Journal of Sensor and Cloud Computing (e-ISSN: 3048-9199)

IoT-Enabled AI System for Real-Time Fake Currency Detection and Denomination Identification

N. B. Mahesh Kumar — 2026-06-01

The proposed system proposes a low-cost, sensor-driven embedded system using an Arduino microcontroller to address the rising threat of counterfeit currency. Unlike resource-heavy machine learning approaches, this architecture utilizes light sensors—such as a Light Dependent Resistor (LDR) or BH1750—to analyze the optical characteristics of banknotes. The system works by illuminating a note with a controlled light source and measuring the reflected or transmitted intensity. Because genuine currency possesses unique materials and security features, it produces consistent optical patterns that differ significantly from counterfeits. These readings are processed via a threshold-based algorithm to verify authenticity instantly. Beyond detection, the system features automatic denomination recognition and counting. Since each note value corresponds to a specific light intensity range, the device identifies the denomination and updates a cumulative total. Results, including authenticity status and total value, are displayed on an LCD module. Designed for portability and energy efficiency, this solution is ideal for small businesses, retail shops, and banks. Experimental evaluations show high accuracy and fast response times, effectively reducing human error in financial transactions. While currently optimized for visible light, future iterations may integrate UV or IR sensing to further enhance robustness against sophisticated forgeries. This approach offers a simple, accessible, and effective alternative to complex, expensive detection hardware.

VisionGuard: Physics-Driven Vehicle Collision Detection and Alerting System using Deep Learning

Rajshekar Gaithond — 2026-06-24

Traffic-related accidents represent a major source of global mortality, often compounded by delayed emergency assistance due to manual reporting constraints. This paper introduces VisionGuard, an automated, web-based collision detection and alerting platform that integrates deep learning-based object detection, multi-object tracking, and physics-driven kinematic reasoning. Operating on live or pre-recorded traffic video streams, the system utilizes the YOLO v11 nano model to localize vehicles across four categories (cars, motorcycles, buses, and trucks) and maps them to a ByteTrack tracker to maintain unique identities across successive frames. A custom physics engine records center-point trajectories in a sliding window buffer, continuously evaluating vehicle pairs for spatial proximity (Intersection over Union and center-point Euclidean distance) and kinematic anomalies. A collision is logged when close proximity coincides with a kinetic shock (sudden velocity drop exceeding 70 %) or an abrupt angular direction change. Confirmed incidents trigger visual overlays, capture localized snapshots, and generate structured, evidence-backed PDF reports. Experimental results on real-world CCTV footage show that VisionGuard achieves a collision detection accuracy of 95.05 %, a specificity of 97.36 %, and a processing speed of 45.5 FPS under GPU acceleration, providing an explainable and reliable automated traffic monitoring layer.

AI-Driven Smart Waste Classification Model for Sustainable Practices in the RMG Industry

Ashikur Rahman Chowdhury — 2026-07-14

Waste management has always been a very complex issue in the ready-made garment (RMG) sector. Each year, almost 92 million tons of textile waste are generated globally. As a result, an effective and efficient waste management system is genuinely required in order to ensure long-term sustainability. This study presents an efficient waste sorting model for the RMG industry. This framework used image-based classification models to uniquely identify different types of common waste, including fabric scraps, buttons, zippers, threads and labels. Various machine learning and deep learning models, including Convolutional Neural Networks (CNN), Transfer Learning models (InceptionV3, ResNet50V2, VGG16), Decision Trees, Support Vector Machines (SVM), Random Forest, and k-Nearest Neighbours (KNN), are being used in this study. The performances of these algorithms were also compared in this research. The datasets were collected from three different major RMG manufacturers in Bangladesh which were further enriched with data augmentation techniques. The categorical cross-entropy loss, the Adam optimizer, and dropout regularization were incorporated in the model training part to enhance generalization. After the evaluation, the final result shows that InceptionV3 achieved the highest classification accuracy at 90.00%, VGG16 at 88.81%, and ResNet50V2 at 87.99%, while the basic CNN model achieved 57.30% accuracy. This result highlights the fact that AI integration has become a dire need in the RMG industry to ensure long-term sustainability.

An Uncertainty-aware Intelligent Digital Twin Framework for Robust Structural Health Monitoring

Shilpi Saxena — 2026-08-05

Civil infrastructure is continuously exposed to operational loads, environmental effects, and material degradation, making accurate Structural Health Monitoring (SHM) essential for ensuring safety and reliability. Conventional SHM methods rely on periodic manual inspections or purely data-driven models, which often lack robustness and fail to generalize under changing operating conditions. The objective of this study is to develop an intelligent digital twin framework that improves damage prediction accuracy while providing reliable uncertainty estimates for informed maintenance decision-making. To achieve this, an uncertainty-aware physics-guided digital twin framework, named UPG-DT, is proposed. The framework integrates physics-guided representation learning, spatio-temporal graph neural networks, and Bayesian uncertainty estimation to capture structural behavior, model spatial correlations among multi-sensor data, and quantify prediction uncertainty. Experimental evaluation on real-world benchmark datasets demonstrates that UPG-DT outperforms existing approaches in predictive accuracy, reduces false-positive detections, and provides better uncertainty calibration. In addition, the proposed framework combines damage predictions and uncertainty estimates into a hybrid risk score that supports maintenance prioritization. These results indicate that UPG-DT offers a reliable and practical solution for intelligent SHM and digital twin-based infrastructure management.

Robotics-driven HealthCare and Assistive Technologies: A Comprehensive Review

Mithlesh Arya — 2026-08-05

This article provides a comprehensive review of the healthcare robotics paradigm, which has evolved from isolated mechanical systems to a pervasive, interconnected cyber-physical ecosystem that underpins digital transformation in medicine. This ecosystem, often termed the Internet of Medical Things (IoMT), encompasses billions of heterogeneous devices, from surgical robotic assistants to nanoscale sensors. As of 2024-2025, the medical robotics market is valued at over USD 12.98 billion, with the assistive robotics sector valued at an additional USD 10.46 billion. This growth, driven by compound annual growth rates (CAGRs) exceeding 16.1% and 20.9%, respectively, is predicated on the integration of intelligent algorithms, advanced connectivity, and novel economic models. Built on the principle of closing the cyber-physical control loop, healthcare robotics translates diagnostic data and clinical commands into precise physical actuation. Despite rapid adoption, significant challenges persist in system interoperability, data security, and the critical need for new workforce competencies. This article analyses the core application domains, from AI-assisted surgery to socially assistive care. It examines the foundational enabling technologies—including 5G for telesurgery, edge computing for real-time inference, and blockchain for data integrity. Furthermore, it analyses the economic shift to Robotics-as-a-Service (RaaS) models that are mitigating long-standing cost barriers. Finally, the article explores future trajectories involving 6G-enabled haptics, Quantum-AI for diagnostics, and neuromorphic computing for adaptive assistive devices.