International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 09 Issue: 04 | Apr 2022
p-ISSN: 2395-0072
www.irjet.net
Survey of ‘Nivritti’: survey of face recognition technologies/methods Sujay Ingle, Tejas Khangal, Yash Bhuse, Nitish Das 1,2,3 Student,
Department of Computer Science and Engineering, MIT School of Engineering Pune, Maharashtra, India 4Professor, Department of Computer Science and Engineering, MIT School of Engineering Pune , Maharashtra , India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Nivritti is a software web application which
There has been a significant amount of increase in corruption as Sometimes the Pension Seeker has to Bribe the Treasury Department Officials to get their Jeevan Pramaan approved.
will provide a Digital Platform to solve the Pension Problem of India using Image Recognition Technology. Nivritti will provide services to verify the Jeevan Pramaan Certificate, the Pension to Bank Service & a special SEVA Service(For the Physically Challenged). With its easy-to-use UI UX, the Pension Seekers will now be able to collect their Pension with very few steps & procedures to follow saving their time & effort.
Standing in long queues to receive the Pension Amount leading to Exhaustion. Threat of theft and delay in receiving pension along with lack of coordination from system.
Key Words: HTML, CSS & Bootstrap, Javascript, Python Script, KERAS, TensorFlow, OpenCV, NumPy
Nivritti solves all the above problems by using the Modern day Technology of Face Recognition. It not only simplifies the process but also increases the Transparency in the system. Hence, as the world is getting Digital it is much necessary to make the Government Schemes and Policies to go Digital. This will not only reduce the Corruption but also increase the Revenue of the Government.
1. INTRODUCTION Receiving Pension has always been a tedious process involving long waiting hours, delay and lack of transparency in the activity of the system. ‘Nivritti’ is an application developed for the Pensioners of India which will reduce the effect of this exhausting process. This web app will solve the issues of the Pensioner by providing the service of online verification process using face recognition techniques for the Jeevan pramaan(Life Certificate for Pensioners) & transferring the pension amount from the Pensioners pension account to their savings account. This Web App is specifically built for the use of the senior citizens (especially the government employees) who want to reduce the effect of receiving their necessary pension. Basically this system which not only improves transparency behind the procedure but also picks out the mismanagement while making it a feasible process uses the Face Recognition Technology to enable the Pension for the Pension Seeker..
1.2 Existing methods used for face recognition Methods used till now gave high accuracy , however the drawback was the amount of data required for training purposes( which becomes even a greater issue when they need to be uploaded to a server). Also, predictions made by these processes were time-consuming. The main reason behind our work is to come up with a way that can do the recognition task in as few images as possible and to make the system deployable. Few methods are described below: DeepFace has been trained on a huge dataset of people from various places. It uses 3D alignment steps to capture facial features. After the alignment step it assumes portions of the face to be fixed at certain pixel levels. Thus it builds its neural net according to this assumption ,as a result of which it requires less convolutional layers.
1.1 Need for Nivritti It is quite evident that our Pension System has been a Subject of concern for the past few decades. The main reason being the lack of Transparency in carrying out the process of issuing the Life Certificate as well as the Transactions involved in Receiving the Pension Amount. Here is the list of Some Major Issues faced by the Pension Seekers:-
© 2022, IRJET
|
Impact Factor value: 7.529
DeepID method becomes more generalized by increasing the classes it needs to classify faces into. Along with deep Convnets, it focuses on high level extraction of over-complete features. Face net optimizes the embedding process. With a small batch of images ,it builds its initial model , which can be reused without training when more batches of images are
|
ISO 9001:2008 Certified Journal
|
Page 3158