Please use this identifier to cite or link to this item:
http://hdl.handle.net/1893/35268
Appears in Collections: | Computing Science and Mathematics Conference Papers and Proceedings |
Author(s): | Swingler, Kevin |
Contact Email: | kevin.swingler@stir.ac.uk |
Title: | A Suite of Incremental Image Degradation Operators for Testing Image Classification Algorithms |
Editor(s): | Back, Thomas van Stein, Bas Wagner, Christian Garibaldi, Jonathan Lam, H K Cottrell, Marie Doctor, Faiyaz Filipe, Joaquim Warwick, Kevin Kacprzyk, Janusz |
Citation: | Swingler K (2022) A Suite of Incremental Image Degradation Operators for Testing Image Classification Algorithms. In: Back T, van Stein B, Wagner C, Garibaldi J, Lam HK, Cottrell M, Doctor F, Filipe J, Warwick K & Kacprzyk J (eds.) 14th International Conference on Neural Computation Theory and Applications, Valletta, Malta, 24.10.2022-26.10.2022. Setubal, Portugal: SCITEPRESS - Science and Technology Publications, pp. 262-272. https://doi.org/10.5220/0011511000003332 |
Issue Date: | 2022 |
Date Deposited: | 23-Mar-2023 |
Conference Name: | 14th International Conference on Neural Computation Theory and Applications |
Conference Dates: | 2022-10-24 - 2022-10-26 |
Conference Location: | Valletta, Malta |
Abstract: | Convolutional Neural Networks (CNN) are extremely popular for modelling sound and images, but they suffer from a lack of robustness that could threaten their usefulness in applications where reliability is important. Recent studies have shown how it is possible to maliciously create adversarial images—those that appear to the human observer as perfect examples of one class but that fool a CNN into assigning them to a different, incorrect class. It takes some effort to make these images as they need to be designed specifically to fool a given network. In this paper we show that images can be degraded in a number of simple ways that do not need careful design and that would not affect the ability of a human observer, but which cause severe deterioration in the performance of three different CNN models. We call the speed of the deterioration in performance due to incremental degradations in image quality the degradation profile of a model and argue that reporting the degradation profile is as important as reporting performance on clean images. |
Status: | VoR - Version of Record |
Rights: | Licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International http://creativecommons.org/licenses/by-nc-nd/4.0/ |
Licence URL(s): | http://creativecommons.org/licenses/by-nc-nd/4.0/ |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
115110.pdf | Fulltext - Published Version | 2.82 MB | Adobe PDF | View/Open |
This item is protected by original copyright |
A file in this item is licensed under a Creative Commons License
Items in the Repository are protected by copyright, with all rights reserved, unless otherwise indicated.
The metadata of the records in the Repository are available under the CC0 public domain dedication: No Rights Reserved https://creativecommons.org/publicdomain/zero/1.0/
If you believe that any material held in STORRE infringes copyright, please contact library@stir.ac.uk providing details and we will remove the Work from public display in STORRE and investigate your claim.