Satellite Image Based Animal Identification System Using Deep Learning Assisted Remote Sensing Strategy A.S. Gowri, Immanuel Yovan, S.D. Sundarsingh Jebaseelan, S.D. Anitha Selvasofia, N. Nandhana Proceedings 3rd International Conference on Advances in Computing Communication and Applied Informatics Accai 2024, 2024 Over the past ten years, tiny Unmanned Aerial Vehicles (UAVs) have exploded in popularity for a variety of aerial monitoring applications, including livestock counting, wildlife tracking in their natural environments, and agricultural region monitoring. When used in conjunction with deep learning, they make it possible to analyze and recognize images automatically. Recently, species population detection and monitoring in remotely sensed data has been made possible with the use of deep learning, an efficient machine learning technology. Animal recognition in satellite and aerial photos is one area where deep learning approaches are finding practical use right now, and this paper intends to give an experimental review of those areas. To simplify the process of animal identification from satellite images, this study presented a new deep learning method called the Learning based Animal Sensing and Classification Model (LASCM). To test how well the method worked, it was cross-validated with the traditional Random Forest (RF) model. The primary obstacles to implementing these deep learning techniques are unbalanced datasets, tiny samples, tiny objects, picture annotation techniques, picture backgrounds, animal counting, evaluation of model performance, and uncertainty calculation. Barely and self-supervised techniques for learning, optimizing either favorable or adverse instances, improving network architecture, and sample annotation methods were all considered as potential answers. The following areas are projected to get increased attention in the next years: video-based detection; detection based on extremely high-resolution satellite images; identification of several species; novel methods for annotation; and the creation of specialized network frameworks and big foundational modelling. The proposed methodology is designed to sort out all these problems and provide an efficient animal identification scheme based on deep learning model from the satellite images.
Efficient Diabetes Detection using Hybrid Machine Learning Model A S Gowri, P Jose, Karthik S M, S Berlin Shaheema, K M Karuppasway, R. Balamurugan International Conference on Distributed Systems Computer Networks and Cybersecurity Icdscnc 2024, 2024 Diabetes is a metabolic disease that affects a large number of the global population and is incurable. The primary causes of death symptoms are kidney failure, heart attacks, strokes, and blindness. In this paper Principal component analysis used for pre-processing the data to extract features and The Efficient Hybrid Machine Learning Model Convolution Neural Network-Support Vector Machine is used to categorize the Pima Indian dataset. CNN and non-linear SVM used in a hybrid model to classify the diabetes data from the Pima Indians. To identify positive classes of pima Indians diabetes, the HCNN-SVM extract the features from the original data expression and learns diabetes data through a novel convolution network environment to the SVM classifier with RBF function and achieve the accuracy as 98,04%.
ACT on Monte Carlo FogRA for Time-Critical Applications of IoT A. S. Gowri, P. Shanthi Bala, Zion Ramdinthara, T. Siva Kumar International Journal of Advanced Computer Science and Applications, 2022 —The need for instantaneous processing for Internet of Things (IoT) has led to the notion of fog computing where computation is performed at the proximity of the data source. Though fog computing reduces the latency and bandwidth bottlenecks, the scarcity of fog nodes hampers its efficiency. Also, due to the heterogeneity and stochastic behavior of IoT, traditional resource allocation technique does not suffice the time-sensitiveness of the applications. Therefore, adopting Artificial Intelligence (AI) based Reinforcement Learning approach that has the ability to self-learn and adapt to the dynamic environment is sought. The purpose of the work is to propose an Auto Centric Threshold (ACT) enabled Monte Carlo FogRA system that maximizes the utilization of Fog’s limited resources with minimum termination time for time-critical IoT requests. FogRA is devised as a Reinforcement Learning (RL) problem, that obtains optimal solutions through continuous interaction with the uncertain environment. Experimental results show that the optimal value achieved by the proposed system is increased by 41% more than the baseline adaptive RA model. The efficiency of FogRA is evaluated under different performance metrics.
Fog Computing Perspective: Technical Trends, Security Practices, and Recommendations C. Kaviyazhiny, P. Shanthi Bala, A.S. Gowri Smart Cyber Ecosystem for Sustainable Development, 2021 As cloud computing is not viable for many Internet of Things (IoT) applications, fog computing is an emerging technology that inherits cloud computing platform and addresses the need for IoT and industry IoT. It reduces the time taken for communication between IoT and the cloud and substantially reduces the bandwidth that affects the IoT performance. The incorporation of IoT applications in fog paves way for a high chance of vulnerabilities in fog. So, the fog layer requires more security to protect the data both in transit and rest. Hence, the security issues of fog computing and the existing solutions are revealed in this chapter. It has been concluded that fog computing provides better performance than its counterparts like Edge Computing, Cloudlet, and Micro-Datacenter. This chapter investigates major security issues in fog computing and provides possible solutions and security recommendations to meet IoT security goals.
Comprehensive Analysis of Resource Allocation and Service Placement in Fog and Cloud Computing A. S. Gowri, P.Shanthi Bala, Immanuel Zion International Journal of Advanced Computer Science and Applications, 2021 The voluminous data produced and consumed by digitalization, need resources that offer compute, storage, and communication facility. To withstand such demands, Cloud and Fog computing architectures are the viable solutions, due to their utility kind and accessibility nature. The success of any computing architecture depends on how efficiently its resources are allocated to the service requests. Among the existing survey articles on Cloud and Fog, issues like scalability and time-critical requirements of the Internet of Things (IoT) are rarely focused on. The proliferation of IoT leads to energy crises too. The proposed survey is aimed to build a Resource Allocation and Service Placement (RASP) strategy that addresses these issues. The survey recommends techniques like Reinforcement Learning (RL) and Energy Efficient Computing (EEC) in Fog and Cloud to escalate the efficacy of RASP. While RL meets the time-critical requirements of IoT with high scalability, EEC empowers RASP by saving cost and energy. As most of the early works are carried out using reactive policy, it paves the way to build RASP solutions using alternate policies. The findings of the survey help the researchers, to focus their attention on the research gaps and devise a robust RASP strategy in Fog and Cloud environment.
Fog resource allocation through machine learning algorithm Gowri A. S., Shanthi Bala P. Architecture and Security Issues in Fog Computing Applications, 2019 Internet of things (IoT) prevails in almost all the equipment of our daily lives including healthcare units, industrial productions, vehicle, banking or insurance. The unconnected dumb objects have started communicating with each other, thus generating a voluminous amount of data at a greater velocity that are handled by cloud. The requirements of IoT applications like heterogeneity, mobility support, and low latency form a big challenge to the cloud ecosystem. Hence, a decentralized and low latency-oriented computing paradigm like fog computing along with cloud provide better solution. The service quality of any computing model depends on resource management. The resources need to be agile by nature, which clearly demarks virtual container as the best choice. This chapter presents the federation of Fog-Cloud and the way it relates to the IoT requirements. Further, the chapter deals with autonomic resource management with reinforcement learning (RL), which will forward the fog computing paradigm to the future generation expectations.
An agent based resource provision for IoT through machine learning in Fog computing A.S. Gowri, P Shanth i Bala 2019 IEEE International Conference on System Computation Automation and Networking Icscan 2019, 2019 Internet of things (IoT) is a large scale distributed system that is growing at rapid fire pace. It is a technological revolution that makes devices smarter, computations intelligent and communications more informative. While IoT still presuming different definitions, its application had set a broader footprint in almost all walks of our daily life. The voluminous amount of data generated by the millions of IoT devices imposes a higher demand on the computation and storage resources. The compute resources to serve the IoT applications need to be chosen depending upon the heterogeneity of the IoT devices. The various constraints of IoT make resource provisioning in the cloud a non trivial task. Fog computing is the apt platform to deal with such constraints of IoT. The IoT challenges like heterogeneity, scalability and low latency can be addressed by fog computing by adapting intelligence features of machine learning in its resource management techniques. In this paper, we propose a resource management technique for fog computing in which an agent adapts centralized learning and distributed scheduling of IoT tasks. This paper considers the micro data center as the resource that have the hardware and software capability to use TCP/IP protocol suite in fog computing paradigm.
