Unsupervised Anomaly Approach to Pedestrian Age Classification from Surveillance Cameras Using an Adversarial Model with Skip-Connections

Anomaly detection is an active research area within the machine learning and scene understanding fields. Despite the ambiguous definition, anomaly detection is considered an outlier detection in a given data based on normality constraints. The biggest problem in real-world anomaly detection applicat...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Husnu Baris Baydargil, Jangsik Park, Ibrahim Furkan Ince
Formato: article
Lenguaje:EN
Publicado: MDPI AG 2021
Materias:
T
Acceso en línea:https://doaj.org/article/c41d3f73adc04e91b5517a74029630e8
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:c41d3f73adc04e91b5517a74029630e8
record_format dspace
spelling oai:doaj.org-article:c41d3f73adc04e91b5517a74029630e82021-11-11T15:00:25ZUnsupervised Anomaly Approach to Pedestrian Age Classification from Surveillance Cameras Using an Adversarial Model with Skip-Connections10.3390/app112199042076-3417https://doaj.org/article/c41d3f73adc04e91b5517a74029630e82021-10-01T00:00:00Zhttps://www.mdpi.com/2076-3417/11/21/9904https://doaj.org/toc/2076-3417Anomaly detection is an active research area within the machine learning and scene understanding fields. Despite the ambiguous definition, anomaly detection is considered an outlier detection in a given data based on normality constraints. The biggest problem in real-world anomaly detection applications is the high bias of the available data due to the class imbalance, meaning a limited amount of all possible anomalous and normal samples, thus making supervised learning model use difficult. This paper introduces an unsupervised and adversarially trained anomaly model with a unique encoder–decoder structure to address this issue. The proposed model distinguishes different age groups of people—namely child, adult, and elderly—from surveillance camera data in Busan, Republic of Korea. The proposed model has three major parts: a parallel-pipeline encoder with a conventional convolutional neural network and a dilated-convolutional neural network. The latent space vectors created at the end of both networks are concatenated. While the convolutional pipeline extracts local features, the dilated convolutional pipeline extracts the global features from the same input image. Concatenation of these features is sent as the input into the decoder, which has partial skip-connection elements from both pipelines. This, along with the concatenated feature vector, improves feature diversity. The input image is reconstructed from the feature vector through the stacked transpose convolution layers. Afterward, both the original input image and the corresponding reconstructed image are sent into the discriminator and are distinguished as real or fake. The image reconstruction loss and its corresponding latent space loss are considered for the training of the model and the adversarial Wasserstein loss. Only normal-designated class images are used during the training. The hypothesis is that if the model is trained with normal class images, then during the inference, the construction loss will be minimal. On the other hand, if the untrained anomalous class images are input through the model, the reconstruction loss value will be very high. This method is applied to distinguish different age clusters of people using unsupervised training. The proposed model outperforms the benchmark models in both the qualitative and the quantitative measurements.Husnu Baris BaydargilJangsik ParkIbrahim Furkan InceMDPI AGarticleanomaly detectioncomputer visionsurveillancedeep learninggenerative adversarial networksTechnologyTEngineering (General). Civil engineering (General)TA1-2040Biology (General)QH301-705.5PhysicsQC1-999ChemistryQD1-999ENApplied Sciences, Vol 11, Iss 9904, p 9904 (2021)
institution DOAJ
collection DOAJ
language EN
topic anomaly detection
computer vision
surveillance
deep learning
generative adversarial networks
Technology
T
Engineering (General). Civil engineering (General)
TA1-2040
Biology (General)
QH301-705.5
Physics
QC1-999
Chemistry
QD1-999
spellingShingle anomaly detection
computer vision
surveillance
deep learning
generative adversarial networks
Technology
T
Engineering (General). Civil engineering (General)
TA1-2040
Biology (General)
QH301-705.5
Physics
QC1-999
Chemistry
QD1-999
Husnu Baris Baydargil
Jangsik Park
Ibrahim Furkan Ince
Unsupervised Anomaly Approach to Pedestrian Age Classification from Surveillance Cameras Using an Adversarial Model with Skip-Connections
description Anomaly detection is an active research area within the machine learning and scene understanding fields. Despite the ambiguous definition, anomaly detection is considered an outlier detection in a given data based on normality constraints. The biggest problem in real-world anomaly detection applications is the high bias of the available data due to the class imbalance, meaning a limited amount of all possible anomalous and normal samples, thus making supervised learning model use difficult. This paper introduces an unsupervised and adversarially trained anomaly model with a unique encoder–decoder structure to address this issue. The proposed model distinguishes different age groups of people—namely child, adult, and elderly—from surveillance camera data in Busan, Republic of Korea. The proposed model has three major parts: a parallel-pipeline encoder with a conventional convolutional neural network and a dilated-convolutional neural network. The latent space vectors created at the end of both networks are concatenated. While the convolutional pipeline extracts local features, the dilated convolutional pipeline extracts the global features from the same input image. Concatenation of these features is sent as the input into the decoder, which has partial skip-connection elements from both pipelines. This, along with the concatenated feature vector, improves feature diversity. The input image is reconstructed from the feature vector through the stacked transpose convolution layers. Afterward, both the original input image and the corresponding reconstructed image are sent into the discriminator and are distinguished as real or fake. The image reconstruction loss and its corresponding latent space loss are considered for the training of the model and the adversarial Wasserstein loss. Only normal-designated class images are used during the training. The hypothesis is that if the model is trained with normal class images, then during the inference, the construction loss will be minimal. On the other hand, if the untrained anomalous class images are input through the model, the reconstruction loss value will be very high. This method is applied to distinguish different age clusters of people using unsupervised training. The proposed model outperforms the benchmark models in both the qualitative and the quantitative measurements.
format article
author Husnu Baris Baydargil
Jangsik Park
Ibrahim Furkan Ince
author_facet Husnu Baris Baydargil
Jangsik Park
Ibrahim Furkan Ince
author_sort Husnu Baris Baydargil
title Unsupervised Anomaly Approach to Pedestrian Age Classification from Surveillance Cameras Using an Adversarial Model with Skip-Connections
title_short Unsupervised Anomaly Approach to Pedestrian Age Classification from Surveillance Cameras Using an Adversarial Model with Skip-Connections
title_full Unsupervised Anomaly Approach to Pedestrian Age Classification from Surveillance Cameras Using an Adversarial Model with Skip-Connections
title_fullStr Unsupervised Anomaly Approach to Pedestrian Age Classification from Surveillance Cameras Using an Adversarial Model with Skip-Connections
title_full_unstemmed Unsupervised Anomaly Approach to Pedestrian Age Classification from Surveillance Cameras Using an Adversarial Model with Skip-Connections
title_sort unsupervised anomaly approach to pedestrian age classification from surveillance cameras using an adversarial model with skip-connections
publisher MDPI AG
publishDate 2021
url https://doaj.org/article/c41d3f73adc04e91b5517a74029630e8
work_keys_str_mv AT husnubarisbaydargil unsupervisedanomalyapproachtopedestrianageclassificationfromsurveillancecamerasusinganadversarialmodelwithskipconnections
AT jangsikpark unsupervisedanomalyapproachtopedestrianageclassificationfromsurveillancecamerasusinganadversarialmodelwithskipconnections
AT ibrahimfurkanince unsupervisedanomalyapproachtopedestrianageclassificationfromsurveillancecamerasusinganadversarialmodelwithskipconnections
_version_ 1718437926295044096