Fault detection method based on high-level feedback and reconstruction for image recognition

Image recognition problem is a typical challenge for artificial intelligence. Deep learning methods have shown great progress in this area in recent years. However, this approach has the limitation of outputting wrong result with high confidence when the input is not well represented in the training...

Full description

Saved in:
Bibliographic Details
Main Author: Xu, Jin
Other Authors: Xie Ming
Format: Thesis-Doctor of Philosophy
Language:English
Published: Nanyang Technological University 2021
Subjects:
Online Access:https://hdl.handle.net/10356/146731
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-146731
record_format dspace
spelling sg-ntu-dr.10356-1467312023-03-11T16:58:10Z Fault detection method based on high-level feedback and reconstruction for image recognition Xu, Jin Xie Ming School of Mechanical and Aerospace Engineering mmxie@ntu.edu.sg Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Image recognition problem is a typical challenge for artificial intelligence. Deep learning methods have shown great progress in this area in recent years. However, this approach has the limitation of outputting wrong result with high confidence when the input is not well represented in the training data, such as adversarial examples (natural images with intentionally designed small perturbations that are imperceptible by human beings) and unusual images that is not covered in the training. This limitation causes great safety concerns for the deployment of existing deep learning systems for critical applications such as autonomous driving and medical diagnosis. However, it is not practical to include all adversarial or unusual images in training because images are highly variable. It is therefore desirable to design an image recognition system with the ability of fault detection. When an “unseen” image, either adversarial or unusual, is input to the system, the fault detection mechanism recognizes the situation and raise a red flag. With the fault detected, further steps such as intervention like double checking by a person can be taken to avoid damages. Inspired by the human perception system, the research reported in this thesis focusses on a novel fault detection mechanism and a reconstruction process. This mechanism, named Detection Based on Feedback and Reconstruction (DBFR), requires no change of the existing recognition system and generates a reconstructed image as an intuitive “internal model” to indicate whether the interpretation is supported by what is learned in the training while performing the inference. By comparing the reconstructed image with the input image, the incorrect interpretation can be detected with high probability. Experiments validate the superior effectiveness of this method over existing defense methods and the reconstructed images help to interpret the internal decision process of the deep learning system, which is currently treated as a black box in most applications. Doctor of Philosophy 2021-03-09T01:29:20Z 2021-03-09T01:29:20Z 2021 Thesis-Doctor of Philosophy Xu, J. (2021). Fault detection method based on high-level feedback and reconstruction for image recognition. Doctoral thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/146731 10.32657/10356/146731 en This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
spellingShingle Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
Xu, Jin
Fault detection method based on high-level feedback and reconstruction for image recognition
description Image recognition problem is a typical challenge for artificial intelligence. Deep learning methods have shown great progress in this area in recent years. However, this approach has the limitation of outputting wrong result with high confidence when the input is not well represented in the training data, such as adversarial examples (natural images with intentionally designed small perturbations that are imperceptible by human beings) and unusual images that is not covered in the training. This limitation causes great safety concerns for the deployment of existing deep learning systems for critical applications such as autonomous driving and medical diagnosis. However, it is not practical to include all adversarial or unusual images in training because images are highly variable. It is therefore desirable to design an image recognition system with the ability of fault detection. When an “unseen” image, either adversarial or unusual, is input to the system, the fault detection mechanism recognizes the situation and raise a red flag. With the fault detected, further steps such as intervention like double checking by a person can be taken to avoid damages. Inspired by the human perception system, the research reported in this thesis focusses on a novel fault detection mechanism and a reconstruction process. This mechanism, named Detection Based on Feedback and Reconstruction (DBFR), requires no change of the existing recognition system and generates a reconstructed image as an intuitive “internal model” to indicate whether the interpretation is supported by what is learned in the training while performing the inference. By comparing the reconstructed image with the input image, the incorrect interpretation can be detected with high probability. Experiments validate the superior effectiveness of this method over existing defense methods and the reconstructed images help to interpret the internal decision process of the deep learning system, which is currently treated as a black box in most applications.
author2 Xie Ming
author_facet Xie Ming
Xu, Jin
format Thesis-Doctor of Philosophy
author Xu, Jin
author_sort Xu, Jin
title Fault detection method based on high-level feedback and reconstruction for image recognition
title_short Fault detection method based on high-level feedback and reconstruction for image recognition
title_full Fault detection method based on high-level feedback and reconstruction for image recognition
title_fullStr Fault detection method based on high-level feedback and reconstruction for image recognition
title_full_unstemmed Fault detection method based on high-level feedback and reconstruction for image recognition
title_sort fault detection method based on high-level feedback and reconstruction for image recognition
publisher Nanyang Technological University
publishDate 2021
url https://hdl.handle.net/10356/146731
_version_ 1761781333484371968