DS2A-Former: Battery Surface Defect Segmentation via Dual-Stream Spatial Attention Transformer Network

With the advancement of industrial automation, surface defect inspection has become crucial for battery manufacturing quality assurance. However, existing public datasets lack sufficient resources for battery surface defects, hindering the process of battery defect inspection. To address this, we developed the multiclass battery surface defect (M-BSD) dataset, which contains 2491 images and 6639 samples spanning three defect types. Meanwhile, deep learning-based methods have made progress in industrial quality inspection but face challenges in handling complex battery defect details, defect category similarities, and small-target defects. These issues hinder accurate, quantitative analysis of defects. To overcome the above challenges, we propose the DS²A-Former, a novel dual-stream spatial attention Transformer network that integrates semantic and spatial information to enhance segmentation performance under complex conditions. The dual-scale spatial cross-attention (DSSCA) module facilitates information exchange across scales and suppresses background noise, while the dual-query gated spatial attention (DQGSA) module uses a spatial gating mechanism to prioritize knowledge between two queries, improving the representation of defect information. Extensive experiments on the M-BSD dataset validate the effectiveness of our approach. We also experimented with it on the Crack-500 dataset, demonstrating its broad applicability in defect segmentation across domains. Our code and dataset are available at https://github.com/yikuizhai/DS2A-Former