Volume 2, Issue 4, Publication: August 2020
1. Real-Time Security Based Video Surveillance
Author(s): Dr. C. Pretty Diana Cyril1 and M. Vidhya2
ABSTRACT: We style as well as apply an internet of Things (IoT) referred to as IoT guard, aimed at a smart, source proficient, as well as protection managing program in real time. The device, comprising of edge fog computational levels, will help with criminal avoidance and also foresee criminal occasions inside a Smart home Environment (SHE) and IoT guard will identify as well as verify criminal, utilizing Artificial Intelligence as well as an event driven method for sending the criminal information to the safety providers as well as police devices allowing quick activity. With this research, we put into action an IoT guard lab testbed prototype, conduct evaluations on the efficiency of it’s for real time protection program. Fog was used for the storage platform. The results indicate much better overall performance through the suggested method of terminology of source effectiveness, agility, and then scalability during the standard IoT surveillance devices as well as state-of-the-art (SoA) solutions. We have achieved the communication response and security level on comparing with existing systems.
2. Innovations of Routing and Secure Dispatching of Packets in MANET
Author(s): Rajalakshmi D1
ABSTRACT: There are incredible influence on broadcasting and the ways in which people use it, One kind of technology area, considered here is mobile ad hoc networking (MANET).It is useful in self-constructing and multi-hop mixed network routing facilities and maintenance. In spite of this, modification and intrusion in connection caused by node mobility and wireless channels sharing undermine communication paths, which makes in constructing routing protocol greatest incitement such as broadcast severe link quality variation and delay. Unlike the OLSR and AODV protocol this proposed approach can provide better performance. The adverse effect of link on communication is relieved by using the following techniques such as dedicated routing; prolong updating, tiny scope retransmission and appending of constant nodes. The simulation results shows improvement of packet delivery ratio, reduced delay and node overhead.
3. A Review Paper on an Automatic Localization of Optic Disc and Segmentation Approaches for Glaucoma Diagnosis
Author(s): S. Prabakaran
ABSTRACT: Glaucoma is a leading cause of stable blindness. Retinal fundus image examination is necessary for early detection of Glaucoma. In order to identify the existence of Glaucoma, the ophthalmologists conclude the localization of optic disc in the fundus image. This review paper is a study of current techniques and segmentation approaches to automatically detect the optic disc, which is one of the most main medical features of the fundus image. The idea behind this paper is to provide overview of different segmentation approaches for automatic detection of Glaucoma by comparing and measuring the different parameters of fundus images with normal images and abnormal images.
4. Clustering Based Feature Subset Selection For Semi-Supervised Data
Author(s): M. Kumaresa Babu
ABSTRACT: The curse feature subset selection magnifies when the training data are semi-supervised. The proposed framework uses both Laplacian score and fisher score to rank the features. To enhance the performance of the subsequent machine learning algorithm, the framework incorporates correlation based maximum spanning tree to eliminate redundancy among the selected subset of features performs better than 3 other feature subset selection algorithms in terms of both classification accuracy, evaluated using 4 different datasets.
5. 3-D Models Using Integration of Grid Computing
Author(s): S. Hannah Pauline
ABSTRACT: Grid Computing is an advanced technique that uses the concept of distributed computing. A collection of computers, storage and other devices that are joined together by means of communication like the internet is called the grid. This is also used to solve the problems among themselves and manage the information. Supercomputers and parallelism are the opted by various researchers and organizations to help deliver the outputs of slow and computationally-intensive tasks at a faster pace. One of the typical scenarios where high performance computing resources are used to complete the job fast and efficiently is that of 3D rendering. The resources available are combined and utilized by the technique of parallelism like grid computing which helps in the processing of computationally-expensive processes and simulations. Accessibility for the available resources can be improved by integrating Service-Oriented Architecture (SOA) into grid computing. This paper demonstrates how SOA and grid computing can be integrated to implement the rendering of POV-Ray 3D models efficiently.