Substation Equipment Appearance Defect Dataset for Complex Inspection Scenarios 核心 · 已核验

mxdsjqe3kk45py83

"This Substation Equipment Defect Dataset consists of 4,235 high-fidelity, real-world inspection images meticulously curated through an extensive field project in direct collaboration with a leading power grid utility. The dataset provides a comprehensive representation of critical equipment health by encompassing five major defect categories: blurred meter readings, damaged housings, silica-gel discoloration, fractured silica-gel bodies, and damaged cover plates.Beyond its volume, the dataset is characterized by extreme diversity, featuring over 200 distinct image resolutions and a strictly partitioned 7:2:1 training, validation, and testing split to ensure rigorous model evaluation. It encapsulates the inherent complexities of industrial field environments, including heterogeneous backgrounds, low-contrast illumination, and significant multi-scale variations of defect targets. Consequently, this dataset serves as a robust benchmark for evaluating the generalization and practical efficacy of advanced computer vision frameworks. It provides a vital resource for developing intelligent power grid maintenance systems, bridging the gap between laboratory research and high-reliability industrial deployment."

本卡描述如含来自官方页的原文片段,其版权归原作者,不在本站 CC-BY 4.0 许可范围(见关于与许可)。

落地页

https://ieee-dataport.org/documents/substation-equipment-appearance-defect-dataset-complex-inspection-scenarios

国内可访问性

国内直连:可达 (2026-07-11 检测) 非直连:可达 (2026-07-11 检测)
「直连超时」表示检测窗口内未完成,系慢或不稳定证据,不构成封锁证据。

PHM 任务

fault_detection