Sustainable Farming Through AI-Enabled Precision Agriculture D. Prabha, R. Siva Subramanian, M.G. Dinesh, P. Girija Artificial Intelligence for Precision Agriculture, 2024 Modern agricultural techniques have seen revolutionary improvements because of Precision Agriculture's quick development, which is driven by the incorporation of artificial intelligence (AI). In-depth information on the symbiotic relationship between AI and Precision Agriculture is presented in this survey study, which emphasizes significant features ranging from fundamental AI ideas to real-world applications. The work discovers a foundational grasp of AI's historical evolution and wide areas of application by beginning with an intelligent introduction to AI. The review then digs into the world of machine learning algorithms, analyzing decision trees, support vector machines, and random forests, and illuminating their vital roles in crop monitoring, disease diagnosis, and yield forecasting. The focus of the investigation moves to deep learning methods, particularly convolutional neural networks and recurrent neural networks, which are used for tasks including yield computation, plant phenotyping, and image analysis. An in-depth analysis of the complex AI model engineering process reveals the iterative path from issue conceptualization to model optimization. The capabilities of TensorFlow, PyTorch, and specialized AutoML tools targeted to the subtleties of agricultural situations are clarified, drawing attention to AI model generation platforms. A study on converting AI models into microservices powered by application programming interfaces (APIs) demonstrates their potential for seamless integration with agricultural systems, raising real-world application. When navigating the deployment phase, the survey takes into account the subtleties of edge computing, cloud deployment, and hybrid architectures to traverse the various techniques for installing AI models in Precision Agriculture. Through insights into data integration and ingestion approaches from various sources, such as remote sensing and IoT devices, the synergy between AI and data is addressed. In addition, the study includes data analytics techniques for extracting forecasts and actionable insights from complicated agricultural data. It highlights the critical role of knowledge visualization tools in bridging the gap between data-driven insights and well-informed decision-making. The study demystifies the fundamental principles of precision and smart agriculture and shows how they have the potential to revolutionize sustainable practices, yield optimization, and resource utilization. The survey article emphasizes the revolutionary potential of AI in Precision Agriculture as a final point, uniting theoretical underpinnings with practical applications. This article provides readers with a comprehensive overview of the dynamic environment where AI and agriculture collide, paving the road for a more sustainable and technologically advanced future by synthesizing a wide range of issues
Quantum processors/networks/sensors Sruthi Anand, Granty Regina J. Elwin, R. Priyadharsini, K. Ranjeetha Priya, D. Prabha Quantum Computing and Artificial Intelligence Training Machine and Deep Learning Algorithms on Quantum Computers, 2023 The concept of quantum computing was introduced to make processing faster so that processes can run in parallel. It harnesses the law of quantum mechanics that aids in solving highly complex problems. The law incorporates the concept of superposition and entanglement, which are the primary properties of quantum. The qubits are those involved in quantum, with the feasibility that a bit can represent any of the two states, unlike classical computers. With the advent of the new technology, it is important to enhance the other units that support in building up the technology of the future. This chapter proposes the different aspects involved in quantum computing, such as quantum sensors, the processing of quantum, the networks established, etc. It also emphasizes on the importance of moving from the classical computers to quantum, with comparisons. The focus is on knowing that there is the possibility of quantum computing in solving problems that cannot be solved in polynomial time. All these aid in uplifting the classic computing power to quantum. As it is understood that this technology is still in the research stage, the chapter provides information on what it is to actually build a quantum system. Having a brief idea on the different units, it also identifies the different applications of quantum with a few case studies.
A Critical Analysis of the Block Chain in Manufacturing System Implementation G Gokilakrishnan, D Prabha, S Srinivasan, S Thirumurugaveerakumar, Vigneshkumar M, Anandakumar H 2023 9th International Conference on Advanced Computing and Communication Systems Icaccs 2023, 2023 As component of the industry 4.0 project, a novel concept of Cyber-Physical Systems (CPS), fog computing, big data analytics, cloud manufacturing, the Internet of Things (IoT), and other technologies have been brought to the manufacturing sector. Some of the projected benefits and possibilities that these innovations might provide include self-prediction, self-maintenance, self-comparison, self-configuration, and self-awareness. In these concepts, the centralised corporate system and third-party trust operations are still in use. However, contemporary manufacturing has a slew of problems, including data security and reliability, adaptability, security, trust, and privacy. This paper provides an analysis of the Blockchain Technology (BT), reviews the application of blockchain in Artificial Intelligence (AI), proposes a model for blockchain-enabled CPS and discusses the potential challenges that face the implementation of blockchain in manufacturing systems.
Automated vehicle number plate recognition system, using convolution long short-term memory technique S. Srinivasan, D. Prabha, N. Mohammed Raffic, K. Ganesh Babu, S. Thirumurugaveerakumar, K. Sangeetha Object Detection with Deep Learning Models Principles and Applications, 2022 With the growing number of cars on the road nowadays, it is required to recognize vehicle number plates during traffic monitoring, tracing stolen vehicles, coordinating parking tolls, highway charge collecting, imposition of red-light negligence, and checks at the border and at customs. Manual detection of number plates from running vehicles in a normal or crowded area is practically not feasible. Existing pattern recognition, template matching and machine learning algorithms lack accuracy due to their complexity for processing in real-time backgrounds. This automatic vehicle number plate recognition system uses deep learning techniques to achieve high accuracy despite realistic challenges like vehicle motion, image size, irregular fonts and different lighting conditions. This technique may be used to locate automobiles that have broken the regulations in crowded places such as malls, universities, hospitals, and other parking lots. Terrorist operations, illegal activities, incorrect number plates, stolen cars, and other illicit acts may all be detected with it. Initially, a grayscale format is obtained from the captured image, and it is preprocessed to remove unwanted noises. Next, the numbers and characters are extracted from the preprocessed image. In this, the relevant pixels are enhanced, the background pixels are weakened, and the super-resolution technique is used to retrieve clear segmented images. Recurrent neural networks (RNN) with convolutional long short-term memory (ConvLSTM) are used to detect objects from continuous images or videos. The ConvLSTM technique with attention layers is implemented to reduce the vanishing gradient problem that traditional RNN faces. To improve the accuracy of the testing, different dropout rates are used. The solution for vehicle number plate identification and detection using a deep learning model is developed and evaluated using Keras. This chapter provides an overview of existing methods, challenges, the role of deep learning, implementation, comparison and discussion of test cases for recognition of license plates.
Ensemble variable selection for Naive Bayes to improve customer behaviour analysis R. Siva Subramanian, D. Prabha Computer Systems Science and Engineering, 2022 Executing customer analysis in a systemic way is one of the possible solutions for each enterprise to understand the behavior of consumer patterns in an efficient and in-depth manner. Further investigation of customer patterns helps the firm to develop efficient decisions and in turn, helps to optimize the enterprise’s business and maximizes consumer satisfaction correspondingly. To conduct an effective assessment about the customers, Naive Bayes(also called Simple Bayes), a machine learning model is utilized. However, the efficacious of the simple Bayes model is utterly relying on the consumer data used, and the existence of uncertain and redundant attributes in the consumer data enables the simple Bayes model to attain the worst prediction in consumer data because of its presumption regarding the attributes applied. However, in practice, the NB premise is not true in consumer data, and the analysis of these redundant attributes enables simple Bayes model to get poor prediction results. In this work, an ensemble attribute selection methodology is performed to overcome the problem with consumer data and to pick a steady uncorrelated attribute set to model with the NB classifier. In ensemble variable selection, two different strategies are applied: one is based upon data perturbation (or homogeneous ensemble, same feature selector is applied to a different subsamples derived from the same learning set) and the other one is based upon function perturbation (or heterogeneous ensemble different feature selector is utilized to the same learning set). Furthermore, the feature set captured from both ensemble strategies is applied to NB individually and the outcome obtained is computed. Finally, the experimental outcomes show that the proposed ensemble strategies perform efficiently in choosing a steady attribute set and increasing NB classification performance efficiently.
A Study on devices for assisting Alzheimer patients D. Surendran, J. Janet, D. Prabha, E. Anisha Proceedings of the International Conference on I Smac Iot in Social Mobile Analytics and Cloud I Smac 2018, 2018
Study of views and perception of engineering graduates in social networking sites to enlarge the quality of education Journal of Advanced Research in Dynamical and Control Systems, 2018
