Manuscripts

Recent Papers

Research Paper

Intelligent assistance and object detection for visually impaired people

This paper is intended to design a smart cane that helps visually-impaired people to walk more confidently. The study hypothesizes that a smart cane that alerts visually-impaired people about obstacles in front. The system is intended to provide artificial vision and object detection, real time assistance via GPS by making use of Raspberry Pi. The system consists of GPS module, and the feedback is received through audio, voice output works through TTS (text to speech). The proposed system detects an object around them and sends feedback in the form of speech and also provides navigation to specific location through GPS. Smart Stick is a device designed to aid the visually impaired person by detecting objects and portray the information to them in the form of speech. This reduces the human effort and gives better understanding of the surrounding. Furthermore it also provides an opportunity for visually impaired people to move from one place to another without being assisted by others. The device can also be used in old age homes where old age people have difficulty in their day to day activities due to decreased vision. The main focus here is to aid people in need to “see” the surroundings. Considering the growth in the field of artificial intelligence and features like object detection is getting easier and computationally feasible, these features are implemented in the proposed system. The main focus is on object detection and classification on pictures which are captured by the device mounted on a stick. This interpreted information can then be relayed to the user in means of sound or speech.

Published by: Pranali Gabale, Shalaka Javheri, Shital Mule, Priti Shete

Author: Pranali Gabale

Paper ID: V7I3-1867

Paper Status: published

Published: June 15, 2021

Full Details
Dissertations

Efficacy of pilates and MC Kenzie exercises on patients with chronic low back pain – A comparative study

The major health issue of human being is lower back pain disease caused by various Lifestyle activities and occupational adaptation. People who continue to in same poster throughout the day like computer professionals and shops may be the major victims of this chronic disease process. Objectives : The study aims to produce the different kinds of Management techniques to deal with chronic low back pain and compared their effects on the pain relieving and core muscle strengthening. Methodology: The study focused on 40 subject into two groups such as group A and group B respectively treated with Pilates and Mc kenzie exercise on management and and core strengthening. The effects of the both types of treatments more assessed by visual analogue scale, back performance scale score and Roland Morris disability questionnaire. Result and Conclusion: As per the various statistical analyses, it was found that both the methods are less effective for relieving pain and group A got effective improvement in core muscle strength. So as this study conclude that the alternative hypothesis has been accepted.

Published by: Kanagaraj R.

Author: Kanagaraj R.

Paper ID: V7I3-1804

Paper Status: published

Published: June 14, 2021

Full Details
Research Paper

The design of ultra wide band circular monopole antenna with Quad-Band Notch characteristics

A novel microstrip fed Ultra wide band (UWB) with quad notch characteristics is presented in this paper to avoid the interference of narrow bands. The proposed antenna is having the size of 55x56x1.6 mm ^3 falsified on FR4 epoxy substrate. Based on the methodologies, the first notch is created to attenuate the WiMAX (World Interoperability for Microwave Access) of range 2.94-3.7GHz by using T slot. The two rectangular slots are used to reject the INSAT (Indian National Satellite) band of the range 4.5-4.8GHz and of WLAN (Wireless LAN) band of the range 5.1-5.9GHz. These three notch band characteristics are attained by etching the notches on the radiation patch and the fourth notch is created by etching U shaped slot in the feed which attenuates the ITU (International Telecommunication Union) band of range 7.4-8.7GHz. All the simulations are performed using HFSS software. The simulated antenna yields a good impedance bandwidth of 2-11GHz with Return loss (S11) ≤ -10dB, VSWR< 2 and exhibits good radiation properties except at the notch band ranges. The peak gain of more than 2.2dBi and efficiency of above 90% is obtained from 2-11GHz excluding at the notching narrow bands.The obtained results depicts that the antenna is very well suitable for real time UWB applications.

Published by: Vaka Gayathri, Gumma Kalyani, Thulabandula Venkata Lakshmi Ravali, Nuti Ramya Krishna, Rayapati Ruchitha

Author: Vaka Gayathri

Paper ID: V7I3-1620

Paper Status: published

Published: June 14, 2021

Full Details
Research Paper

IoT Based Healthcare System

Looking at the current pandemic situation in the world, every human being's health has been a major concern. In this scenario, the best option for dealing with an epidemic is an IoT-based health monitoring system. The Internet of Things (IoT) is a recent internet trend that is becoming a popular research topic, especially in health care. The key goal of this project is to keep track of a patient's heart rate, temperature, and oxygen saturation (Spo2) in real-time. Aside from that, the device monitors the patient’s positions using tilt, temperature, humidity, and CO2 level in the room. Tilt is a core aspect of our project. The Arduino Nano is wired to all the sensors. The Arduino nano is a microcontroller that manages the system's functions. Arduino uses a Wi-Fi module to send data to an IoT platform through the ThingSpeak wireless network. In this system, Arduino uses a Wi-Fi module to send data to an IoT platform called ThingSpeak. ThingSpeak is used to store, retrieve and visualize our real-time data. So authorized individuals can access the patient's measured data from any location and take appropriate actions if it is required.

Published by: Shivani Purohit, Drashti Sojitra, Mohit Chawla, Kriti Vishesh Jaiswal

Author: Shivani Purohit

Paper ID: V7I3-1773

Paper Status: published

Published: June 14, 2021

Full Details
Research Paper

Vehicle detection and classification using Deep Neural Networks

Insightful transportation frameworks have recognized a proportion of consideration somewhat recently. In this space vehicle arrangement and restriction is the key errand. In this assignment the greatest test is to separate the highlights of various vehicles. Further, vehicle grouping and identification is a difficult issue to recognize and find in light of the fact that wide assortment of vehicles doesn’t follow the path discipline. In this article, to distinguish and find, we have made a convolution neural organization without any preparation to group and identify objects utilizing a cutting edge Deep neural organization dependent on quick locales. In this work we have considered three kinds of vehicles like transport, vehicle and bicycle for grouping and recognition. Our methodology will utilize the whole picture as information and make a bouncing box with likelihood evaluations of the element classes as yield. The aftereffects of the investigation have shown that the projected framework can significantly improve the exactness of the discovery..

Published by: Sathishkumar M., Kavya N., Kowsalya P., Nandhini G, Kaviyadharshini S.

Author: Sathishkumar M.

Paper ID: V7I3-1706

Paper Status: published

Published: June 14, 2021

Full Details
Research Paper

Video anomaly detection

Security is always a main concern in every domain, due to a rise in crime rate in crowded events or suspicious lonely areas. Abnormal detection and monitoring have major applications of computer vision to tackle various problems. Due to the growing demand in the protection of safety, security, and personal properties, the needs and deployment of video surveillance systems can catch all the anomaly and anomalies play important role in surveillance detection. Anomaly detection is used to identify the odd pattern or abnormal behaviors in the surrounding. Surveillance videos are able to capture and store a variety of anomalies in the surrounding. People detection and tracking are of important research fields that have gained a lot of attention in the last few years. Most of the time, the timely and accurate detection of video anomalies is the main objective of security applications. The video anomalies such as anomalous activities and anomalous entities are defined as abnormal or irregular events that are different from normal events happening every day. In this paper, we present how surveillance videos are used to detect anomalies.

Published by: Vishal Singh, Keyur Madane, Himanshu Sugandhi, Ankit Joshi, B. S. Satpute

Author: Vishal Singh

Paper ID: V7I3-1859

Paper Status: published

Published: June 14, 2021

Full Details