Article Per Year (5 Year)
p-Index From 2021 - 2026
0.702
P-Index
This Author published in this journals
All Journal Formosa Journal of Science and Technology (FJST) Indonesian Journal of Advanced Research (IJAR)
Ahmed, Nisher
Unknown Affiliation
Humanities Civil Engineering, Building, Construction & Architecture Computer Science & IT Decision Sciences, Operations Research & Management Education Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Social Sciences

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
Evaluating the Performance of Vision Transformers and Convolutional Neural Networks for Hostile Image Detection Hossain, Zakir; Hossain, Md Emran; Ahmed, Nisher; Kabir, Md Farhad; Hossain, Iffat Sania
Indonesian Journal of Advanced Research Vol. 4 No. 1 (2025): January 2025
Publisher : PT FORMOSA CENDEKIA GLOBAL

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55927/ijar.v4i1.13681

Abstract

Detecting malicious or adversarial images, for example in security and surveillance systems, is an important problem in computer vision. These results highlight the effectiveness of ViTs when compared to CNNs when confronting hostile images. However, CNNs have stiff competition from ViTs and have been the go-to architecture for image classification and object detection for many years, due to the existence of spatial hierarchies in images. Using benchmark datasets containing a combination of adversarial and clean images, this study compares the ability of both models to (i) detect hostile images, (ii) generalize to unseen dataset, and (iii) the overall computational efficiency of both models. While ViTs can be even more computationally expensive than incurred with task3 input, we demonstrate that, in fact, our architecture generalizes truncation -- both in power and action -- exceptionally well and can simply outperform performance-per-dollar in more robust pattern recognition tasks, especially under adversarial perturbations. In contrast, CNNs are faster to inference and less likely to overfit on small data. This finding informed decisions showing trade-offs between the two architectures, including a potential path for hybrid approaches and future enhancements in the adversarial defense against hostile image detection.
Understanding the Capabilities and Implications of Agentic AI in Surveillance Systems Ahmed, Nisher; Hossain, Md Emran; Hossain, Zakir; Kabir, Md Farhad; Hossain, Iffat Sania
Indonesian Journal of Advanced Research Vol. 4 No. 1 (2025): January 2025
Publisher : PT FORMOSA CENDEKIA GLOBAL

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55927/ijar.v4i1.13682

Abstract

Agentic AI can also mean the speeding up of new levels of surveillance systems the size of which has never been encountered before which grants automation decision making and realtime responses. Examples of stateful models of an agent are agentic AI, where an agentic AI isn't just a static function, but has capabilities to reason and learn with reference to the environment and goals. The paper explores the possible implications of embedding Agentic AI in surveillance systems, demonstrating how it could revolutionize monitoring, identification of threats, and response systems. Agentic AI: how this would take surveillance to a whole new levelThis is better than never breaking your objectivity at all, but leaves you to micromanage thousands of processes each requiring situational analysis and prediction in real time to create data packets, imploding the challenges of complex environments into just a number/sensible statistic. Yet, the application of Agentic AI for surveillance raises several ethical, legal, and social issues, such as about privacy, accountability, and the risk of misuse. It also outlines challenges to transparency, neutrality and oversight of Agentic AI systems (in their design and deployment), and emphasises the need for powerful regulatory frameworks able to confront risk. Moving forward the ability of Agentic AI to enhance surveillance technologies is significant, but the implementation of such a technology must be implemented correctly while remaining at or below the existing moral paradigms of our society and protecting the fundamental human rights of the individual.
AIPowered Trust and Security: Enhancing ECommerce with Blockchain and Machine Learning Ahmed, Nisher; Hossain, Md Emran; Hossain, Zakir; Kabir, Md Farhad; Hossain, Iffat Sania
Formosa Journal of Science and Technology Vol. 4 No. 1 (2025): January 2025
Publisher : PT FORMOSA CENDEKIA GLOBAL

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55927/fjst.v4i1.13680

Abstract

The ecommerce wave we saw over the years, not just added more opportunities, but added more challenges, especially in the area of trust and security. Fraud, data theft and a lack of transparency remain causes for concern for both businesses and consumers. This paper explores new possibilities of using blockchain and machine learning in designing a robust Artificial Intelligence (AI)based secure ecommerce ecosystem. The immutability of data, transparency, and decentralized control of blockchain act against counterfeit products, payment fraud, and integrity of supply chains. In parallel, machine learning algorithms provides realtime threat detection, predictive analytics, and personalized security measures to detect and counteract threats preemptively. The solution that is proposed leverages the benefits of these technologies to enhance trust among all parties involved, improve operational efficiency, and offer a more secure and trustworthy ecommerce environment. We address the underlying tech stack, realworld application, and next steps in leveraging the convergence of blockchain and ML technologies to transform ecommerce security for a clean and secure digital market.
Classifying and Triggering Events in Close-Contact Bullying Scenarios: A Study on the Effectiveness of Remote Deep Neural Networks Hossain, Zakir; Ahmed, Nisher; Hossain, Md Emran; Kabir, Md Farhad; Hossain, Iffat Sania
Indonesian Journal of Advanced Research Vol. 4 No. 2 (2025): February 2025
Publisher : PT FORMOSA CENDEKIA GLOBAL

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55927/ijar.v4i2.13782

Abstract

Cyber bullying in close-contact environments, especially schools, is another huge social problem that has a deep impact on victims psychologically and mentally. Recent developments in deep learning and remote monitoring technologies offer the potential to improve real-time detection and intervention strategies. In this work, we explore the effectiveness of remote deep neural networks (DNNs) for classifying and identifying trigger events in close-contact bullying events. Using spatial-temporal video data and multimodal sensor inputs, we present a hierarchical DNN framework that fuses real-time audio, video, and physiological signals to accurately identify bullying events.In our proposed system, we use transfer learning using pre-trained vision transformers (ViTs) and convolutional neural networks (CNNs) to extract the key visual features, while Bidirectional Long Short-Term Memory (Bi-LSTM) networks analyze the speech and contextual cues. We develop a hierarchical user model to classify events into verbal,physical, and psychological bullying. The deployed system on edge devices, with cloud-assisted inference, yields real-time low-latency detection.
Co-Authors Hossain, Iffat Sania Hossain, Md Emran Hossain, Zakir Kabir, Md Farhad