An Accurate and Real-time Self-blast Glass Insulator Location Method Based On Faster R-CNN and U-net with Aerial Images

TL;DR

This paper presents a deep learning-based method combining Faster R-CNN and U-net for accurate, real-time detection of self-blast glass insulators in aerial images, improving maintenance efficiency.

Contribution

The paper introduces a novel two-module framework integrating object detection and pixel classification for insulator localization in aerial imagery.

Findings

High detection accuracy demonstrated on Chinese aerial image dataset

Real-time performance achieved with the proposed method

Outperforms existing approaches in accuracy and speed

Abstract

The location of broken insulators in aerial images is a challenging task. This paper, focusing on the self-blast glass insulator, proposes a deep learning solution. We address the broken insulators location problem as a low signal-noise-ratio image location framework with two modules: 1) object detection based on Fast R-CNN, and 2) classification of pixels based on U-net. A diverse aerial image set of some grid in China is tested to validated the proposed approach. Furthermore, a comparison is made among different methods and the result shows that our approach is accurate and real-time.