Low-pass U-Net: a segmentation method to improve strip steel defect detection
- 7 December 2022
- journal article
- research article
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 34 (3) , 035405
Abstract
The detection of strip steel surface defects is critical to ensuring the quality of strip steel products. Many deep learning-based methods have been presented and can achieve outstanding performance. However, most of these methods ignore the frequency information among defect areas, which plays an important role in defect detection. This paper proposes a deep learning method to further improve defect segmentation effects based on existing methods, called low-pass U-Net. Since most defects in strip steel are located in high-frequency areas, we implement a low-pass filter before downsampling in the encoder, which prevents aliasing and separates out high-frequency information. The high-frequency feature is transferred into the decoder to assist segmentation. Following previous studies, we propose an adaptive variance Gaussian low-pass layer to generate different filters according to each spatial location of the feature map, with lower computing resource use. Furthermore, to detect defects at significantly different scales, an improved Hypercolumn module is adopted at the end of the decoder to upsample and fuse the feature maps in different resolutions, where Subpixel replaces the bilinear interpolation to refine the upsampled results. The proposed method is validated on practical datasets and achieves considerable performance improvement (with a best Dice coefficient of 0.903), which demonstrates the effectiveness of low-pass U-Net. The introduction of the adaptive variance Gaussian low-pass filter layer results in a 3% increase in Dice coefficient in a comparative inference time, which achieves a balance in performance, inference time and complexity.Keywords
Funding Information
- National Natural Science Foundation of China (62076015)
- the grant of China Scholarship Council
This publication has 15 references indexed in Scilit:
- A deep-learning-based approach for fast and robust steel surface defects classificationOptics and Lasers in Engineering, 2019
- Defect Detection of Steel Surfaces with Global Adaptive Percentile Thresholding of Gradient ImageJournal of The Institution of Engineers (India): Series B, 2017
- A surface defects inspection method based on multidirectional gray-level fluctuationInternational Journal of Advanced Robotic Systems, 2017
- Deep Learning‐Based Crack Damage Detection Using Convolutional Neural NetworksComputer-Aided Civil and Infrastructure Engineering, 2017
- Detection of Pinholes in Steel Slabs Using Gabor Filter Combination and Morphological FeaturesISIJ International, 2017
- An End‐to‐End Steel Strip Surface Defects Recognition System Based on Convolutional Neural NetworksSteel Research International, 2016
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Using fractal features of digital images for the detection of surface defectsPattern Recognition and Image Analysis, 2015
- A noise robust method based on completed local binary patterns for hot-rolled steel strip surface defectsApplied Surface Science, 2013
- Surface Detection of Continuous Casting Slabs Based on Curvelet Transform and Kernel Locality Preserving ProjectionsJournal of Iron and Steel Research International, 2013