Coarse-to-Fine Few-Shot Defect Recognition With Dynamic Weighting and Joint Metric
- 21 July 2022
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 71 (00189456) , 1-10
Abstract
Deep-learning-based methods have been widely used in defect recognition and achieved great success. However, deep-learning-based methods need a large-scale dataset. While in real industrial scenarios, the training samples are always insufficient since the defect data acquisition is difficult and time-consuming. Therefore, in this article, the few-shot learning theory is introduced to address the challenge. We propose to achieve the few-shot defect recognition (FSDR) in a coarse-to-fine manner with dynamic weighting and joint metric. In the coarse-grained phase, following feature embedding, we propose an affine dynamic weighting (ADW) module to control the embedding output of all the channels according to the global context. By the dynamic weighting, the model can extract discriminative features better with learnable affine parameters. In the fine-grained phase, we propose a joint metric method which contains a Kullback–Leibler (K-L) divergence-based covariance metric module (KLCM) and cosine classifier. In this method, KLCM exploits the covariance matrix from the local descriptors to represent the distribution of a special defect class and then measures the similarity between the support and query defect images. A novel FSDR dataset which contains a variety of defects from four different surfaces is constructed to evaluate our method. The results show the state-of-the-art performance compared with the mainstream methods.Keywords
Funding Information
- “Pioneer” and “Leading Goose” Research and Development Program of Zhejiang (2022C01064)
- National Natural Science Foundation of China (51935009, 52075480)
- High-Level Talent Special Support Plan of Zhejiang Province (2020R52004)
This publication has 23 references indexed in Scilit:
- An End-to-End Steel Surface Defect Detection Approach via Fusing Multiple Hierarchical FeaturesIEEE Transactions on Instrumentation and Measurement, 2019
- Surface defect saliency of magnetic tileThe Visual Computer, 2018
- Learning to Compare: Relation Network for Few-Shot Learning2018
- Automatic Defect Detection of Fasteners on the Catenary Support Device Using Deep Convolutional Neural NetworkIEEE Transactions on Instrumentation and Measurement, 2017
- Deep Residual Learning for Image Recognition2016
- Design of deep convolutional neural network architectures for automated feature extraction in industrial inspectionCIRP Annals, 2016
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- A noise robust method based on completed local binary patterns for hot-rolled steel strip surface defectsApplied Surface Science, 2013
- Multi-scale pyramidal pooling network for generic steel defect classification2013
- The distance between two random vectors with given dispersion matricesLinear Algebra and its Applications, 1982