Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

A Generative Adversarial Network Based Framework For Unsupervised Visual Surface Inspection

Abstract: 

Visual surface inspection is a challenging task due to the highly inconsistent appearance of the target surfaces and the abnormal regions. Most of the state-of-the-art methods are highly dependent on the labelled training samples, which are difficult to collect in practical industrial applications. To address this problem, we propose a generative adversarial network based framework for unsupervised surface inspection. The generative adversarial network is trained to generate the fake images analogous to the normal surface images. It implies that a well-trained GAN indeed learns a good representation of the normal surface images in a latent feature space. And consequently, the discriminator of GAN can serve as a naturally one-class classifier. We use the first three conventional layer of the discriminator as the feature extractor, whose response is sensitive to the abnormal regions. Particularly, a multi-scale fusion strategy is adopted to fuse the responses of the three convolution layers and thus improve the segmentation performance of abnormal detection. Various experimental results demonstrate the effectiveness of our proposed method.

up
0 users have voted:

Paper Details

Authors:
Wei Zhai,Jiang Zhu,Yang Cao, Zengfu Wang
Submitted On:
13 April 2018 - 5:01am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Wei Zhai
Paper Code:
2935
Document Year:
2018
Cite

Document Files

This is the presentation slides of ICASSP 2018.

(8 downloads)

Subscribe

[1] Wei Zhai,Jiang Zhu,Yang Cao, Zengfu Wang, "A Generative Adversarial Network Based Framework For Unsupervised Visual Surface Inspection", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2660. Accessed: Apr. 20, 2018.
@article{2660-18,
url = {http://sigport.org/2660},
author = {Wei Zhai;Jiang Zhu;Yang Cao; Zengfu Wang },
publisher = {IEEE SigPort},
title = {A Generative Adversarial Network Based Framework For Unsupervised Visual Surface Inspection},
year = {2018} }
TY - EJOUR
T1 - A Generative Adversarial Network Based Framework For Unsupervised Visual Surface Inspection
AU - Wei Zhai;Jiang Zhu;Yang Cao; Zengfu Wang
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2660
ER -
Wei Zhai,Jiang Zhu,Yang Cao, Zengfu Wang. (2018). A Generative Adversarial Network Based Framework For Unsupervised Visual Surface Inspection. IEEE SigPort. http://sigport.org/2660
Wei Zhai,Jiang Zhu,Yang Cao, Zengfu Wang, 2018. A Generative Adversarial Network Based Framework For Unsupervised Visual Surface Inspection. Available at: http://sigport.org/2660.
Wei Zhai,Jiang Zhu,Yang Cao, Zengfu Wang. (2018). "A Generative Adversarial Network Based Framework For Unsupervised Visual Surface Inspection." Web.
1. Wei Zhai,Jiang Zhu,Yang Cao, Zengfu Wang. A Generative Adversarial Network Based Framework For Unsupervised Visual Surface Inspection [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2660