Paper Title

IDENTIFICATION OF CORONAVIRUS DISEASE FROM CHEST X-RAY PICTURES USING CNN

Abstract

The detection of severe acute respiratory syndrome coronavirus 2 (SARS CoV-2), which is responsible for coronavirus disease 2019 (COVID-19), using chest X-ray images has life-saving importance for both patients and doctors. Therefore, it is very important to evaluate the affected patients quickly and cost-effectively to combat this disease. Radiographic examination is one of the best viable steps to achieve this goal, with chest radiography being the least effort and least expensive option that can obtain. This article proposed a whole convolutional neural network-based solution to locate five COVID-19+ patients using chest X-ray imaging. Several recent CNN fads, DenseNet201, Resnet50V2, and Inceptionv3, have been adopted within the proposed work. For my part, they have been adept at making unbiased predictions. The models are then shuffled to wait for the category rate. To verify the efficacy of the response, we openly used chest X-ray images of COVID + ve and –ve cases. A mean sensitivity of 93.84%, mean specificity of 99.18%, mean accuracy of 98.50%, and mean receiver operating characteristics–area under the curve scores of 96.51% are achieved. A convolutional neural network without pre-processing and with minimized layers is capable of detecting COVID-19 in a limited number of, and in imbalanced, chest X-ray images.
KEYWORDS : COVID-19, pneumonia, X-ray, convolutional neural networks, coronavirus

Paper Title

LOSSLESS MEDICAL IMAGE COMPRESSION USING REGION OF INTEREST

Abstract

Typically conducted on the small portion of an image in therapeutic field scanning, but because of its immense image size, it comes to high costs and less transmission speed. Gradually, it is incredibly difficult to reduce the scope of the image without data loss. In this case, we will reduce the size of the image by using certain compression techniques that have the potential to achieve higher reconstruction efficiency. Be that as it might a picture\'s existence plays a key role in the prescription. We have proposed optimized compression methods for this purpose. We perform region-based segmentation techniques as part of this technique by dividing the image into two regions, ROI and NONROI, which are monitored by applying compression techniques for LOSS and LOSSLESS. We can measure the compression ratio and discern the best suited wavelet technique by applying different wavelets for ROI and NONROI pictures. We will fulfill our prerequisites at that point i.e. picture with high speed of transmission, less cost and less size. Finally, compacted images should be transmitted through flexible administrations and web administrations. In this article, it assures assistance for the vulnerable community with regard to the cost-effectiveness of disease sufferers.
KEYWORDS : Field scanning, ROI and NONROI

Paper Title

A RESEARCH ON NETWORK SECURITY IN CLOUD COMPUTING

Abstract

This paper is the research of facts within the public cloud and associated views to it concerning safety. The paper will pass into subtleties of records safety techniques and techniques utilized sooner or later in the arena to assure maximum immoderate records coverage with the resource of lessening dangers and risks. Accessibility of information within the cloud is extraordinary for a few applications however it provides risks through supplying facts to applications that may add as of now have protection getaway clauses in them. Furthermore, utilization of virtualization for allocated computing may also additionally hazard facts whilst a tourist OS is administered over a hypervisor without knowledge of the unwavering excellent of the traveller OS which also can have a protection getaway clause in it. This paper will likewise deliver expertise on statistics safety views for information-in-Transit and records-at-rest.
KEYWORDS : Virtualization, Public Cloud, Hypervisor, information-in-Transit, records-at-rest.

Paper Title

PRIVACY PRESERVING IN DATAMING BY USING DIFFERENT ALOGRITHEMS ON HYBRID PARTITIONAL DATA SETS

Abstract

Maintenance of privateers in records mining has emerged as an absolute prerequisite for replacing private records in phrases of records analysis, validation, and publishing. Ever-escalating net phishing posed intense danger on vast propagation of touchy statistics over the web. Conversely, the doubtful emotions and contentions mediated unwillingness of various records carriers toward the reliability safety of facts from disclosure regularly consequences utter rejection in facts sharing or incorrect information sharing. This article affords a panoramic assessment on new angle and systematic interpretation of a listing posted literatures through their meticulous company in subcategories. The essential notions of the prevailing privacy preserving records mining strategies, deserves, and shortcomings are supplied. The cutting-edge privacy retaining statistics mining strategies are labelled based on distortion, association rule, disguise affiliation rule, taxonomy, clustering, associative category, outsourced information mining, distributed, and k-anonymity, in which their extraordinary blessings and downsides are emphasized. This careful scrutiny reveals the beyond improvement, gift studies challenges, destiny traits, the gaps and weaknesses. Similarly big upgrades for extra robust privateness protection and preservation are affirmed to be mandatory.
KEYWORDS : Distributed data mining, privacy preservation, association rule, clustering, classification, secure multiparty computation, trusted third party.

Paper Title

SAFETY MANAGEMENT IN BUILDING CONSTRUCTION SITES: INDIAN CONTEXT

Abstract

India is a making creating and most patterns are happening in building industry. India has most labor supply to utilize as alternatively accessible to the construction work to development industry. Across the world, it is felt that acceptable and assurance being developed venture is depending of concern. In India this is among quite possibly the most fundamental issue. In India insurance of worker is currently not taken as prevalent difficulty because of which numerous representatives lose their ways of life because of absence of assurance on building site. The preeminent reason for this learns about is to hold best of work and to diminish mishap on building site. The aim of the study is to identify and evaluate the safety management in building construction sites. The study also focused on health and safety issues and minimizes the accidents during construction work. The review of literature associated with the research is the beginning step in obtaining the facts about the related studies. The review of literature provides background knowledge about safety management that guided the design of the questionnaire. Questionnaire is distributed among the management of the construction industry. The study was conducted among the 20 management personal working in building construction industry. Questionnaire is used to collect responses from experienced respondents. The result of this study is analyses the different responses regarding safety management. In questionnaire survey Likert scale 1 to 5 used for strongly disagree, disagree, moderately, agree and strongly agree respectively. The result of the study determined via relative importance index method and their respective ranks has been generated against each response. This study also helps to know the problems occurred to the management for implementation the safety rules.
KEYWORDS : Safety Management, Building, Construction Sites

Paper Title

DETECTION OF FAULTS IN TRANSMISSION LINE SYSTEM WITH GSM

Abstract

The efficiency of power systems is largely determined by the effectiveness of the inbuilt power equipment. Monitoring transmission lines for faults and quick isolation of the system from faults helps to improve the efficiency of the power systems reliability. Technologies such as Power line carrier communication and the use of power based communication systems have their respective demerits. In this paper we are going to design a system to provide a reliable monitoring and fault detection in transmission lines. If any fault occurs in transmission lines, it is highly difficult task to find out which type offault has occurred as well as a person should monitor all those things for every time. By chance if there is no one in a controlling room, if at that time fault occurs it causes dangerous hazard. So to avoid this we are going to implement Monitoring and alert in Electrical Transmission Lines System with GSM. By using this system we will identify which type of fault occur in transmission lines and that will be displayed on LCD. Whenever fault occurred a buzzer sound will be given as alert. For reference herein this paper we are taking Transmission lines at different Voltages and finding open circuit fault and Short Circuit fault, whenever fault occurs it can display on LCD which type of fault occurred and as well as give the alert to particular area. And we will be using GSM technology to send a message to the controller at sub-station.
KEYWORDS : GSM, Raspberry Pi