Research

LIST OF PAPERS PRESENTED AT NATIONAL CONFERENCE HELD AT THE DEPT ON 17-06 -2020
Sl. No.	TITLE AND NAME OF AUTHORS
1	Traffic Congestion Control System Usingimage Processing Monisha R, Yeshodharani, Anusha M., Prof. Puneeth H J.
2	Real Time Driver Drowsiness monitor assisted with Cell Phone detection based on video sequences of the Cabin Saathvik R Khatavkar, Shiv Swarup, Shreyas M S, Varun H H, Ghousia Begum S.
3	Data Extraction of Digital Text and Conversion into an Audio-Visual Output Niharika Rajaram, Nikhila M.Y, Rajeev Koushik Y G, Monisha R, K. Venkata Rao
4	Aerial crop monitoring system for site specific crop management Nikhil Kevin Jones, Nikhil Sanju J, Rahul M S, Dr. S V Uma
5	Advanced Secure Voting System using IOT Suhas M Angadi, Mayank Kumar, Sunny Kumar, Dr. Mamatha A S
6	Object detection for visually impaired M Darshith, M S Karthik, Smitha R D, Monica R.
7	Machine Learning analysis of Emotion detection in Adults Mohan H, Dr Smitha N
8	Face Emotion Detection Dr. Suresh D, Rohit Kumar K, Subin S, Sanket Shanbhag, Naveena N
9	Automatic extraction of retinal vascular tree Divya B K, Shri Ramya S, Meghana T J, Yashaswini V, Manohar P
10	Application for Gender, Age and FaceRecognition Using OpenCV Ms. Prasanna N, Mr. K. Venkata Rao
11	Efficient Data Embedding in Image Steganography using Deep Learning Mrs.Sangeetha K, Manoj R, Darshan S N, Kushal S, I S Goutam.
12	Study of Image Compression Models & Algorithms Shraddha Singh, Sharath Pratap, Sharmistha Kulshrestha.
13	Brain Tumor Segmentation And Classification Using CNN Vijay Devaramani, Vinyas, Vishal Singh Bais, Chandan BM, Ms.Leena Chandrashekar.
14	A Survey on Cancer Detection and Diagnosis using Image Processing and Machine Learning Ramana gowda G, Manjula V K.
15	Real-time Sign Language Recognition with Convolutional Neural Networks Akshay Kumar Rai, Kundan, Sanith Nair, Prakash Tunga.
16	Medical Prescription Management System Abhas Agrawal, Rahul Kumar, Siddharth Gupta, Dr. Vipula Singh
17	Compression of Hyperspectral Images Aman Prakash Singh, Meghana V Patil, Saumya, Sushant Shekhar, Dr. Mamatha A S.
18	Image Registration Using Convolution Neural Network For Brain Tumor Detection Chandan Shekar, Manoj Raj M, Vinay R, Naveen Rao V, Leena C.
19	Realtime Multi-person Human Pose Estimation and View-independent action recognition Amith Mukesh Gaba, Anirudh M, Jaya Sai Abhishek Reddy Munagala, S R Vigneshwar, Dr. Vipula Singh

  

  List of Papers presented on the day of National conference on 03/05/2019:

Conference Title : IETE Sponsored Second National Conference on Emerging Trends in Engineering, Science and Technology

LIST OF PAPERS PRESENTED AT NATIONAL CONFERENCE HELD AT THE DEPT ON 03-05-2019
Sl. No.	TITLE AND NAME OF AUTHORS
1	Wavelet Filters Performance Analysis for Image Compression Devika S , Shreya A, Bharathi Gururaj
2	Digital Image Filtering using α -Centroidal and Heron Means Vivek B A, V Keshav Bharadwaj, Abhinav Narayan, Sampath Kumar R
3	IOT - Camcorder Based Anti-Piracy Screen with Robust High Capacity video Steganography Rakshith kumar H, Dr. Rajini V Honnungar
4	The Implementation of Convolution Neural Network in Landmine Detection using Feature Extraction. Aditya S Chippalkatti, Akash R, Anubhav Verma, Dheeraj H A, Smitha N
5	Silkworm Egg Counting and Monitoring Based on Image Processing and Internet of Things. Mahalakshmi Bandi, Madhushree.R.P , Lokesh, Navneeth Achar  , S V Uma
6	Remote Sensing Satellite Image Segmentation based on Colour-thresholding ARAVINDHA. M, SAMRAH, HARISH DEVADIGA, PRAVEEN. G
7	Image Processing based autonomous fetchbot for garbage collection. Sanjay V Joshi, Sathish R , Shreeharsha C M , T S Suhas, Ganesh Yernally
8	Image Processing techniques using genrative adversarial networks K.Hari vinayak sai, Kaustubh dilip gaikwad, A.Mahesh reddy, G.Sri vidya
9	A Roboust Secure Image Transmission Technique via secret-fragment-visible mosaic images. K.Sundeep, S.V.S.S.Narasimharao

LIST OF PAPERS PRESENTED IN NATIONAL CONFERENCE HELD AT THE DEPT ON 04-05-2018
PAPER No.	TITLE AND NAME OF AUTHORS
1	Portable GET Sindhu P, Shwetha S R, Spoorthy D, Yogesh Babu S, Narendra Kumar
2	Glaucoma Automated Diagnosis System Classification using Advanced Transform Techniques of various Features and Wavelet Types Mohan kumar N, Dr . Suresh D
3	ECG and PPG evaluation for blood vessel stiffness estimation Premila, Dr.C.N Prabhavathi
4	An efficient Diabetic foot monitoring system Swathi G,Dr. Mamatha A.S
5	Image processing and bone conduction for the deaf and blind people M V Ritu,,Manasa N, L,Ranjani, Dr C.N Prabhavathi
6	Volumetric statistical estimate preparation instrumental analysis Akash.R
7	CCIT and IBM Whitening in image processing Roshni U Singh, Manjula V.K
8	A Neuroevolution Approach to an Arcade Game called Mario using NEAT Algorithm K. Hari Vinayak Sai
9	Image registration for ct and MRI brain images Aishwarya C, Anupriya D V, Leena Chandrashekar
10	Hand Gesture Interpreter that interprets sign and converts it into Speech and Text using a Smart Glove Shyam Sundar. D.G, Sushwath Chandra D, Shridhar Ramakrishna Choudhari
11	Face recognition security system Amruth.R, Bharatesh Bhat, Deepak P, Ganesh S, Prakash Tunga
12	Detection of Pedestrians, Sign boards and Road condition using Computer Vision and Deep Learning for Highly Automated Driving Priyanka, Pranesh A G Rao, M S Swaroop, Nikhil Joshi, Ghousia Begaum
13	Design and Development of a Prototype System for NadiPariksha Sri Harsha.M, Prashant, Jya Surya G K, Suhas N, Narendra Kumar
14	Brain Tumor Segmentation using K-means AmulyaNanjajjar, Disha D Bhat, Chandana B G, Ankitha Uday Hulekal
15	Water meter reading using image processing Abhishek S K, Adarsh M Kashyap, Anush N,Geetha Ganesh
16	Wireless body area network with emg and inertial parameter sensing and processing Aditya D Sharma, Dheeraj V, Manjunath S K,Ohileswari M.S
17	Brain Tumor Segmentation of Fused MRI and CT Images U.P. Prajval,Apoorva K.N, ShreyalTelang,Sunaina B.N,Leena Chandrashekar

Abstract of all the papers

Paper1

Abstract - There has been an exponential increase in health care costs in the last decade. Seniors have to make frequent visits to their doctor to get their vital signs measured. There is a  huge market for non-invasive methods of measurement of these vital signs. This paper gives an idea about the design and implementation of a reliable, cheap, low powered, non-intrusive, and accurate system that can be used on a regular basis to monitors the vital signs and displays the output to the user’s cell phone. This data is also easily accessible by the physician through wireless network. The proposed model called the portable HET specifically deals with the signal conditioning and data acquisition of three vital signs: heart rate, ECG, and body temperature. Electrocardiogram is obtained by attaching skin surface electrodes on the patient’s wrists and legs. The heart rate is measured using a heart rate sensor. Body temperature is measured using LM35 sensor. The theory, design procedures, experimental results and discussions of these systems are presented.

Paper2

Abstract: Correntropy is a measure of statistical and temporal information that is becoming more and more significant with the expansion of non-linear domain. The result, varies with different frequency bands proposed in recent years. This paper briefly discusses the different techniques like transform techniques such as Emipirical wavelet transform, image morphological operation using subjective analysis method. To begin with, applying an pre-processing technique, the optic disk, cup and vessel extraction is done. The cup disk ratio describes the statistical features and grey level cooccurance matrix The proposed method is implemented on a database in order to measure the performance characteristics such as Signal-to-noise ratio, Peak-signal-tonoise ratio, and area. The same has been implemented on a wavelet feature extraction .Mexican , Morlet and Harr wavelets are classified using SVM classifier along with neural networks.

Paper3

 Abstract—  Numerous methods present extraction of heart rate with PPG signal when physical activity are carried out , But during physical exercise the signal is tend to have large interference which causes. Also PPG has a greater  role in detecting heart rate variability , When PPG is combined with ECG a greater insight into body circulatory system can be achieved. Here in this paper for ECG signals, we propose a new spectral masking approach which modifies a pan Tompkins algorithm with the use of wavelet, and for PPG signals we use the principle of first derivate to extract notch region and other parameters such as peak and valley. To add further PPG can be used to derive coarctation in vessels. Performance is verified using Qripto-Coarc monitor. Instantaneous rate is observed to provide more information than the average heart rate which gives us a information into nervous system activity. With instantaneous rate, PPG variation from radial, femoral artery and dorsal Pedis would help in detection of aortic coarctation a profound disease in neonates.

Paper4

Abstract : Computer aided diagnosis is a recent advancement in the field of healthcare diagnosis, which involves data mining combined with classification. The main objective of CAD system is aiding doctor to perform a better diagnosis by having an objective visualization of the disease and hence have a better validation and treatment plan. Here, we focus on pre-processing and segmentation methods to effectively extract diabetic ulcer from foot, by using phase congruency analysis and Segmentation algorithm. The work would aid doctors to treat diabetes patients who would have issues related to leg ulcer which would take longer time to heal as they would be prone to have gangrene. With the necessary algorithms continuous monitoring of this wound portion could be possible so that it would help to heal the patient at prior stage. The proposed methodology has been developed in MATLAb environment and the efficiency of methodology is evaluated using statistical parameters.

Paper5

Abstract- When hearing and sight fail, the human need for communication continues to remain intact. Deaf-blindness is a unique disability that combines varying degrees of both hearing and visual impairment. Over time and with the advancement in technology, a variety of communication technologies are available for deaf-blind individuals. Despite the available communication technology for deaf-blind individuals and the advancement in the field of Image Processing, we still lack a system/device to help deaf-blind individuals go independently. With the help of Image Processing, bone conduction and the education provided to them we can develop a warning system that provides environmental awareness to deaf-blind people, thus helping them participate in social norms independently. This article provides the methodology to design and implement a device for the deaf-blind people. Finally, some potential topics and challenges for future research are highlighted.

   

 Paper6

Abstract- This mechanism estimates the weight of vegetables and prepares a periodic estimate of requirements and provides dynamic and statistical analysis. The main concept of this mechanism is to be able to analyze and estimate the weight of a physical commodity without actually weighing it. The dynamic and highly volatile vegetable market is taken into consideration here because it is one of the prime quantitatively analyzed market which keeps continuously changing with respect to two factors – quantity and cost. Taking these two factors into consideration, the mechanism is able to assist and help give the marketer a complete view of the economics of the business and prepares an estimate for future stock renewal requirements.

Paper7

Abstract— Communication system aims to transmit and receive the data successfully without any errors. A transmission system is usually optimized for a certain frequency range, and the transmitting components may not operate in optimal range if the average signal input is not centered at required frequency. A data stream could generate a modulation signal with low frequency and does not pass easily through receiving components. By randomizing the data with a known set of values and then recovering back the transmitted signal at receiving end, the signal values will be more closer to the center of transmitting frequency range. This paper describes the method of randomizing , data whitening process.

Paper8

Abstract: In recent years, the rise of AI has been quite prominent both in academic and commercial ventures. One aspect that has been explored in the use of AI is using computer games as a method to both test and train AI. Video and computer games often use Artificial Intelligence (AI) code to give life to nonplayable characters. A major challenge for AI is to develop agents that can learn to perform many different tasks. This paper describes such agent capable of playing a classic arcade game - Mario without any prior knowledge using NEAT algorithm. It takes ‘fitness’ as a parameter and evolves new species of neurons through mutation and breeding and after a few generations, the Mario completes a level by itself.

Paper9

Abstract— This paper deals with different procedures to perform image registration. Image registration is the process of transforming different sets of data into one coordinate system. Data may be multiple photographs, data from different sensors, times, depths, or viewpoints. It is used in computer vision, medical imaging, military automatic target recognition, and compiling and analyzing images and data from satellites. Registration is necessary in order to be able to compare or integrate the data obtained from these different measurements. Here we will deal with two basic procedures of registration, that is affine and DFT registration.

Paper10

Abstract---In the present generation with all latest technology which helps physically disabled persons, there are very few problems where in the technology has been restricted to certain group of people One kind of such problem is the communication between a deaf and dumb person with a normal person, where these person tries to communicate with their hand gestures which are certainly not understood by normal persons.  Sign language is the communication medium for deaf and mute people.        This paper proposes a novel approach of interpreting the sign language using a portable smart glove. The flex sensors present on each finger senses the signing gestures and couples the analog voltage to the arduino. The arduino converts these analog values to digital samples and through the course of transceiver and arduino  pair the sign language is converted into an audio speech which is useful for the mute people  and also displays the intended message on an LCD display which is useful for the deaf persons. Basically, this smart device is used to recognize sign language of deaf and dumb.

Paper11

Abstract - in today’s world, face recognition is an important part for the purpose of security and surveillance. Hence there is a need for an efficient and cost effective system. Our goal is to explore the feasibility of implementing face recognition system using conventional face detection and recognition techniques such as Haar detection and LBPH (Local Binary Patterns Histogram). This paper aims at taking face recognition to a level in which the system can replace the use of passwords and RF I-Cards for access to high security systems and buildings. We aim at making the cost effective and easy to use system, with high performance.

Paper12

Abstract—Automated driving system requires reliable, accurate detection and recognition of surrounding objects in real time environments. In this project, deep learning methods like Convolutional Neural Networks and computer vision are explored and applied to three major tasks of detection such as Pedestrians, Sign boards and Road conditions which concerns the development behind automated driving. This study goes one step further and addresses the problem of object detection using Deep Neural Networks, the approach is not only based on classifying but also precisely localizing objects of various classes. This project aims at presenting a simple and yet powerful formulation of object detection and then regressing bounding-boxes for pixels with a high object confidence score. Thus, this project aims at understanding deep learning methods and attempts to create a knowledge transition process where computer vision and learning representations are connected with a real-life application.

Paper13

Abstract— With the ever-increasing number of diseases in today's world, there is a need for a system to provide early diagnosis of human health. Ayurveda is a traditional medicine and natural healing system in India. Nadi-Nidan (pulse-based diagnosis) or Nadi Pariksha is an ancient medical technique, which can be traced back to ancient Indian and Chinese literature. Nadi-Nidan is a prominent method in Ayurveda, and is known to dictate all the salient features of a human body.  Wrist pulses or arterial pulses are sensed to diagnose about the health status. In this paper, we aim to design a non-invasive system for nadi pariksha to assist doctors in routine diagnostic procedures and provide details of our procedure for obtaining the complete spectrum of the nadi signals as a time series. We analyse the obtained time series signals for features, which provide more insight into this technique. An ayurvedic practitioners and physicians can use this prototype for pulse reading and analysis.

Paper14

Abstract: Brain tumor segmentation is a crucial step in surgical planning and treatment planning. A significant medical informatics task is indexing patient databases according to size, location, and other characteristics of brain tumors possibly based on magnetic resonance (MR) imagery. This requires segmenting tumors within images from different MR modalities. Automated brain tumor from MR modalities remains a challenging, computationally intensive task. This paper presents a study of K-means clustering approach for segmenting brain tumor from MRI images​.

Paper15

Abstract: As a smart city initiative by the government of India, digitalization of various retro techniques are being done. Water meter reading is one such. Image processing is a technique used to process the image and extract information from it. In this paper we propose a system to extract data from the image of the meter using image processing

Paper16

Abstract : This project is aimed at generating EMG (Electromyogram) signals through movement of muscles and analysis of electromyogram (EMG) signal processing and its application to identify human muscle strength of rehabilitation purpose. Single channel EMG signal was obtained from human muscle using non-invasive electrodes and further process by signal acquisition circuit to get a suitable signal to be process. In the first part of signal acquisition, the amplification circuit for the small EMG signal has been designed successfully using PSOC 5lp. After amplification stage EMG signal was digitized through analog to digital converter (ADC) then further process through Fast Fourier Transform (FFT) of the obtained EMG samples to convert it into frequency domain. Finally, the processed EMG signal was classified into 6 different levels in order to display the muscle strength level of the user. This EMG device can be used to help the weak person or an elderly to identity their strength level of muscle for clinical rehabilitation purpose.

Paper17
Abstract: Medical image segmentation plays an important role in diagnosis and various medical evaluations. Detection and segmentation of Brain tumor accurately is a challenging task. In the past, Different kinds of segmentation algorithms have been proposed for segmenting tumors in CT and MRI images individually. This paper proposes an approach for combining two multimodal images like CT and MRI containing tumor cell. Segmentation techniques like Edge detection, Thresholding, Watershed, K-means are implemented on fused MRI and CT images. Experiments show that K-means gives good segmentation results.