Attention-Based Multiscale Feature Fusion for Efficient Surface Defect Detection

Deep-learning-based detection methods have been widely applied to industrial defect inspection. However, directly using vanilla detection methods fails to achieve satisfying performance due to the lack of identifiable features. In this article, a novel attention-based multiscale feature fusion module (AMFF) is proposed, aiming to enhance defect features and improve defect identification by leveraging attention mechanism in the feature fusion. AMFF includes self-enhanced attention module (SEAM) and cross-enhanced attention module (CEAM). SEAM is performed on a single feature map, which first adopts multiple dilation convolutions to enrich contextual information without compromising resolution and then utilizes the intralayer attention on the current feature map. CEAM takes both the current feature map and the adjacent feature map as input to perform cross-layer attention. The adjacent feature map is modulated with the guidance of the current feature map, which is then combined with the current feature map and the output of SEAM for final prediction. AMFF is utilized in current feature fusion networks, e.g., feature pyramid network (FPN) and path aggregation FPN (PAFPN), and is further integrated into prevalent detectors to guide them to pay more attention to defects rather than the background. Extensive experiments are conducted on two real industrial datasets released by Tianchi platform, i.e., fabric and aluminum defect datasets. For each dataset, 500 images are randomly selected for test and the rest for training. The proposed AMFF is demonstrated to significantly boost defect detection accuracy with acceptable computational cost, and the real-time performance could fully satisfy practical requirements.

I. Introduction

Surface defect detection [1] plays a pivotal role in industrial applications, particularly in the realm of quality control. Manual defect detection is labor-intensive, time-consuming, and sometimes causes secondary damage, which does not satisfy the growing demand of industrial applications. Additionally, human subjectivity will inevitably affect the consistency of product quality, especially for scenes with weak visual effects. Vision-based defect detection provides a promising solution to deal with the difficulties due to its high efficiency and reliable performance.