{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/112952"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/112952","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Exploring anomaly detection methods using features extracted with a neural network","abstract":"Anomaly (or outlier) detection is a problem with the goal of detecting outliers—or anomalous samples—within a dataset or distribution. Anomaly detection has important applications in computer vision, cybersecurity, biomedical imaging, and more. As the data becomes bigger (more samples collected) and more complex (more variables observed), it becomes increasingly difficult to detect outliers within a dataset. In this thesis, we explore the problem of outlier detection in computer vision applications, in which the input images can be interpreted as vectors with hundreds of dimensions. While an effective anomaly detection method is important to achieve good performance in detecting outliers, it is also often important to first extract useful features from the data. We explore both extracting useful features from the data (using neural networks) and the use of different anomaly detection methods that use the extracted features. We first train a neural network on the training set (containing only inlier images) with the goal of image classification, and then obtain feature vectors for both the training and test sets by performing a forward pass of the training and test images through the trained network. We then use an anomaly detection method—such as nearest subspace distance or Mahalanobis distance—that makes use of the extracted features in order to obtain an anomaly score for each image in the test set, which can then be used to detect the outliers in the test set. We show that this outlier detection scheme (i.e. using an anomaly detection method on features extracted using a neural network) often outperforms the use of the maximum softmax probability (MSP) as an anomaly score. We also demonstrate that the features extracted using neural networks often lead to better anomaly detection performance than the features obtained by simply reducing the dimensions of the input images (e.g. using principal component analysis).","abstract_html":"Anomaly (or outlier) detection is a problem with the goal of detecting outliers—or anomalous samples—within a dataset or distribution. Anomaly detection has important applications in computer vision, cybersecurity, biomedical imaging, and more. As the data becomes bigger (more samples collected) and more complex (more variables observed), it becomes increasingly difficult to detect outliers within a dataset. In this thesis, we explore the problem of outlier detection in computer vision applications, in which the input images can be interpreted as vectors with hundreds of dimensions. While an effective anomaly detection method is important to achieve good performance in detecting outliers, it is also often important to first extract useful features from the data. We explore both extracting useful features from the data (using neural networks) and the use of different anomaly detection methods that use the extracted features. We first train a neural network on the training set (containing only inlier images) with the goal of image classification, and then obtain feature vectors for both the training and test sets by performing a forward pass of the training and test images through the trained network. We then use an anomaly detection method—such as nearest subspace distance or Mahalanobis distance—that makes use of the extracted features in order to obtain an anomaly score for each image in the test set, which can then be used to detect the outliers in the test set. We show that this outlier detection scheme (i.e. using an anomaly detection method on features extracted using a neural network) often outperforms the use of the maximum softmax probability (MSP) as an anomaly score. We also demonstrate that the features extracted using neural networks often lead to better anomaly detection performance than the features obtained by simply reducing the dimensions of the input images (e.g. using principal component analysis).","abstract_has_math":false,"creators":["Yeom, Jaewook"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Electrical & Computer Engr","degree_department":null,"school":null,"contributors":["Bresler, Yoram"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022-01-12T21:40:31Z","date_published":"2022-01-12T21:40:31Z","updated_at":"2026-07-22T22:24:52Z","subjects":["Anomaly detection","outlier detection","neural networks","computer vision","information theory","dimensionality reduction","random projection","coding rate"],"languages":["en"],"rights":["Copyright 2021 Jaewook Yeom"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/112952","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Bresler, Yoram"]},{"key":"dc:creator","label":"Author","values":["Yeom, Jaewook"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2022-01-12T21:40:31Z","2021-05-26","2021-08"]},{"key":"dc:type","label":"Dc Type","values":["text","Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical & Computer Engr"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Anomaly detection","outlier detection","neural networks","computer vision","information theory","dimensionality reduction","random projection","coding rate"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2021 Jaewook Yeom"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/112952"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Anomaly (or outlier) detection is a problem with the goal of detecting outliers—or anomalous samples—within a dataset or distribution. Anomaly detection has important applications in computer vision, cybersecurity, biomedical imaging, and more. As the data becomes bigger (more samples collected) and more complex (more variables observed), it becomes increasingly difficult to detect outliers within a dataset. In this thesis, we explore the problem of outlier detection in computer vision applications, in which the input images can be interpreted as vectors with hundreds of dimensions. While an effective anomaly detection method is important to achieve good performance in detecting outliers, it is also often important to first extract useful features from the data. We explore both extracting useful features from the data (using neural networks) and the use of different anomaly detection methods that use the extracted features. We first train a neural network on the training set (containing only inlier images) with the goal of image classification, and then obtain feature vectors for both the training and test sets by performing a forward pass of the training and test images through the trained network. We then use an anomaly detection method—such as nearest subspace distance or Mahalanobis distance—that makes use of the extracted features in order to obtain an anomaly score for each image in the test set, which can then be used to detect the outliers in the test set. We show that this outlier detection scheme (i.e. using an anomaly detection method on features extracted using a neural network) often outperforms the use of the maximum softmax probability (MSP) as an anomaly score. We also demonstrate that the features extracted using neural networks often lead to better anomaly detection performance than the features obtained by simply reducing the dimensions of the input images (e.g. using principal component analysis).","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-01-12 without embargo terms","The student, Jaewook Yeom, accepted the attached license on 2021-05-25 at 09:52.","The student, Jaewook Yeom, submitted this Thesis for approval on 2021-05-25 at 10:22.","This Thesis was approved for publication on 2021-05-26 at 13:24.","DSpace SAF Submission Ingestion Package generated from Vireo submission #16657 on 2022-01-12 at 12:42:48","Made available in DSpace on 2022-01-12T21:40:31Z (GMT). No. of bitstreams: 3 YEOM-THESIS-2021.pdf: 426977 bytes, checksum: 41492bdd59b287a4bd7df316ad3af524 (MD5) MS Thesis Latex Source File.tex: 178926 bytes, checksum: d3b66a45e4dfc7cfbe8db86efabb41aa (MD5) LICENSE.txt: 4209 bytes, checksum: cfacd3a03ad50f9c3503bb28a9531273 (MD5) Previous issue date: 2021-05-26"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Exploring anomaly detection methods using features extracted with a neural network"]}]}],"canonical_facts":{"dc:contributor":["Bresler, Yoram"],"dc:creator":["Yeom, Jaewook"],"dc:date":["2022-01-12T21:40:31Z","2021-05-26","2021-08"],"dc:description":["Anomaly (or outlier) detection is a problem with the goal of detecting outliers—or anomalous samples—within a dataset or distribution. Anomaly detection has important applications in computer vision, cybersecurity, biomedical imaging, and more. As the data becomes bigger (more samples collected) and more complex (more variables observed), it becomes increasingly difficult to detect outliers within a dataset. In this thesis, we explore the problem of outlier detection in computer vision applications, in which the input images can be interpreted as vectors with hundreds of dimensions. While an effective anomaly detection method is important to achieve good performance in detecting outliers, it is also often important to first extract useful features from the data. We explore both extracting useful features from the data (using neural networks) and the use of different anomaly detection methods that use the extracted features. We first train a neural network on the training set (containing only inlier images) with the goal of image classification, and then obtain feature vectors for both the training and test sets by performing a forward pass of the training and test images through the trained network. We then use an anomaly detection method—such as nearest subspace distance or Mahalanobis distance—that makes use of the extracted features in order to obtain an anomaly score for each image in the test set, which can then be used to detect the outliers in the test set. We show that this outlier detection scheme (i.e. using an anomaly detection method on features extracted using a neural network) often outperforms the use of the maximum softmax probability (MSP) as an anomaly score. We also demonstrate that the features extracted using neural networks often lead to better anomaly detection performance than the features obtained by simply reducing the dimensions of the input images (e.g. using principal component analysis).","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-01-12 without embargo terms","The student, Jaewook Yeom, accepted the attached license on 2021-05-25 at 09:52.","The student, Jaewook Yeom, submitted this Thesis for approval on 2021-05-25 at 10:22.","This Thesis was approved for publication on 2021-05-26 at 13:24.","DSpace SAF Submission Ingestion Package generated from Vireo submission #16657 on 2022-01-12 at 12:42:48","Made available in DSpace on 2022-01-12T21:40:31Z (GMT). No. of bitstreams: 3 YEOM-THESIS-2021.pdf: 426977 bytes, checksum: 41492bdd59b287a4bd7df316ad3af524 (MD5) MS Thesis Latex Source File.tex: 178926 bytes, checksum: d3b66a45e4dfc7cfbe8db86efabb41aa (MD5) LICENSE.txt: 4209 bytes, checksum: cfacd3a03ad50f9c3503bb28a9531273 (MD5) Previous issue date: 2021-05-26"],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/112952"],"dc:language":["en"],"dc:rights":["Copyright 2021 Jaewook Yeom"],"dc:subject":["Anomaly detection","outlier detection","neural networks","computer vision","information theory","dimensionality reduction","random projection","coding rate"],"dc:title":["Exploring anomaly detection methods using features extracted with a neural network"],"dc:type":["text","Thesis"],"thesis:degree_discipline":["Electrical & Computer Engr"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:24:52Z"}