Detection of Fake and Real Violence Using Hierarchical CNN Model DOI Open Access

Lucky Rajpoot,

Rosy Madaan

International Journal of Electronics and Communication Engineering, Journal Year: 2024, Volume and Issue: 11(6), P. 114 - 121

Published: June 30, 2024

This investigation delves into the intersection of deep learning and image processing for early detection classification violence, with a primary focus on differentiating between movie fights (staged or fake) true violence. Leveraging "Violence Non-violence Images Dataset," along collected fight images dataset, proposed methodology involves Training Model3 (Hierarchal combination Model1 Model2). The hierarchy enhances performance significantly improves specificity scores, even in dataset biased toward nonviolence cases. model achieves an impressive accuracy 98.33%, showcasing its potential crime detection.

Language: Английский

A Review of Recent Hardware and Software Advances in GPU-Accelerated Edge-Computing Single-Board Computers (SBCs) for Computer Vision DOI Creative Commons
Umair Iqbal, Tim Davies, Pascal Perez

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(15), P. 4830 - 4830

Published: July 25, 2024

Computer Vision (CV) has become increasingly important for Single-Board Computers (SBCs) due to their widespread deployment in addressing real-world problems. Specifically, the context of smart cities, there is an emerging trend developing end-to-end video analytics solutions designed address urban challenges such as traffic management, disaster response, and waste management. However, deploying CV on SBCs presents several pressing (e.g., limited computation power, inefficient energy real-time processing needs) hindering use at scale. Graphical Processing Units (GPUs) software-level developments have emerged recently these enable elevated performance SBCs; however, it still active area research. There a gap literature comprehensive review recent rapidly evolving advancements both software hardware fronts. The presented provides detailed overview existing GPU-accelerated edge-computing including algorithm optimization techniques, packages, development frameworks, specific packages. This subjective comparative analysis based critical factors help applied Artificial Intelligence (AI) researchers demonstrating state art selecting best suited combinations use-case. At end, paper also discusses potential limitations highlights future research directions this domain.

Language: Английский

Citations

7
Boxing behavior recognition based on artificial intelligence convolutional neural network with sports psychology assistant DOI Creative Commons

Yuanhui Kong,

Zhiyuan Duan

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: April 1, 2024

Abstract The purpose of this study is to deeply understand the psychological state boxers before competition, and explore an efficient boxing action classification recognition model supported by artificial intelligence (AI) technology through these characteristics. Firstly, systematically measures key dimensions boxers, such as anxiety level, self-confidence, team identity, opponent attitude, scale survey obtain detailed data. Then, based on data, innovatively constructs a BERT fusion 3D-ResNet, which not only comprehensively considers information, but also carefully characteristics improve accuracy actions. performance evaluation shows that proposed in significantly superior traditional terms loss value, F1 reaches 96.86%. Therefore, comprehensive application psychology deep learning, successfully can fully provides strong support for training boxers.

Language: Английский

Citations

1
Perceptions of Women’s Safety in Transient Environments and the Potential Role of AI in Enhancing Safety: An Inclusive Mobility Study in India DOI Open Access
Guilhermina Torrão, Amal Htait, Shun Ha Sylvia Wong

et al.

Sustainability, Journal Year: 2024, Volume and Issue: 16(19), P. 8631 - 8631

Published: Oct. 5, 2024

Travel safety for women is a concern, particularly in India, where gender-based violence and harassment are significant issues. This study examines how the perception of influences women’s travel behaviour assesses potential technology solutions to ensure their safety. Additionally, it explores AI machine learning techniques may be leveraged enhance A comprehensive mobility survey was designed uncover complex relationship between behaviour, reasons mode choice, built environment, feelings, future mobility, technological solutions. The responses revealed that security most critical factors affecting choices, with 54% 41%, respectively. Moreover, over 80% indicated willingness change after experiencing fear, anxiety, or danger during everyday journeys. Participants were 24% less willing use ride-sharing services than ride-hailing services, which could affect transition towards more sustainable transportation options. Furthermore, AI-based sentiment analysis 46% respondents exhibited signs ‘anger’ regarding what help feel safer transient environments. practical implications this study’s findings discussed, highlighting optimise transport planning.

Language: Английский

Citations

1
Detection of Fake and Real Violence Using Hierarchical CNN Model DOI Open Access

Lucky Rajpoot,

Rosy Madaan

International Journal of Electronics and Communication Engineering, Journal Year: 2024, Volume and Issue: 11(6), P. 114 - 121

Published: June 30, 2024

This investigation delves into the intersection of deep learning and image processing for early detection classification violence, with a primary focus on differentiating between movie fights (staged or fake) true violence. Leveraging "Violence Non-violence Images Dataset," along collected fight images dataset, proposed methodology involves Training Model3 (Hierarchal combination Model1 Model2). The hierarchy enhances performance significantly improves specificity scores, even in dataset biased toward nonviolence cases. model achieves an impressive accuracy 98.33%, showcasing its potential crime detection.

Language: Английский

Citations

0