Abstract
Mixed defects have become increasingly popular in defect detection and one of the hottest research areas in wafer maps. Postprocessing methods used to solve the overlapping problem in mass mixed defects have a poor detection speed, which is insufficient for rapid defect detection. In this paper, the fast‐soft nonmaximum suppression (fs‐NMS) method is proposed to solve this problem. The score of the detection box is updated by optimizing the penalty distribution function. Further, this paper analyzes the performance of the fs‐NMS method in wafer defect detection. As a penalty, the logistic function is used, and experiments are conducted using single‐stage and two‐stage detectors. The final results show that, compared to the soft‐NMS, the efficiency for the single‐stage and two‐stage detectors is increased on average by 9.63% and 21.72%, respectively.
Highlights
Defect detection is an important application of object detection that has received a lot of attention
In detecting complex mixed wafer defects based on deep convolutional neural networks, greedy NMS drastically reduces the screening of false-positive boxes in the postprocessing stage, but mixed defects are still difficult to detect
The greedy NMS is used in the region of interest (RoI) extraction stage, while the fast-soft nonmaximum suppression (fs-NMS) is used in the classification and localization stages
Summary
Defect detection is an important application of object detection that has received a lot of attention. In detecting complex mixed wafer defects based on deep convolutional neural networks, greedy NMS drastically reduces the screening of false-positive boxes in the postprocessing stage, but mixed defects are still difficult to detect. This causes the detector lose mass positive boxes at a certain threshold, while causes a decrease in the average precision. In industrial detection detecting thousands of wafer defects is very inefficient and often insufficient To solve this problem, this paper proposes an improved fast-soft nonmaximum suppression (fs-NMS) postprocessing algorithm to improve detection efficiency by optimizing the distribution of penalty terms in soft-NMS [11], so as to better apply to large quantities of industrial production. It is concluded that our approach is effective for both single-stage and two-stage detectors
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
