72‐3: Deep Learning Based Visual Defect Detection in Noisy and Imbalanced Data
Authors:
- 28 June 2022
- journal article
- research article
- Published by Wiley in SID Symposium Digest of Technical Papers
- Vol. 53 (1) , 971-974
Abstract
Despite the immense potential, deep learning‐based visual defect detection is often impaired by the inadequate quality of real‐world data. Here, we leverage Confident Learning and One‐class Classification to tackle two major data issues: 1) mislabeling and 2) limited defects for model development, achieving ≥0.85 AUROC on intentionally polluted industrial data.Keywords
This publication has 9 references indexed in Scilit:
- Multimodal Machine Learning for Display Panel Defect Layer Identification2021
- 16‐4: Invited Paper: Region‐Based Machine Learning for OLED Mura Defects DetectionSID Symposium Digest of Technical Papers, 2021
- Confident Learning: Estimating Uncertainty in Dataset LabelsJournal of Artificial Intelligence Research, 2021
- Uninformed Students: Student-Teacher Anomaly Detection With Discriminative Latent Embeddings2020
- Visual-Based Defect Detection and Classification Approaches for Industrial Applications—A SURVEYSensors, 2020
- Automatic Fabric Defect Detection with a Multi-Scale Convolutional Denoising Autoencoder Network ModelSensors, 2018
- Making Deep Neural Networks Robust to Label Noise: A Loss Correction Approach2017
- Steel defect classification with Max-Pooling Convolutional Neural Networks2012
- Learning from Imbalanced DataIEEE Transactions on Knowledge and Data Engineering, 2009