Anomalous Behaviour Detection using Transfer Learning Algorithm of Series and DAG Network

Razak, Hana’ Abd; Almisreb, Ali Abd; Tahir, Nooritawati Md

doi:10.1109/icsengt.2019.8906470

Cited by 3 publications

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Transfer learning is a machine learning method that leverages knowledge acquired in one environment to improve performance in another, even when modeling a different set of tasks. Transfer learning has greatly benefited deep learning with image data by providing the best architecture of pre-trained CNN models and enabling better results with fewer data [7,20,21].…”

Section: Transfer Learningmentioning

confidence: 99%

Real Time Snatch Theft Detection using Deep Learning Networks

Zamri¹,

Tahir²,

ALI³

et al. 2023

ARASET

View full text Add to dashboard Cite

Snatch theft is a common crime in urban areas that poses a serious threat to public safety. It involves forcefully grabbing a victim's personal belongings, such as purses or mobile phones, before quickly fleeing the scene. Detecting snatch theft incidents in real-time is a challenging task due to the speed at which they occur. The current methods used to detect snatch theft incidents rely heavily on human intervention, which can lead to significant delays and potential errors. Therefore, there is a need for an automated technique that can accurately and efficiently detect these incidents in real-time. Hence, the study aims to detect snatch theft using a transfer learning approach based on eight pre-trained convolutional neural networks (CNNs) as classifiers: AlexNet, VGG16, VGG19, GoogleNet, InceptionV3, ResNet-18, ResNet-50, and ResNet-101. The modified pre-trained CNN models are evaluated in both offline and real-time modes. Based on the offline mode, VGG19 achieved 100% training accuracy, and ResNet50 had the highest testing accuracy of 98.9%. In the offline mode, all models accurately classified normal scenes, with ResNet-10 having the lowest false negative rate and ResNet-50 achieving the lowest false positive rate with only 44 misclassified anomaly frames related to snatch theft. The study further evaluated and validated the eight models in real-time mode, and the results showed that AlexNet and ResNet-18 were the only models capable of categorizing snatch theft scenarios with promising findings.

show abstract

Section: Transfer Learningmentioning

confidence: 99%

Real Time Snatch Theft Detection using Deep Learning Networks

Zamri¹,

Tahir²,

ALI³

et al. 2023

ARASET

View full text Add to dashboard Cite

show abstract

A Review of Video Object Detection: Datasets, Metrics and Methods

Zhu

Wei

et al. 2020

Applied Sciences

View full text Add to dashboard Cite

Although there are well established object detection methods based on static images, their application to video data on a frame by frame basis faces two shortcomings: (i) lack of computational efficiency due to redundancy across image frames or by not using a temporal and spatial correlation of features across image frames, and (ii) lack of robustness to real-world conditions such as motion blur and occlusion. Since the introduction of the ImageNet Large Scale Visual Recognition Challenge (ILSVRC) in 2015, a growing number of methods have appeared in the literature on video object detection, many of which have utilized deep learning models. The aim of this paper is to provide a review of these papers on video object detection. An overview of the existing datasets for video object detection together with commonly used evaluation metrics is first presented. Video object detection methods are then categorized and a description of each of them is stated. Two comparison tables are provided to see their differences in terms of both accuracy and computational efficiency. Finally, some future trends in video object detection to address the challenges involved are noted.

show abstract

Big Data Detection utilizing Cloud Networks with Video Vision Techniques

Ahmed

Aljuboori

2023

EAI Endorsed Scal Inf Syst

View full text Add to dashboard Cite

Regardless of the number of grounded object identification procedures reliant upon still pictures, their application to edge video information through the system hypothesis faces two drawbacks: (1) the deficit of computational throughput in view of abundance across picture follows or through the shortfall of usage of a transient and spatial relationship for parts across the edges of the image, and (ii) a shortfall of energy for authentic conditions, e.g., muddled turn of events and impediment. Since the Visual Recognition challenge has been by and large introduced, different methods have emerged recorded as a printed version around video object distinguishing proof, countless which have used significant learning norms. The mark of this assessment is to present a twofold framework for a total investigation of the principle methodologies of video object acknowledgment regardless the methodology of murkiness associations. It presents a chart of existing datasets for video object location close by appraisal estimations ordinarily used connected with fleecy frameworks organization methodologies. The video data acknowledgment advancements are then arranged and each one imparted. Two test tables are given to know the differences between them to the extent that accuracy and math ability. Finally, a couple of future examples in video object recognition have been believed to address embedded difficulties.

show abstract

Anomalous Behaviour Detection using Transfer Learning Algorithm of Series and DAG Network

Cited by 3 publications

References 15 publications

Real Time Snatch Theft Detection using Deep Learning Networks

Real Time Snatch Theft Detection using Deep Learning Networks

A Review of Video Object Detection: Datasets, Metrics and Methods

Big Data Detection utilizing Cloud Networks with Video Vision Techniques

Contact Info

Product

Resources

About