Safety Threat Detection in Construction Zones using Dense YOLOv8 with White Shark Nested Attention Network Rajeeth T J, Chitaranjan Dalai, Vanitha G Naik, Vipul Vekariya, Kannadasan B, Harshal Patil 8th International Conference on I Smac Iot in Social Mobile Analytics and Cloud I Smac 2024 Proceedings, 2024 The construction industry faces numerous hazards and risks, many of which remain unmonitored, leading to accidents and injuries. To address these challenges and improve safety in construction areas, this paper proposes a novel approach based on the Dense YOLOv8 White Shark Nested Attention Network (DYWS-NAN) for safety threat detection. Utilizing the Construction Site Safety Image Dataset, which contains a wide range of images depicting safety compliance and violations, the proposed method ensures precise identification of safety threats. The dataset undergoes rigorous pre-processing using the Grid-Constrained Data Cleansing Method to remove noise and enhance image quality, enabling the model to focus on critical safety features. Following pre-processing, feature extraction and classification are performed using the DYWS-NAN architecture, which combines YOLOv8's real-time object detection capabilities with the White Shark Nested Attention mechanism. This integration enhances both object feature extraction and classification performance, significantly improving safety threat detection accuracy. Simulations conducted in the Python environment demonstrate an accuracy level of 99.4%, highlighting the method's effectiveness in reducing false alarms and enabling timely safety interventions in construction zones. This approach enhances safety management and ensures better compliance with safety regulations.
Efficient complexity based adaptive system for cloud resources Sarvesh Kumar, Dyagala Naga Sudha, Anupama Anupama, B. Kannadasan, Ajay Singh Yadav, Dinesh Goyal Journal of Interdisciplinary Mathematics, 2023 Cloud computing is a new IT concept, that is no longer solely applicable to the economic system however as nicely very beneficial in science. The issue of asset distribution and income expansion is likewise similarly significant, particularly about cloud security. This achieves the need of various displaying strategies including however not restricted, to security danger, asset assignment, and income boost models. These offerings are billed on a utilization basis. The cloud services are provided by the CSPs to the end users in an optimized way by using our mathematical proposed algorithm. This proposed algorithm is simulated in the cloud simulator. It prompts financially savvy arrangements by lessening the execution season of enormous application testing. As a piece of framework assets, cloud testing can accomplish its productivity by dealing with the boundaries like organization traffic, Circle Stockpiling, and Smash speed. In this paper, we propose another fluffy numerical model to accomplish a superior degree for the above boundaries. The results of the outcomes informs that the proposed algorithm CROS (Cloud Resource Optimized System) performances are better.
Revolutionizing Waste Management Through Intelligent Garbage Segregation Bin K M Veerabhadraswamy, Karthik Kumar, A Indhuja, Amit Barve, B Kannadasan, Ruhi Bakhare 3rd International Conference on Innovative Mechanisms for Industry Applications Icimia 2023 Proceedings, 2023 Efficient waste management stands as a paramount concern in the current environmental landscape. The proper handling, sorting, transportation, and disposal of waste materials are essential to mitigate environmental hazards. The separation of trash into several categories allows for a more accurate evaluation. Conventional manual waste segregation systems necessitate a significant amount of human labor, time, and money. This study proposes a waste management system integrated using advanced technologies such as the Internet of Things (IoT) and Artificial Intelligence (AI), allowing for easy and cost-effective garbage segregation and tracking. The technique is intended to separate garbage into several categories such as glass, plastic, cardboard, paper, metal, and trash. A sample dataset acquired from Kaggle was rigorously processed to design and evaluate the system, which includes operations like as image resizing and rescaling. The DenseNet, is a Deep Learning (DL) model used for accurate garbage recognition. This unique smart segregation bin makes use of the potential of AI and IoT technology. The model is embedded in a controller, allowing it to properly recognize different waste items and trigger the associated actuator for accuracy sorting. A user-friendly Graphical User Interface (GUI) structure makes the entire process visible to the public. This interface enables users to watch the system’s operation in real-time, providing a clear perspective of garbage identification and sorting. By revolutionizing waste management through intelligent garbage segregation, this system enhances the efficiency of waste sorting, contributing to environmental preservation and sustainable waste management practices.
Advanced AI Techniques for Autonomous Moving Object Detection and Tracking G. Meenakshi Sundaram, Nalluru Mourya Sai Eatesh, Manjiri Ulhas Karande, Warish Patel, B Kannadasan, Harshal Patil 3rd International Conference on Innovative Mechanisms for Industry Applications Icimia 2023 Proceedings, 2023 In the domain of autonomous vehicle systems, the accurate detection and tracking of moving objects, particularly vehicles, are of paramount importance for ensuring safe and efficient operations. Conventional methods based solely on image information face substantial challenges when confronted with complex and dynamic environments. To address these challenges, this research utilizes deep learning (DL) techniques for vehicle detection and tracking. The study draws upon data from the UA-DETRAC dataset, a valuable resource providing real-world scenarios for experimentation. For vehicle detection, DL models, including Faster Region Convolutional Neural Network (Faster R-CNN), You Only Look Once (YOLO), and Single Shot MultiBox Detector (SSD), are employed. These models are assessed using essential metrics like mAP (mean Average Precision), IoU (Intersection over Union), Precision, and Recall. When it comes to object tracking, the study delves into the capabilities of the SORT and DeepSORT algorithms, which represent advanced tracking solutions. Evaluation metrics like MOTA (Multiple Object Tracking Accuracy), IDF1 (Identification F1 score), MT (Mostly Tracked), and ML (Mostly Lost), offer a comprehensive view of tracking performance. The outcomes of this research indicate that YOLO stands out as the top performer in vehicle detection, while DeepSORT excels in the critical task of vehicle tracking. This study’s findings contribute significantly to the advancement of AI techniques, enhancing the capabilities of autonomous systems for object detection and tracking in complex and dynamic real-world environments.
Smart Transport System for Passenger Comfort using IoT Awari Mahesh Babu, T. Thulasimani, D. Sundaranarayana, B. Kannadasan, R. Salini, K. Vanisree Mysurucon 2022 2022 IEEE 2nd Mysore Sub Section International Conference, 2022
I have vast of strong industry experience in Geographic Information Systems (GIS), Remote Sensing, Spatial Data Modelling, and Geospatial Consulting, working across India, USA, and Indonesia. My work involved developing high-quality geospatial datasets, handling large-volume imagery, and delivering accurate spatial solutions for government agencies, utility companies, environmental firms, and global consulting organizations.
