KINJAL MISTREE | Uka Tarsadiya University (original) (raw)

Papers by KINJAL MISTREE

International Journal of Intelligent Computing and Cybernetics, Dec 16, 2022

International Journal of Information Technologies and Systems Approach

Sign language recognition and translation is a crucial step towards improving communication betwe... more Sign language recognition and translation is a crucial step towards improving communication between the deaf and the rest of the society. According to the Indian Sign Language Research and Training Centre (ISLRTC), India has around 300 certified human interpreters. With such a shortage of human interpreters, an alternative service is desired that helps people to achieve smooth communicate with deaf. In this study, an approach is presented that translates ISL sentences in English text using MobileNetV2 model and neural machine translation (NMT). The system features ISL corpus created from Brown corpus using ISL grammar rules. The approach converts the ISL videos into ISL gloss sequence using MobileNetV2 model and recognised ISL gloss sequence is then fed to machine translation module. MobileNetV2 was proven best-suited model for recognition of ISL sentences and NMT gives better result than statistical machine translation (SMT) to convert ISL gloss sequence into English text. The auto...

The purpose of this paper is to find the percentage purity of rice grain sample by image processi... more The purpose of this paper is to find the percentage purity of rice grain sample by image processing technique. The most common test is done on the basis of size of the rice grain. Traditionally the grain types and their quality were promptly measured through visual assessment by human inspectors, whose decisions are subjected to exterior influences such as tiredness and is also time consuming. Machine vision is an alternative to an automated, cost-effective and non-destructive technique. This method proposes a method for rice seeds using image processing. This technique begins by acquiring images of rice samples which are converted to gray image and then to binary image by using thresholding. The rice grains area is measured using regionprops property of Matlab. KeywordsImage Processing; Quality Grading; Thresholding; Regionprops; Matlab.

2017 International Conference on Innovations in Information, Embedded and Communication Systems (ICIIECS), 2017

Crowd analysis becomes very popular research topic in the area of computer vision. A growing requ... more Crowd analysis becomes very popular research topic in the area of computer vision. A growing requirement for smarter video surveillance of private and public space using intelligent vision systems which can differentiate what is semantically important in the direction of the human observer as normal behaviors and abnormal behaviors. People counting, people tracking and crowd behavior analysis are different stages for computer based crowd analysis algorithm. This paper focus on crowd behavior analysis which can detect normal behavior or abnormal behavior.

2016 International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET), 2016

Human age estimation is very interesting challenge today also. The prediction of age based on dif... more Human age estimation is very interesting challenge today also. The prediction of age based on different techniques and by human, itself is very similar now. The major challenges for machine based age estimation are still alive. Challenges are uncontrolled environment, different light settings, variant posing and expression, ageing effects and life style. Human age estimation started with craniofacial growth has now reached to the condition where challenges are changing into opportunities for developing of applications. Applications include Biometrics, forensic investigation during crime, Relationship management over e-commerce, security control, Monitoring and surveillance, entertainment, cosmetology. Computer-based age estimation via facial images includes stages such as pre-processing, feature extraction, feature reduction and lastly classification or regression. Pre-processing provides invariability against different light settings and environment. Feature extraction deals with extraction of features like facial landmarks, shape and texture for further analysis. Feature reduction is considered for faster analysis of feature vectors. Lastly, regression, classification or combination is used for estimation of the age.

2015 International Conference on Information Processing (ICIP), 2015

Real time object tracking is the process of locating moving objects over time using the camera in... more Real time object tracking is the process of locating moving objects over time using the camera in video sequences in real time. The objective of object tracking is to associate target objects in consecutive video frames. Object tracking requires location and shape or features of objects in the video frames. Every tracking algorithm needs to detect moving object. So object detection is the preceding step of object tracking in computer vision applications. After that, detected object can be extracted by the feature of moving object to track that moving object into video scene. It is challenging task in image processing to track the objects into consecutive frames. Various challenges can arise due to complex object motion, irregular shape of object, occlusion of object to object and object to scene and real time processing requirements. Object tracking has a variety of uses, some of which are: surveillance and security, traffic monitoring, video communication, robot vision and animation.

This Image Mosaicing is a process of assembling multiple overlapping images of the same scene int... more This Image Mosaicing is a process of assembling multiple overlapping images of the same scene into a large image. The output of the image Mosaicing operation is the union of the two or more overlapping input images. This technology is widely used in photography, digital video, motion analysis, medical image processing, remote sensing image processing, document image processing and other fields. This paper describes a combined approach of Harris-SIFT feature detection for Image Mosaicing. Firstly, feature points are detected by using Harris corner detector, then after SIFT descriptor is computed to store feature vector for each detected keypoints and then feature matching is applied. The RANSAC Homography algorithm is used to detect wrong matches for improving the stability of the algorithm and for estimating the transformation model. Finally Image Blending procedure is used for smooth and seamless panorama. Combined approach of Harris corner detector and stable invariant SIFT descriptor is reliable in extracting the translation, scale and rotation invariant features, less affected by illumination issues and increases the speed of the feature detection process which also affects the matching accuracy.

2015 International Conference on Electrical, Electronics, Signals, Communication and Optimization (EESCO), 2015

Object tracking is the process of locating moving objects over time using the camera in video seq... more Object tracking is the process of locating moving objects over time using the camera in video sequences. The objective of object tracking is to associate target objects in consecutive video frames. Object tracking requires location and shape or features of objects in the video frames. So, object detection and object classification is the preceding steps of object tracking in computer vision application. To detect or locate the moving object in frame, Object detection is first stage in tracking. After that, detected object can be classified as vehicles, human, swaying tree, birds and other moving objects. It is challenging or difficult task in the image processing to track the objects into consecutive frames. Various challenges can arise due to complex object motion, irregular shape of object, occlusion of object to object and object to scene and real time processing requirements. Object tracking has a variety of uses, some of which are: surveillance and security, traffic monitoring, video communication, robot vision and animation. This paper presents the various techniques of object tracking in video sequences through different phases using image processing.

Classification is vital for the evaluation of agricultural produce. However, the high costs, subj... more Classification is vital for the evaluation of agricultural produce. However, the high costs, subjectivity, tediousness and inconsistency associated with manual sorting have been forcing the post harvest industry to apply automation in sorting operations. Fuzzy logic (FL) was applied as a decision making support to grade apples in this study. Quality features such as the color, size and defects of apples were measured through different equipment. The same set of apples was graded by both a human expert and a FL system designed for this purpose. Grading results obtained from FL showed 89% general agreement with the results from the human expert, providing good flexibility in reflecting the expert's expectations and grading standards into the results. This application of apple grading can be fully automated by measuring the required features by means of high-tech sensors or machine vision and making the grading decision using FL.

bvmengineering.ac.in

... [3] Fraser, C., Hanley, H., “Bias Compensation in rational functions for IKONOS satellite ima... more ... [3] Fraser, C., Hanley, H., “Bias Compensation in rational functions for IKONOS satellite imagery", PE & RS, 2003, pp 53-58 [4] Sanjay Singh, Devakanth Naidu, TP Srinivasan, B. Gopala Krishna, PK Srivastava, “Rational Polynomial Modelling For Cartosat-1 Data”, ISPRS, 2006 ...

Sign language is a complete natural language used by deaf and dumb people. It has its own grammar... more Sign language is a complete natural language used by deaf and dumb people. It has its own grammar and it differs with spoken language to a great extent. Since people without hearing and speech impairment lack the knowledge of the sign language, the deaf and dumb people find it difficult to communicate with them. The conception of system that would be able to translate the sign language into text would facilitate understanding of sign language without human interpreter. This paper describes a systematic approach that takes Indian Sign Language (ISL) video as input and converts it into text using frame sequence generator and image augmentation techniques. By incorporating these two concepts, we have increased dataset size and reduced overfitting. It is demonstrated that using simple image manipulation techniques and batch of shifted frames of videos, performance of sign language recognition can be significantly improved. Approach described in this paper achieves 99.57% accuracy on the...

Image enhancement is an important processing task in image processing field. By applying image en... more Image enhancement is an important processing task in image processing field. By applying image enhancement, blur or any type of noise in the image can be removed so that the resultant image quality is better. Image enhancement is used in various fields like medical diagnosis, remote sensing, agriculture, geology, oceanography. There are numbers of techniques for image enhancement. Image interpolation is used to do enhancement of any image. This paper gives overview about different interpolation techniques like nearest neighbor, bilinear, bicubic, new edge-directed interpolation (NEDI), data dependent triangulation (DDT), and iterative curvature-based interpolation (ICBI).

2019 IEEE 5th International Conference for Convergence in Technology (I2CT), 2019

Text summarization is the process to create short and concise summaries from the text document wi... more Text summarization is the process to create short and concise summaries from the text document with the aim of provide most important or salient content in a condensed form from the source document. As the information is overloaded on the web, text summarization becomes necessary. Extractive and abstractive are two broad approaches of text summarization. Extractive method assembles summaries from the document, which directly take a sentence or phrase from the original document while abstractive method may generate new words to make a summary. This paper gives the comparison of various neural network based abstractive text summarization models and also discuss the types of summarization based on categories and different approaches of abstractive text summarization. At last, comparison of these models are also given.

Communications in Computer and Information Science, 2021

International Journal of Advanced Computer Science and Applications

Sign language, a language used by Deaf community, is a fully visual language with its own grammar... more Sign language, a language used by Deaf community, is a fully visual language with its own grammar. The Deaf people find it very difficult to express their feelings to the other people, since the other people lack the knowledge of the sign language used by the Deaf community. Due to the differences in vocabulary and grammar of the sign languages, complete adoption of methods used for other international sign languages is not possible for Indian Sign Language (ISL) recognition. It is difficult to handle continuous sign language sentence recognition and translation into text as no large video dataset for ISL sentences is available. INSIGNVID-the first Indian Sign Language video dataset has been proposed and with this dataset as input, a novel approach is presented that converts video of ISL sentence in appropriate English sentence using transfer learning. The proposed approach gives promising results on our dataset with MobilNetV2 as pretrained model.

IJETAE, 2013