Vellore Institute of Technology (VIT) is a private university located in&nbsp;Tamil Nadu, India. Founded in 1984, as Vellore Engineering College, the institution offers 20 undergraduate, 34 postgraduate, four integrated and four research programs. It has campuses in Vellore, Amravati, Bhopal and Chennai.

VIT is one of the top ranked private universities in India according to NIRF, THE and QS Rankings.&nbsp;Govt. of India has recognized&nbsp;VIT, Vellore as an&nbsp;Institution of Eminence. This has allowed VIT to take independent quality initiatives and move up in world ranking.

&nbsp;

&nbsp;

VIT University

This paper discusses relationship extraction among actors/nodes in the text provided. Given a text data, relationships are extracted using natural language processing and shown in a graph. Proper identification of “Nouns” and “Pronouns” can help to identify the actors/characters in the given text. Any relation usually would be between/among the nouns, so this paper first aims to extract “Nouns” and “Pronouns” in the given text and discusses how relation is extracted between/among them. Paper discusses step by step procedure of finding relation among actors/characters in the given plot/text. There are two prominent phases; one is, extracting “Nouns” and “Pronouns,” and the other is the relation among the characters/actors. This methodology is applied on movie story plot and tested. The above said methodology is applied on movie “Baahubali,” as it contains many characters and relations. © 2020, Springer Nature Singapore Pte Ltd.

World University root node

Department of Information Security

School of Computer Science and Engineering

Vellore Campus

Lecture Notes in Networks and Systems

Relation Extraction and Visualization Using Natural Language Processing

Comprehending the environment accurately and proficiently is one of the fundamental undertakings for the visually impaired. Many electronic travel aid devices are utilizing stereo vision techniques. However, there is no existing technology that helps to fulfill the requirements of the visually impaired and be economical at the same time. All the current technologies are implemented only for a specific purpose and have some constraints. In this paper, we propose a new framework that integrates the critical aspects of stereo vision with some additional features for aiding the visually impaired to solve their existing problems efficiently. In this system, both stereo vision and ultrasonic sensors have been integrated to get a precise awareness of the obstacles present in the real-time surroundings. The central concept is to categorize point vertices for the detection of objects in three-dimensional space according to height, width and slope of the adjacent vertices present in the images captured by CCD cameras. The intended assistive system serves to be more cost-effective for the visually impaired with the usage of CCD cameras and low-budget sensors. The assistive system in its recognition phase employs the proposed obstacle detection algorithm using stereo vision to recognize and track impediments that are at a farther distance. While, it employs ultrasonic sensors to detect obstacles present at shorter distances. The system also overcomes the problem of finding holes and descending stairs, faced by devices that use monoscopic vision. Additional features, like the Buzzer, GPS, Voice module, Alert message, are being incorporated. The principle intention of this paper is to contribute our knowledge to assist the visually impaired to become independent and help them to adapt to various daily situations.

Fulltext

SN Applied Sciences

Obstacle avoidance using stereo vision and depth maps for visual aid devices

Security has become an important factor. Intruders have become prominent factors for all the data/property theft. The basic idea in this paper is to identify the intruder and alert owner/administrator in different possible ways. This paper discusses different ways such as “a message (SMS)”, “WhatsApp message”, “location of intruder”, “an immediate call”, and “intruder’s image to owner’s/administrator’s WhatsApp” to alert owner/administrator. For identifying the intruder, machine learning algorithm is used. A camera placed at the locality is trained such that it can identify the familiar people and it is “on” all the time. Whenever an unknown/unidentified person comes to the vicinity of the camera, all the above-said features get activated and the owner gets alerted. The idea can be applied in many real-life situations, like thief identification near the house.

VIT Unitversity

Innovations in Computer Science and Engineering Lecture Notes in Networks and Systems

Machine Learning Technique for Automatic Intruder Identification and Alerting

Investigation of Geometrical Properties of Kernels Belonging to Seeds

Web-based games and simulations are becoming popular because they are easily accessible and are mostly hardware independent. This gives scope for improvements for the field of game development from a Web perspective. The Game Engine of this paper is proposed and developed to meet the increasing needs of a more straightforward way to develop Web-based 3D games and other visualizations. The Game Engine of this paper is made using certain existing tools such as three.js. The Game Engine gives flexibility to users by combining physi.js objects and three.js objects and introduces new classes of objects and also supports particle systems providing good control to developers without compromising performance. The Engine also features a HTML5-based editor that lets developers to give more prominence to game logic. This also streamlines the development process by splitting up the tasks among various streams of development.

WebGL-Based Game Engine

One dimensional nanostructured branched wires of tin dioxide (SnO2) were prepared in an alumina boat by a self catalytic method at different temperatures in a flow of argon and oxygen. The SEM images exhibit the different morphologies of the nanowires with an i.d. of around 40 nm, and presence of core/cluster shell nanowires and of nano/microbelt structures. XRD studies show the product to possess a typical tin dioxide nanowire structure. FTIR and Raman studies revealed the bond information. The sensing properties were studied by monitoring the changes in conductivity on exposure to hydrogen gas. Hydrogen can be detected in the 1000 ppm to 3000 ppm concentration range, and the limit of detection for H2 is 1000 ppm. [Figure not available: see fulltext.]. © 2017, Springer-Verlag Wien.

Department of Bio-Sciences

School of Bio Sciences and Technology

Centre for Nanotechnology Research

Microchimica Acta

Overhead projector sheets as substrate for deposition of one-dimensional tin dioxide nanostructures for use as a chemoresistive sensor for hydrogen

The aim of the paper is to automate the processing of gaze tracking data through soft clustering techniques. Standard analysis software for eye gaze tracking data requires users to define areas of interest, which may not be best option for exploratory analysis, where users may want to analyze eye gaze tracking data to know the area of interest. We have presented results on using Fuzzy c-means and Expectation Maximization algorithms on gaze tracking data and using an entropy based cluster validation index, we tried to automate identification of areas of interest. In our study, data from search task in digitally rendered 2D architectural plans have been explored and results indicated that irrespective of clustering technique, users fixated attention only 2 or 3 times for individual image. We have also presented GUI of a tool that can automatically identify areas of interest for any gaze tracking data sample using FCM or EM Algorithms.

2017 International Conference on Intelligent Computing, Instrumentation and Control Technologies (ICICICT)

Automating the process of gaze tracking data using soft clustering

One of the fears in the field of communication is insecurity of data. Pseudo random noise (PRN), which is similar to noise and also fulfills a greater number of the standard tests for statistical randomness, can be used as a key for encryption. Using Legendre PN sequence which is based on prime numbers, an image encryption technique has been proposed in this paper. Based on the number of pixels in the image, Legendre sequences are generated for a set of prime numbers and concatenated to generate the required length sequence. This sequence is then used to encrypt the secret image that generates noise like pattern. This encrypted image is later embedded into a larger image using DCT technique to have double security. The proposed algorithm using Legendre sequence when tested on various images has always produced a PSNR value of cover image above 20, proving that changes made to cover image is very less. Identification of key to decrypt the original image is impossible to detect as multiple Legendre sequences are used to form PN sequence. Steganography with DCT adds to improvement in security. The above algorithm can be extended to color images and the robustness of the algorithm can be tested using different attacking methods.

2017 International Conference on Recent Advances in Electronics and Communication Technology (ICRAECT)

Image Encryption Using Legendre Sequence and DCT

Separating tumor in MRI(Magnetic Resonance Image) brain images is a stimulating task,since there are variation of tumor shape,tumorsize and tumor location on the brain. The detection of tumor is still developing in medical field. This study targeted to extract the brain tumor from the brain images and execute the exact tumor location and its size. MRI scans images are widely used in medical research and also used to detect abnormalities in our body especially tumors. In this work,the process of image segmentation is performed by Morphological operation and threshold operation. Finally the result of the segmentation is tumor region and the size of the tumor. © 2016,International Journal of Pharmacy and Technology. All rights reserved.

International Journal of Pharmacy and Technology

Brain tumor detection in MRI brain imagesusing threshold operation

Nowadays, the image fusion technique plays very important role in these world and it is used for various fields such as medical imaging applications, battle field applications, satellite imaging applications and military applications. The image fusion is the process of combining two or more images with relevant information to produce a single image, which gives more information compare to the original source images. Here, the foreground image and background image are taken as an input source images. The foreground image is MRI brain image i.e functional parts and the background image is CT brain image i.e anatomical parts. Due to this, the information of the MRI i.e functional parts as well as CT i.e structural part of that source image combine to gives better views in fused image. The histogram equalization is the method, which is used to improve the contrast of the input source images. Based on this, to remove the noise in the fused image, the adaptive filter and median filter is used. Finally, the edge detection algorithm is used to detect the edge of the final output image. © 2014 by CESER PUBLICATIONS.

International Journal of Tomography and Simulation

Enhancement of MRI brain and CT brain image fusion by adaptive filter

The segmentation of brain tumors in magnetic resonance imaging (MRI) is a challenging and difficult task because of the variety of their possible shapes, locations, image intensities. In this paper the part of the image that has the tumor has more intensity in that portion and we can make our assumptions about the radius of the tumor in the image, these are the basic things considered in the algorithm. First of all some image enhancement and noise reduction techniques are used to enhance the image quality, after that some morphological operations are applied to detect the tumor in the image. The morphological operations are basically applied on some assumptions about the size and shape of the tumor and in the end the tumor is mapped onto the original gray scale image with 255 intensity to make visible the tumor in the image. The algorithm has been tried on a number of different images from different angles and has always given the correct desired result. © Research India Publications.

International Journal of Applied Engineering Research

Detection of brain tumor from MRI scan using segmentation methods

The field of medical imaging is gaining importance with an increase in the demand for automated, reliable, fast and efficient diagnosis. breast cancer accounts for 22.9% of illnesses amongst women worldwide. identification plays an important part in the diagnosis of benign and malignant tumors. a prime reason behind an increase in the number of cancer patients worldwide is the ignorance towards treatment of a tumor in its early stages. this project is initiated with basic pre-processing steps using filtering, contrast adjustment and stretching techniques. these techniques aid in noise removal,thus obtaining a well-defined tumor to work upon. in order to clearly demarcate the tumor from the rest of the original image, several methods have been applied, which include image subtraction and extending the minima function of the image. this is then followed by filling in holes near the tumor and applying segmentation techniques to the image. therefore, the final image displays the tumor, encircled within a border, against the original image.

International Journal of Engineering and Technology

Detection of tumor in mammogram images using extended local minima threshold

to design an image transformation system is Depending on the transform chosen, the input and output images may appear entirely different and have different interpretations. Image Transformation with the help of certain module like input image, image cluster index, object in cluster and color index transformation of image. K-means clustering algorithm is used to cluster the image for better segmentation. In the proposed method parallel saliency algorithm with K-means clustering is used to avoid local minima and to find the saliency map. The region behind that of using parallel saliency algorithm is proved to be more than exiting saliency algorithm.

Department of Analytics

Image transformation using modified K-means clustering algorithm for parallel saliency map

Medical Image Processing is a complex and challenging field nowadays. Processing of MRI images is one of the parts of this field. This paper proposes a strategy for efficient detection of a brain tumor in MRI brain images. The methodology consists of the following steps: preprocessing by using sharpening and median filters, enhancement of image is performed by histogram equalization, segmentation of the image is performed by thresholding. This approach is then followed by the further application of morphological operations. Finally the tumor region can be obtained by using the technique of image subtraction.

Centre for Bio-Separation and Technology

2012 IEEE International Conference on Computational Intelligence and Computing Research

Tumor detection using threshold operation in MRI brain images

Image fusion is the process of combining relevant information from two or more images into a single image. The resulting image will be more informative than any of the input images to retaining the important features of each image. The resulting image contains more information as compared to individual images. Multiple image fusion is an important technique used in aerial and satellite imaging, medical imaging, robot vision, digital camera application and battle field monitoring. The histogram equalization method is useful in images with backgrounds and foregrounds that are both bright or both dark. In particular, the method can lead to better views of functional and structural parts in MRI and CT images. Based on this, the removal of noise in the image by using adaptive filtering, the system is evaluated and the accurate output will be produced. The fusion algorithm of foreground images and background images is studied. Here, the foreground image is MRI and background image is CT image.