Transformer Core Structure Recognition Dataset

The current industrial sector faces significant challenges in ensuring the quality and alignment of transformer core structures during manufacturing. The existing solutions often fall short in accuracy and efficiency, leading to increased costs and production delays. This dataset aims to address the technical issue of precise identification of core structures and their quality status, which is critical for enhancing the reliability of transformers. Data collection was conducted using high-resolution cameras in factory environments, ensuring diverse lighting conditions and angles. Quality control measures include multi-round annotations, consistency checks among annotators, and expert reviews to validate the data. The dataset is organized in a JPEG format, structured with clear labeling for each image. Each image is accompanied by metadata detailing the core structure characteristics and quality assessments.