Detection of anomaly in surveillance videos using quantum convolutional neural networks

Amin, Javeria; Anjum, Muhammad Almas; Ibrar, Kainat; Sharif, Muhammad; Kadry, Seifedine; Crespo, Rubén González

doi:10.1016/j.imavis.2023.104710

Cited by 23 publications

(6 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By using the Inception model as the visual backbone, the hybrid system can effectively process and understand the visual information from the surveillance footage. In addition, the Inception model has been used and showed exceptional performance for human activity recognition tasks [2,3,12,28].…”

Section: Proposed Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Shoplifting Detection Using Hybrid Neural Network CNN-BiLSMT and Development of Benchmark Dataset

et al. 2023

View full text Add to dashboard Cite

Shoplifting poses a significant challenge for shop owners as well as other stakeholders, including law enforcement agencies. In recent years, the task of shoplifting detection has gained the interest of researchers due to video surveillance generating vast quantities of data that cannot be processed in real-time by human staff. In previous studies, different datasets and methods have been developed for the task of shoplifting detection. However, there is a lack of a large benchmark dataset containing different behaviors of shoplifting and standard methods for the task of shoplifting detection. To overcome this limitation, in this study, a large benchmark dataset has been developed, having 900 instances with 450 cases of shoplifting and 450 of non-shoplifting with manual annotation based on five different ways of shoplifting. Moreover, a method for the detection of shoplifting is proposed for evaluating the developed dataset. The dataset is also evaluated with methods as baseline methods, including 2D CNN and 3D CNN. Our proposed method, which is a combination of Inception V3 and BILSTM, outperforms all baseline methods with 81 % accuracy. The developed dataset will be publicly available to foster in various areas related to human activity recognition. These areas encompass the development of systems for detecting behaviors such as robbery, identifying human movements, enhancing safety measures, and detecting instances of theft.

show abstract

Section: Proposed Methodsmentioning

confidence: 99%

“…Prior research has primarily concentrated on identifying shoplifting by analyzing human behavior using methods such as 2D CNN [8], 3D CNN [9][10][11][12], RNN [3], and LSTM [13]. Previous studies on shoplifting detection have some limitations, such as training on the small size of the dataset and limited cases of shoplifting.…”

Section: Introductionmentioning

confidence: 99%

Shoplifting Detection Using Hybrid Neural Network CNN-BiLSMT and Development of Benchmark Dataset

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Irfanullah et al (2022) utilized MobileNet model for intelligent reconnaissance video examination in real time to cater a practical solution for the difficult human undertaking. A recent work (Amin et al, 2023), attempted to define Quantum CNNs to address the sufficiently large data or model dimension limitation. It classified the number of violent robberies such as armed thefts containing handguns or knives, and robberies displaying varying levels of viciousness.…”

Section: Related Workmentioning

confidence: 99%

Fight detection with spatial and channel wise attention‐based ConvLSTM model

Chaturvedi,

Dhiman,

Vishwakarma

2023

Expert Systems

View full text Add to dashboard Cite

An automated detection of aggressive and violent behaviour in videos has immense potential. It enables efficient online content filtering by restricting access to extreme content and also, when integrated with security systems, helps to monitor violence in surveillance videos. In this work, a convolutional neural network is combined with the proposed Spatial and Channel wise Attention‐based ConvLSTM encoder (SCan‐ConvLSTM). The proposed architecture performs an efficient spatiotemporal fusion of the features extracted from the video sequences containing fight scenes. In order to focus selectively on regions of utmost importance, this blended attention mechanism adjusts the weights of outputs in different locations and across different channels. This recurrent attention mechanism enhances the sequential refinement of activation maps and boosts the model performance. Finally, the experimental results have been presented that show the proposed architecture achieves superior results on the benchmark datasets (RWF‐2000, Violent‐flow, Hockey‐fights, and Movies).

show abstract

“…In 2023, Amin, J., et al, [19] design an J. DCNN to analyze the abnormal behavior of anomaly detection using real-time video segment. The suggested models are built from the ground up for the purpose of detecting anomalies in the most di cult publicly accessible video surveillance datasets, including UNI-Crime and UCF Crime.…”

Section: Literature Surveymentioning

confidence: 99%

YOLO ROBBERY: Video Surveillance based Robbery Prediction and Alert System Using YOLO V7

Chidambaram,

Chandrasekaran

2023

Preprint

View full text Add to dashboard Cite

Detection of anonymous behavior is a method of detecting the behavior of people who are insignificant. By using video surveillance and anomaly detection, it is possible to automatically see when something that does not fit the usual pattern is captured by the camera. Although it is a challenging task, it is crucial to automate, improve, and lower expenses in order to detect crimes and other calamities. In this paper, a novel YOLO-Robbery network has been introduced for enhance the security by identifying the threat activities in the supermarket and send the alert message to the shop owner automatically. Initially, the surveillance camera's real-time footage is collected and transformed into image frames for subsequent processing. These frames are pre-processed using multi-scale retinex to remove distortions and augmented to increase the data frames. This work utilizes the YOLO V7 network to extract features from surveillance camera images to quite effective at recognizing and classifying threats at supermarket. Finally, Greedy snake optimization is used to fine-tune the hyperparameters of YOLO V7 network it is trained using DCSASS dataset for efficient image recognition and the alert message is sent to the shop owner automatically. The proposed method has been simulated using MATLAB. The experimental result shows that the YOLO-Robbery method performance was evaluated using the DCSASS dataset in terms of accuracy, precision, recall, and specificity. The proposed YOLO-Robbery achieves the overall accuracy of 99.15%. The proposed YOLO-Robbery increases the overall accuracy range by 13.15%, 2.15%, and 6.24 better than CLSTM-NN, J. DCNN, and ANFIS respectively.

show abstract

Detection of anomaly in surveillance videos using quantum convolutional neural networks

Cited by 23 publications

References 40 publications

Shoplifting Detection Using Hybrid Neural Network CNN-BiLSMT and Development of Benchmark Dataset

Shoplifting Detection Using Hybrid Neural Network CNN-BiLSMT and Development of Benchmark Dataset

Fight detection with spatial and channel wise attention‐based ConvLSTM model

YOLO ROBBERY: Video Surveillance based Robbery Prediction and Alert System Using YOLO V7

Contact Info

Product

Resources

About