SIFT and SURF based feature extraction for the anomaly detection
Abstract
In this paper, we suggest a way to use SIFT and SURF algorithms to extract the image features for anomaly detection. We use those feature vectors to train various classifiers on a real-world dataset in the semi-supervised (with a small number of faulty samples) manner with a large number of classifiers and in the one-class (with no faulty samples) manner using the SVDD and SVM classifier. We prove, that the SIFT and SURF algorithms could be used as feature extractors, that they could be used to train a semi-supervised and one-class classifier with an accuracy around 89% and that the performance of the one-class classifier could be comparable to the semi-supervised one. We also made our dataset and source code publicly available.
Keywords
Anomaly detection, Object descriptors, Machine Learning, SIFT, SURFPersistent identifier
http://hdl.handle.net/11012/209385Document type
Peer reviewedDocument version
Final PDFSource
Proceedings I of the 28st Conference STUDENT EEICT 2022: General papers. s. 459-464. ISBN 978-80-214-6029-4https://conf.feec.vutbr.cz/eeict/index/pages/view/ke_stazeni
Collections
- Student EEICT 2022 [108]