I. Introduction
Transmission infrastructure makes it possible for power producers and consumers to connect directly [1]. Electricity grid maintenance is a big cost burden for transmission system operators because there are so many different components [2]. Cost savings could be achieved by automatic defect detection. Nevertheless, specific infrastructure must first be located and separated using an automated visual system before any form of defect detection can be carried out [3]. In order to ascertain its viability, the proposed deep learning-based method for segmenting and identifying problematic portions (such as insulators) in the video will be tested [4]. If a deep-learning-based strategy proved effective, drones might automatically monitor equipment [5].