Wafer Surface Defect Detection Based on Feature Enhancement and Predicted Box Aggregation

Abstract

For wafer surface defect detection, a new method based on improved Faster RCNN is proposed here to solve the problems of missing detection due to small objects and multiple boxes detection due to discontinuous objects. First, focusing on the problem of small objects missing detection, a feature enhancement module (FEM) based on dynamic convolution is proposed to extract high-frequency image features, enrich the semantic information of shallow feature maps, and improve detection performance for small-scale defects. Second, for the multiple boxes detection caused by discontinuous objects, a predicted box aggregation method is proposed to aggregate redundant predicted boxes and fine-tune real predicted boxes to further improve positioning accuracy. Experimental results show that the mean average precision of the proposed method, when validated on a self-developed dataset, reached 87.5%, and the detection speed was 0.26 s per image. The proposed method has a certain engineering application value.

Multi-Modal Fusion for Multi-Task Fuzzy Detection of Rail Anomalies

“…For the detection of anomalies on rail surfaces, Zhang et al [1] introduce an enhanced approach employing YOLOX and image enhancement techniques for the detection of rail surface defects. The method yields a notable 2.42% improvement in the mean average precision (mAP) of the YOLOX network.Yang et al [2] combined the grayscale features of different parts of the image and proposed a rail surface segmentation method based on sliding window grayscale maximum value, achieving an improved detection accuracy of 2.78%.…”

RSDNet: A New Multiscale Rail Surface Defect Detection Model

“…Wang et al [35] designed spatial attention sharpening filters based on YOLOv5s to enhance attention to the defects at the edge location of the rail defects and constructed M-ASFF to enhance the details of the underlying features of tiny defects. Zhang et al [36] used BiFPN for feature fusion at the neck of YOLOX and also fused the NAM attention mechanism to improve the image feature expression ability, and the experimental results showed that the defect recognition rate was improved by 2.42% compared to YOLOX. Wang et al [37] addressed the problem of detecting small targets and dense occlusion on the surface of rails by introducing the SPD-Conv building block in YOLOv8 to improve detection attention to small and medium-sized targets, and the Focal-SIoU loss function was used to adjust the sample weights to improve the model's ability to recognize complex samples.…”