Power Line Detection for Aerial Navigation Using Deep Learning

¹ Department of Computer Science and Engineering, Brainware University, Barasat Kolkata, West Bengal, 700125, India.
² School of Computer Science and Engineering, VIT-AP University, Amaravati, Andhra Pradesh, 522237, India.
³ Department of Computer Science and Engineering, Aliah University, IIA/27, AA II, Newtown, Kolkata, 700160, West Bengal, India.

^* e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

Power line detection is crucial for ensuring the safety of low-altitude aircraft such as uncrewed aerial vehicles and helicopters. Al-though many traditional techniques have been used for this task, their detection accuracy has often been limited. To address this issue, we propose a deep learning–based approach for power line detection that demonstrates strong performance on both visible-spectrum and infrared images, achieving detection accuracies of 91.75% and 99.12%, respectively. The proposed method effectively identifies power lines across varying environmental conditions and complex backgrounds. Its stable performance over a wide range of experimental settings highlights the robustness of the approach. In addition, the superior effectiveness of infrared imagery compared to visible-band images, as observed in this study, is particularly noteworthy.