发表机构
Incheon National University（仁川国立大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对半导体划痕检测难题，提出ScratNet框架，集成改进Swin Transformer主干与定制解码器，含多尺度扩张聚合等模块，经阶段自适应特征聚合和边界感知细化，在检测薄且不规则缺陷上精度高，优于现有方法。
AI 中文摘要
半导体制造中的表面划痕缺陷因其形状不规则、对比度低和尺度变化而带来重大挑战。传统检测方法难以可靠检测此类缺陷，尤其是在复杂成像场景中。基于卷积神经网络的深度学习方法虽提高了准确性，但常无法捕捉细粒度边缘细节。为此提出ScratNet，它将改进的Swin Transformer主干与定制解码器集成。解码器包含多尺度扩张聚合模块捕捉局部和全局上下文、茎集成模块恢复空间细节以及精度细化分支用各向异性卷积增强边界清晰度。通过阶段自适应特征聚合和边界感知细化实现高精度，实验表明其优于现有方法，为高精度制造中的自动划痕检测提供了可扩展且强大的解决方案。
英文摘要
Surface scratch defects in semiconductor manufacturing pose significant challenges due to their irregular shapes, low contrast, and varying scales. Traditional inspection methods often struggle to detect such defects reliably, especially in complex imaging scenarios. While deep learning approaches based on Convolutional Neural Networks (CNNs) have improved accuracy, they often fail to capture fine-grained edge details. To address these limitations, we propose ScratNet, a novel end-to-end scratch segmentation framework that integrates a modified Swin Transformer backbone with a tailored decoder. The decoder incorporates a Multi-Scale Dilated Aggregation (MDA) module to capture both local and global context, a Stem Integration Module (SIM) to restore spatial detail, and a Precision Refinement (PR) branch that enhances boundary sharpness using anisotropic convolutions. Through this stage-adaptive feature aggregation and boundary-aware refinement, ScratNet achieves superior accuracy on thin and irregular defects. Extensive experiments demonstrate that ScratNet consistently outperforms existing methods, providing a scalable and robust solution for automated scratch inspection in high-precision manufacturing.
