Global Journal of Computer Science and Technology (GJCST)
Not a member yet
1830 research outputs found
Sort by
Smart Lock using Image Recognition
Get PDFIn the 21st century, for a digital lifestyle in a smart city, security is one of the core ingredients to ensure digital continuity. The existing smart security technologies use a smartphone, card, and fingerprint that need additional devices to carry or can spread out infectious diseases. Now it is high time to think about an additional device free and more timeconsuming technique. So, an intelligent system is proposed in this paper to secure our doors with face authentication. The human face is a unique and easy identifying feature of a human. The registered user2019;s image is saved in the device as a dataset to train the lock. The system recognizes the registered faces very fast and controls the hardware part to be unlocked. The system is capable of detecting and recognizing human faces from a real-time video. It is usable in the door lock, car lock, hutch, and many more security purposes
Classification of Hyperspectral Image using SVM Post-Processing for Shape Preserving Filter and PCA
Get PDFThis paper is based on an experimentation to preserve shapes of the natural classes in a hyperspectral image post classification of the image using SVM. The classifier classifies the vegetation types present in the hyperspectral image and then estimates the crop types present in the image. In doing so it preserves the spatial shapes of the vegetation types spread in the image using an Edge-preserving filter. The shape-preserving filter was applied prior to dimension reduction where by the low information content spectral components are discarded using Principal Component Analysis. The classification of the features is performed using SVM. The result has been found very effective in characterizing significant spectral and spatial structures of objects in a scene.
Optimized Load Centroid and Rabin Onion Secured Routing in Wireless Sensor Network for IoT
Get PDFAdvances in wireless communication have geared up extensive insights wherein the sensors can themselves communicate with other sensors that form significant parts of the Internet of Things (IoT). However, the large-scale acceptance of WSN for IoT is still surfacing threats and controversies that apprehend the security aspects. There are a lot of attacks that can manipulate the routein WSN for IoT. In this work, an Optimized Load Centroid and Rabin Onion Routing (OLC-ROR) method are designed to improve the throughput rate with minimum routing overhead and latency. The proposed method is based on a Centroid and Rabin Signature, a Digital Signature technique. First, the optimal route is identified by considering both the load and residual energy using Load Centroid function. Then onion routing is used for selecting secured route amongst the optimality. Besides, the node genuineness is checked by applying the Rabin Signature
Acceptance of Self-Driving Cars in United Arab Emirates
Get PDFTransportation industry witnessing a revolution of the emerging of self-driving cars which are autonomous vehicles that drive by itself without human involvement. It is expected that self-driving cars would have powerful feature and would provide a lot of benefits such as reducing traveling time, reducing traffic jams, reducing car accidents and many other benefits. The government of United Arab Emirates adopt technology implementation in all life aspects in the country starting by turning into smart government and then smart education and many other implementations of using technology in different aspects of the country. This adoption of technology positively affected UAE people2019;s intention toward accepting technology. As UAE government always adopt best technology practices, it is expected that United Arab Emirates would adopt the using of autonomous cars. The aim of this research paper is to investigate UAE people2019;s intention to turn into using self-driving cars. Researcher aim as well to explore the most common factors that may affect people2019;s intention to turn into using self-driving cars. This research paper methodology based on quantitative methods for gathering data in which questionnaire developed and sent to people live in United Arab Emirates
A Systematic Approach to English to Bangla Sentence Translator
Get PDFThis paper deals with the design and development of an expert sentence translation system. In this translator, the source language is English, and the target language is Bangla. The implemented translation system determines the relationship among different forms of English and Bengali sentences and makes appropriate correspondence between English and Bengali grammar. Here, we have been developing a top-down parsing program. The system incorporates itself with the dictionary and gives the corresponding Bengali meaning. The system performs translation procedure in three steps. The lexical analyzer reads the English sentence, tokenizes into words, and stores information into a stack. The lexical analyzer uses the English to Bangla dictionary and word morphology for finding lexical information. The parser parses the input sentence and identifies the types of it and finds tense, phrase, clauses, etc. The generator generates a Bangla sentence, which is equivalent to the given input English sentence. It uses the output of the lexical analyzer and the parser to make Bengali sentence. This system can translate all kinds of sentences. But the limitation is that it cannot handle semantic and contextual problem
Specifying a Model for the Study of Social Entrepreneurship
Get PDFThe purpose of the work was to specify a model for the study of social understanding A documentary cross-sectional and exploratory study was carried out with a non-probabilistic selection of sources indexed to international repositories considering the indexation year of publication and impact factor Discussion axes were established to address the problem although the research design limited the findings to the sample analyzed suggesting the extension of other repositories years and quality indicator
A Comparative Study between a Simulation of Machine Learning and Extreme Learning Machine Techniques on Breast Cancer Diagnosis
No full textBreast Cancer is a developing and most normal disease among ladies around the globe Breast malignancy is an uncontrolled and exorbitant development of abnormal cells in the Breast because of hereditary hormonal and way of life factors During the starting stages the tumor is restricted to the Breast and in the latter part it can spread to lymph hubs in the armpit and different organs like the liver bones lungs and cerebrum At the point when the bosom disease spreads to different pieces of the body it is going to metastasize The sickness is repairable in the beginning periods yet it is identified in later stages which is the fundamental driver for the passing of such a large number of ladies in this entire world Clinical tests led in medical clinics for deciding the malady are a lot of costly just as tedious as wel
Machine Learning Algorithm for Development of Enhanced Support Vector Machine Technique to Predict Stress
Get PDFStress is a common risk factor for many diseases A correct and efficient prediction model is required to predict stress levels for targeted prevention and intervention in the personal healthcare domain Before preventing the event of stress-related diseases stress should be detected and managed early However surveys are used to evaluate an individual s stress condition with ease of measurement and requiring little time However anything that puts high demands on a person makes it stressful This includes positive events such as getting married buying a house going to college or receiving a promotion Of course not all stress is caused by external factors Stress can also be internal or self-generated when a person worries excessively about something that may or may not happen or have irrational pessimistic thoughts about life This article aims to develop a predictive model to find the interruption of stress using an efficient way One of the successive machine learning algorithm is SVM This paper proposed to enhance the parameters of SVM which is used to improve the efficiency for predicting stress This article proposed an Enhanced Support Vector Machine classifier to predict Stress The stress dataset is downloaded from the Kaggle repository with 951 instances and 21 attribute
Impact Analysis of Advising and Counseling System for Under-graduation Students in Bangladesh
Get PDFIn the development process of a student, counseling and advising system plays a crucial role. By providing proper counseling through empathetic relationship, we can learn about the problems of a student. With the help of this learning, we can find out all of the negative issues that are happening with the students. Such as- why their results are poor, why they are irregular in their study and many more. Even providing counseling, we can know their weakness, strength, and opportunities. Not only that, but it also helps to build a good relationship between teachers and students. Students can overcome their gap that exists between students and teachers, and that will ensure good bonding among them. So counseling is a must factor to develop a qualified student. Before taking any decision against any student, it2019;s important to learn their present condition through counseling. That2019;s why proper counseling with an effective way is necessary for all students. To understand the effects of counseling and academic advising, we surveyed on Daffodil International University. With this survey, we tried to realize all of the facts related to it. Based on the analysis, we have also mentioned some recommendations which may be helpful for the betterment of the undergraduate students
A Survey on Load Balancing in Cloud Computing
No full textCloud computing is a biggest technology. It is different cloud services are SaaS, PaaS, aaS. loud users are able to access software and applications from whenever they need while it is being host by an outside party in cloud. Cloud services are bought on a pay-as-user-go or subscription basis. The cloud users getting into increased day by day. In this paper mainly focused on Load balancing within the cloud computing surroundings has a crucial impact on the performance. Smart load balancing makes cloud computing additional economical and improves user satisfaction. Load balancing may be a pc networking technique to distribute work load across multiple computer or a multiple cluster