Manuscripts

Recent Papers

Research Paper

Face recognition based attendance monitoring system

This project is to construct a face recognition-based attendance monitoring system for various educational institutions for taking attendance of students accurately. Previously attendance was taken manually where the professor/ teacher used to record attendance of students in log registers. This process is time-consuming and hectic. The professor/teacher might miss marking the attendance for few students or an illegal third party might manipulate the log register. The accuracy and security were at stake always. Then there was a new attendance system using fingerprints. This system might be accurate but there is a chance of stealing fingerprints from the biometric device for malicious activities. In this project, we use the face which is the most important part of our body. The facial features of individuals are recognized and analyzed for taking attendance. In this project, we use one of the most popular programming languages Python, and its library called OpenCV which is used to capture live images and videos. The face images and details like registration number and name of the student will be saved in files and attendance notes can be seen in Excel Spreadsheet. The image detection and processing are done using the LBPH algorithm which is known for its simple yet efficient performance.

Published by: Keerthi Sree Ganne, B. Prajna, Golagani Naga Kalyani, G. Neha Gagana, Gurugubelli Rajani

Author: Keerthi Sree Ganne

Paper ID: V7I4-1314

Paper Status: published

Published: July 12, 2021

Full Details
Research Paper

Impact of COVID-19 Lockdown on Air Pollution and Air Quality Index (AQI) in India

From the start of industrialization, air pollution in India and around the globe started increasing with new pollutants coming into existence. The extent of air pollution has also increased in the last few decades, very rapidly affecting human health and the earth's ecosystem badly. During pandemics like it happened in other countries, the same way, in India also government-imposed lockdown that resulted in the termination of all economic activities. The pandemic and imposed lockdown provided improved air quality as a gift to mankind. This duration also shows insight into the relationship between the economy and good air quality. We also get some insight into how to improve air quality without affecting the basic needs of society. This paper aims to examine the variation of the concentration of 5 main contaminants (PM2.5, PM10, NO2, CO and Ozone) before and after the lockdown imposed in 5 major Indian cities. After reading this paper, we get an insight into to what extent we can reduce air pollution by imposing lockdown and how the socio-economic, geographical variance across cities decides the extent of reduction in air pollution. We also get direction for future management strategies and policies to regain economic strength with sustainable improvements. So at the moment, we have to focus on both air pollution management and economic recovery. This study provides insight into the air quality index across the geographical variation of India.

Published by: Ankit Kumar

Author: Ankit Kumar

Paper ID: V7I4-1309

Paper Status: published

Published: July 12, 2021

Full Details
Research Paper

Traffic rule violation detection and reporting system

The increasing number of vehicles in cities can cause high volumes of traffic, and implies that traffic violations become more critical nowadays in India and also around the world. This causes severe destruction of property and more accidents that may endanger the lives of the people. To solve the alarming problem and prevent such unfathomable consequences, traffic violation detection systems are needed. For which the system enforces proper traffic regulations at all times, and apprehends those who do not comply

Published by: B. G. Vinayak, Rakesh M. R., Santosh Reddy P.

Author: B. G. Vinayak

Paper ID: V7I4-1301

Paper Status: published

Published: July 12, 2021

Full Details
Research Paper

Reducing crude protein level in commercial layer chicken feed by balancing digestible amino acids

A total of 10 commercial layer farms in the stage of phase-I laying period (25 to 40 weeks) reared in raised platform caged house with uniform management wereselected for the experiment. In each farm, 4000 birds will be selected and divided into two treatment groups of 2000 birds each. The feed formulation with reduced crude protein level (15 per cent) compared to standard level of 17 per cent will be formulated by balancing the digestible amino acids (Lysine, methionine+cysteine, threonine, tryptophan, arginine, isoleucine and valine). Feed ingredients used in feed formulation were maize, cumbu, broken rice, soya bean meal, sun flower cake, deoiled ricebran, rapeseed meal, fish meal,etc. Based on the above study it could be concluded that crude protein reduction is one of the way to reduce feed cost in commercial layer chickens, but should be properly balanced with all the essential digestible amino acids rather than total amino acids basis. But the quality of feed ingredients, loss in body weight and egg weight is the area of concern.

Published by: Sakthivel D., Mani K., Kannan D., Vasanthakumar P.

Author: Sakthivel D.

Paper ID: V7I4-1288

Paper Status: published

Published: July 12, 2021

Full Details
Research Paper

Algorithmic Trading with Deep Learning

In this paper, Artificial Neural Network (ANN) is used to predict trading signals like Strong Buy, Buy, Neutral, Strong Sell and Sell a total of 5 signals. The ANN outputs one of the trading signals based on the market data of cryptocurrencies. The ANN makes use of the candle-stick data along with additional data like Volume, Number of Trades, etc and technical indicators to predict the signals. The ANN is trained on multiple intervals of the market, i.e. 1 minute, 3 minutes, 5 minutes and so on up to 3 days in the intervals. Along with the market data, the paper also makes use of the news articles to better predict the trading signals, this is only used when the ANN predicts the Buy or Sell, and by using the news articles those Buy and Sell signals are converted to Strong Buy and Strong Sell respectively. The ANN was trained on 3 years of data, and Google’s BERT [1] model was trained on almost 1000 news article’s titles on topics related to cryptocurrencies and the model outputs whether it’s a positive or negative title. The accuracy of the ANN models for all the markets is in the range of 85 - 90% and the same accuracy is observed with the BERT model which is trained on the news article’s title.

Published by: Mayank Anuragi, Rakshit Ramnath Naik, Manoj Galanki, Kavitha S. Patil

Author: Mayank Anuragi

Paper ID: V7I4-1286

Paper Status: published

Published: July 12, 2021

Full Details
Research Paper

City cleanliness using geo-tagged images with experimental results

During the process of smart city construction, city planners and managers always spend a lot of energy and money for cleaning street garbage due to the random appearances of street garbage. Consequently, visual street cleanliness assessment is particularly important. However, the existing assessment approaches have some clear disadvantages, such as the collection of street garbage information is not automated and street cleanliness information is not real-time. To address these disadvantages, this paper proposes a novel urban street cleanliness assessment approach using mobile edge computing and deep learning. First, the high resolution cameras installed on vehicles collect the street images. Mobile edge servers are used to store and extract street image information temporarily. Second, these processed street data is transmitted to the cloud data centre for analysis through city networks. At the same time, Faster Region-Convolutional Neural Network (Faster R-CNN) is used to identify the street garbage categories and count the number of garbage. Finally, the results are incorporated into the street cleanliness calculation framework to ultimately visualize the street cleanliness levels, which provides convenience for city managers to arrange clean-up personnel effectively. Index Terms— Smart cities, street cleaning, garbage detection, deep learning, mobile edge computing.

Published by: Krishna Sonawane, Prachi Pingle, Siddhant Lunawat, Trupti Zagade, Dr. Meenakshi Thalor

Author: Krishna Sonawane

Paper ID: V7I4-1285

Paper Status: published

Published: July 12, 2021

Full Details