# A Novel 24 h × 7 Days Broken Wire Detection and Segmentation Framework Based on Dynamic Multi-Window Attention and Meta-Transfer Learning

## TL;DR

This paper introduces a new framework for detecting and segmenting broken wires in substations that works reliably under different lighting conditions.

## Contribution

The framework combines dynamic multi-window attention and meta-transfer learning for improved performance with limited data.

## Key findings

- The framework improves detection and segmentation accuracy under varying lighting conditions.
- A dataset of 3760 RGB images was used to evaluate performance across six lighting levels.
- The method supports small-sample training while reducing negative transfer effects.

## Abstract

Detecting and segmenting damaged wires in substations is challenging due to varying lighting conditions and limited annotated data, which degrade model accuracy and robustness. In this paper, a novel 24 h × 7 days broken wire detection and segmentation framework based on dynamic multi-window attention and meta-transfer learning is proposed, comprising a low-light image enhancement module, an improved detection and segmentation network with dynamic multi-scale window attention (DMWA) based on YOLOv11n, and a multi-stage meta-transfer learning strategy to support small-sample training while mitigating negative transfer. An RGB dataset of 3760 images is constructed, and performance is evaluated under six lighting conditions ranging from 10 to 200,000 lux. Experimental results demonstrate that the proposed framework markedly improves detection and segmentation performance, as well as robustness across varying lighting conditions.