Multi-State Recognition Method of Substation Switchgear Based on Image Enhancement and Deep Learning

s. Since the current substation robot inspection process exists in high-voltage switchgear status recognition is highly susceptible to influence of complex environments such as low image contrast, light clutter interference, and blurred reading details, this paper proposes an enhancement deep learning based state method. A multi-scale Retinex-based proposed enhance adaptability outdoor images changes; improve YOLOx target detection network introduces a lightweight ECA attention mechanism without dimensionality reduction on original model’s backbone CSPDarknet, allowing model learn classification features while also focusing spatial features. The experimental results show that improved can accurately identify boundary information anomalies, quality its prediction will not be reduced for noise-containing data, shows strong generalization, robustness, accuracy rapidity, providing certain conditions realizing equipment condition monitoring.