Abstract
Surface quality inspection is important for vehicle rearview mirrors. Surface quality defects, such as bubbles, particles, cracks, and scratches, may appear during the production process. Traditionally, manual inspection is time-consuming, laborious, low in accuracy, and costly. In recent years, object detectors based on deep learning have been used for defect detection [such as single shot detectors (SSDs)], and object detectors mostly detect the details of small objects at a shallow level. However, the lack of shallow semantic information will lead to inaccuracy. The deep layer has more semantic information, but the deep layer cannot be detected due to the effect of the complete convolutional layer in the convolutional neural network. In this article, we propose an enhanced SSD method to detect micro-defects on the nonplanar surface of vehicle rearview mirrors. We call it the multi-scale atrous single-shot detector (MSASSD). Specifically, we first replace the maximum pool depth layer with an unconscious convolutional layer to expand the receiving field without reducing the size of the input image. Then, we link the shallow layer to the deep layer through the fusion block to form new and rich fusion features for object detection. Finally, we use multi-scale features (including deep features and fusion features) to predict defects. The results show that our MSASSD method can improve the average accuracy of defect detection (about 1.2% compared with the SSD method), while the detection speed is equivalent (low about two frames per second compared to the SSD method).
Highlights
Producing surfaces of high quality is one of the key goals in the manufacturing of vehicle mirrors
We replace the max-pooling operation with the atrous convolution one in the single shot detectors (SSDs) deep layers to enlarge the receptive field without reducing the input image size
Shallow and deep layers are linked by fusion blocks to form new fusion features, which contain rich information for object detection
Summary
Producing surfaces of high quality is one of the key goals in the manufacturing of vehicle mirrors. Vehicle mirror manufacturers relied on human workers to manually inspect mirror surfaces and find potential defects This manual process is undoubtedly time-consuming, effortful, error-prone, and costly. Methods based on machine vision have been increasingly applied in surface inspection and defect detection.. Compared with manual defect detection methods, machine-vision-based automatic methods typically have higher detection accuracies and better real-time performance and require faster processing. In these automated methods, a dedicated high-resolution CCD/CMOS gray-scale camera is essentially used for image collection. Numerous machine-vision methods have been investigated for defect detection in vehicle mirrors and other types of surfaces These methods can be mainly divided into two categories, namely, traditional image processing methods and deep learning methods. A deep learning architecture is generally based on multi-layer neural networks, whose performance is typically superior to classical learning models with hand-crafted features
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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.
