Powered by OpenAIRE graph
Found an issue? Give us feedback
OpenMETUarrow_drop_down
OpenMETU
Master thesis . 2025
License: CC BY NC ND
Data sources: OpenMETU
addClaim

This Research product is the result of merged Research products in OpenAIRE.

You have already added 0 works in your ORCID record related to the merged Research product.

A NOVEL DEPTHWISE CONVOLUTIONAL VARIATIONAL AUTOENCODER FOR ANOMALY DETECTION IN COMPLEX TRAFFIC SCENARIOS FROM UAV VIEWS

İHA görünümlerinden Karmaşık Trafik Senaryolarında Anomali Tespiti için Yeni Bir Derinlik Yönlü Evrişimsel Varyasyonel Otokodlayıcı
Authors: Saleem, Arslan;

A NOVEL DEPTHWISE CONVOLUTIONAL VARIATIONAL AUTOENCODER FOR ANOMALY DETECTION IN COMPLEX TRAFFIC SCENARIOS FROM UAV VIEWS

Abstract

Trafik anomali tespiti (AD), akıllı gözetim sistemlerinde kamu güvenliğini artırmak, riskleri azaltmak ve hızlı müdahale sağlamak için hayati öneme sahiptir. Özellikle dronlar kullanılarak yapılan havadan trafik izleme, dinamik kentsel ortamlar gibi zorlukların üstesinden gelme potansiyeli nedeniyle dikkat çekmiş, ancak henüz yeterince keşfedilmemiştir. Drone ile çekilen videolarda anormallikleri tespit etmek, nadir olaylar, küçük ve örtüşen nesneler, çok ölçekli hedefler ve karmaşık arka planlar gibi benzersiz engelleri beraberinde getirir. Bu zorlukların üstesinden gelmek için, drone tabanlı trafik gözetiminde anomali tespitini geliştirmek üzere tasarlanmış Derinlemesine Evrişimli Varyasyonel Otomatik Kodlayıcı (DwCVAE) adında yeni bir model öneriyoruz. DwCVAE, verimli ve ayrıntılı özellik çıkarımı sağlayan, modelin ince ve çok ölçekli anomalilere karşı hassasiyetini artıran derinlemesine evrişimlerden yararlanır. Varyasyonel otomatik kodlayıcı (VAE) mimarisi üzerine inşa edilen DwCVAE, normal trafik modellerini yakalayan kompakt gizli gösterimler oluşturarak sapmaların güvenilir bir şekilde tespit edilmesini sağlar. Bu derinlemesine yaklaşım, hem hesaplama verimliliğini hem de tespit doğruluğunu optimize eden temel bir yenilik işaret etmektedir. DwCVAE'nin etkinliğini sistematik olarak değerlendirmek için dört ek model tasarladık: Evrişimli Varyasyonel Otomatik Kodlayıcı (CVAE), Genişletilmiş Evrişimli VAE (DCVAE), Ayrılabilir Evrişimli VAE (SCVAE) ve Evrişimli LSTM VAE (CLSTMVAE). Ek olarak, DwCVAE'yi Drone-Anomaly ve UIT-Adrone adlı iki kıyaslama veri kümesinde en son teknolojiye sahip zayıf denetimli ve denetimsiz modellerle karşılaştırdık. Deneysel sonuçlar, DwCVAE'nin rakip modelleri geride bıraktığını, Drone-Anomaly ve UIT-Adrone üzerinde sırasıyla 74.95 ve 79.77 AUC ile 0.30 ve 0.27 EER elde ettiğini göstererek karmaşık havadan gözetim görevlerinde üstün performansını ortaya koymaktadır.

Traffic anomaly detection (AD) is essential for improving public safety, reducing risks, and enabling quick responses in intelligent surveillance systems. Aerial traffic monitoring, particularly using drones, has gained attention due to its potential to address challenges like dynamic urban environments, yet it remains underexplored. Detecting anomalies in drone captured video involves unique obstacles: rare events, small and overlapping objects, multi-scale targets, and complex backgrounds. To address these challenges, we propose the Depthwise Convolutional Variational Autoencoder (DwCVAE), a novel model designed to enhance AD in drone-based traffic surveillance. DwCVAE leverages depthwise convolutions, which allow efficient and detailed feature extraction, improving model sensitivity to subtle and multi-scale anomalies. Built on variational autoencoder (VAE) architecture, DwCVAE creates compact latent representations that capture normal traffic patterns, enabling reliable detection of deviations. This depthwise approach marks a key innovation, optimizing both computational efficiency and detection accuracy. We design four additional models: Convolutional Variational Autoencoder (CVAE), Dilated Convolutional VAE (DCVAE), Separable Convolutional VAE (SCVAE), and Convolutional LSTM VAE (CLSTMVAE) to systematically assess the effectiveness of DwCVAE. Additionally, we evaluate DwCVAE against state-of-the-art weakly supervised and unsupervised models on two benchmark datasets, Drone-Anomaly and UIT-Adrone. Experimental results show that DwCVAE outperforms competing models, achieving an AUC of 74.95 and 79.77 and an EER of 0.30 and 0.27 on Drone-Anomaly and UITAdrone, respectively, demonstrating its superior performance in complex aerial surveillance tasks.

Related Organizations
Keywords

Variational Autoencoder, Anomaly Detection, Unsupervised Learning, Depthwise Convolutional, Traffic Surveillance

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Upload OA version
Are you the author of this publication? Upload your Open Access version to Zenodo!
It’s fast and easy, just two clicks!