmission lines is of significant importance to the economy and society. However, external factors such as plastic film and kites can cause damage to the lines, potentially leading to power outages. Traditional detection methods are inefficient, and the accuracy of automated systems is limited in complex background environments.This paper proposes a weighted spatial attention (WSA) network model to address the problems of low accuracy caused by background texture occlusion for identifying extraneous materials within electrical transmission infrastructure. First, this article uses color space conversion, image enhancement, and improved LSKNet (large selective kernel network) technology in the model preprocessing stage to boost the model's proficiency to detect foreign objects in intricate surroundings. Second, to accurately capture and identify the characteristic information of foreign objects in power lines, the model adopts the dynamic sparse BSAM (BiLevel spatial attention module) structure proposed in this article in the feature extraction stage. In the feature pyramid stage, by replacing the feature pyramid network structure, reasonable weights are allocated to the bidirectional feature pyramid network (BiFPN) to optimize the feature fusion results, thereby ensuring that the semantic information of foreign objects in the power line output by the network is effectively identified and processed. The outcomes of the experiments reveal that test recognition accuracy of the proposed WSA model on the PL (power line) dataset is improved by three percentage points compared with that of the YOLOv8 model, reaching 97.6%. This enhancement demonstrates the WSA model's superior capability in detecting foreign objects on power lines, even in complex environmental backgrounds. The integration of advanced image preprocessing techniques, the dynamic sparse BSAM structure, and the BiFPN has proven effective in improving detection accuracy and can potentially transform the approach to monitoring and maintaining power transmission infrastructure.