A5050615367- Chaos-based Image/Signal Encryption
- Advanced Steganography and Watermarking Techniques
- Face and Expression Recognition
- Digital Media Forensic Detection
- Cryptographic Implementations and Security
- Stock Market Forecasting Methods
- Artificial Intelligence in Healthcare
- Remote-Sensing Image Classification
- Face recognition and analysis
- Biometric Identification and Security
- Cellular Automata and Applications
- Image and Video Stabilization
- Image Retrieval and Classification Techniques
- Generative Adversarial Networks and Image Synthesis
- Machine Learning in Healthcare
- Advanced Data Compression Techniques
- Forensic Fingerprint Detection Methods
- Water Quality Monitoring Technologies
- Nausea and vomiting management
- IoT and Edge/Fog Computing
- Text and Document Classification Technologies
- Healthcare Policy and Management
- Healthcare Operations and Scheduling Optimization
- Advanced Queuing Theory Analysis
- Forensic and Genetic Research
SASTRA University
2018-2025
SRM University
2010-2020
SRM Institute of Science and Technology
2009-2018
Vellore Institute of Technology University
2013
Singapore-MIT Alliance for Research and Technology
2013
Cloud services offer doctors and data scientists access to medical from multiple locations using different devices (laptops, desktops, tablets, smartphones, etc.). Therefore, cyber threats at rest, in transit when used by applications need be pinpointed prevented preemptively through a host of proven cryptographical solutions. The presented work integrates adaptive key generation, neural-based confusion non-XOR, namely DNA diffusion, which offers more extensive unique key, unpredictable...
With the extensive growth in usage of cars, newly produced cars are unable to reach customers for various reasons like high prices, less availability, financial incapability, and so on. Hence used car market is escalated across globe but India, a very nascent stage mostly dominated by unorganized sector. This gives chance fraud while buying car. precision model required which will estimate price an with none bias towards customer or merchandiser.In this model, A Supervised learning-based...
In the healthcare sector, e-diagnosis through medical images is essential in a multi-speciality hospital; securing becomes crucial for preserving an individual's privacy e-healthcare applications. So, this paper has proposed novel encryption scheme implemented on reconfigurable hardware. Realising image schemes FPGA hardware platforms offers substantial advantages over software implementations. The image-specific key and Hopfield Neural Network (HNN) carry out diffusion process using...
Non-Insulin Dependent Diabetes Mellitus or Type2 is one of the critical diseases and many people are suffering from it. Every year, approximately 2 to 5 million losing their lives as Diabetics. If predicted earlier, it can be controlled also, deadly risks such diabetes cardiac stroke, nephropathy other disorders associated with prevented. Therefore, early prediction helps in maintaining good health. With recent development machine learning (ML), being applied various aspects medical The Pima...
Traditional agriculture has been the global foundation for development centuries. However, to meet this demand and exponential growth of population, farmers need water irrigate their property. Farmers require a fix that modifies business practices because scarcity using resource. To keep up with satisfy Demand, Agriculture 4.0 become reality thanks new technologies. Automated Irrigation: Smart irrigation systems can automatically adjust schedules based on real-time data, such as weather...
Machine Learning is considered as a subfield of Artificial Intelligence and it concerned with the development techniques methods which enable computer to learn.In classification problems generalization control obtained by maximizing margin, corresponds minimization weight vector.The vector can be used in regression problems, loss function.The problem for linearly separable data introduces concept margin essence SVM -margin maximization.In this paper gives soft idea slack variables trade-off...
Background: Recent advances in medical associated technologies have drastically increased the amount of electronic records collected, stored and transferred through network. Considering significance level sensitivity collected data, security transmitted data has become a very vital challenging task for researchers. The protection these images with embedded is usually guaranteed using encryption or hiding techniques. Conventional techniques that employ are often insecure also time-consuming...
Securities trade data is a high dimensional time course of action cash related that positions exceptional computational challenges. Stock variable with respect to time, suspecting the future example expenses trying task. The segments effect consistency stock can't be judged as same variables may affect estimation always. We propose burrowing approach for desire advancement securities trade. It consolidates using innate pre taking care and cross breed packing strategy Hierarchical gathering...
Background and Objective: Iris recognition is one of the popular winning biometric frameworks, giving promising outcomes in identity authentication access control systems.In this study, an efficient, fast robust segmentation methodology suitable for non-cooperative noisy iris images proposed.Materials Methods: This proposed considers both shape spatial feature properties taken from visible spectrum near infrared spectrum.Circular hough transform applied to input image outer boundary...
Information security is becoming more crucial to data transmission and storage in this era of digital communication. In numerous processes, images are frequently used. As a result, protecting the image from misuse equally crucial. Image encryption solution protect visual hackers. The chaotic algorithm unique choice, but software algorithms can be hacked anywhere. paper proposes scheme along with standalone device using FPGA. Even if compromised, FPGA keys secure. Because developed on core...
The common vector (CV) method is a linear subspace classifier for datasets, such as those arising in image and word recognition. In this approach, class modeled from the features of all samples corresponding class. Since are separate each feature domain, there overlapping between these subspaces loss information which turn reduces recognition performance. CV followed criterion considers only scatter matrices. Thus neglecting influence neighboring classes also Generally fails to extract...
Forecasting stock price is a important challenging task in the real world because more and money involved they are affected by many social, economic, political psychological factors. Numerous machine learning procedures have been used to fore see developments cost. Machine classifiers include expanding past encounters into future actives. The proposed framework presents another hereditary calculation for forecast of monetary execution with information sets from known source. objective...