MIT-World Peace University (MIT-WPU) is a top educational institute located in Pune, India. It is recognized within&nbsp;UGC under Government of Maharashtra Act No. XXXV 2017.&nbsp;

It is part of the MIT group of institutions (established in 1983) which covers 10+ campuses across India and has more than 50,000+ students enrolled across various disciplines at a given point of time. The institute houses 15 schools ranging from Engineering &amp; Technology, Design, Pharmacy, Commerce, Governance among others. It offers undergraduate, postgraduate and PhD programs among different disciplines.&nbsp;

&nbsp;

MIT World Peace University

School of Computer Engineering & Technology

In Optical Character Recognition [OCR], machine printed or handwritten characters/numerals are recognized. There are plentiful approaches that deal with problem of detection of numerals/character depending on the sort of feature extracted and different way of extracting them. This paper presents a novel method for handwritten Marathi (Devnagari) numerals; most popular in Indian subcontinent. The proposed work focuses on feature extraction i.e. the global features; namely, Fourier descriptors and chain code of the binary images representing isolated Marathi numerals are extracted as features. These feature vectors are fed to SVM classifier that has been trained to recognize Marathi numerals. The system has been tested on varieties of image of isolated handwritten Marathi numerals. Experimentation result shows the recognition rate of 98.1%.

World University root node

Engineering and Technology

International Journal of Computational Intelligence Research

Isolated handwritten Marathi numerals recognition based upon Fourier descriptors and freeman chain code

International Journal of Computer Applications

Fourier Descriptor based Isolated Marathi Handwritten Numeral Recognition

CAPTCHA stands for Completely Automated Public Turing Tests to Tell Computers and Humans Apart.A CAPTCHA is a program that protects websites against bots -automated scripts by generating and grading tests that humans can pass but current computer programs cannot. For example, humans can read distorted text but current computer programs can't. The threats from bots is growing rapidly and also increasing in technical sophistication causing various types of attacks. These attacks are affecting social aspects of human beings. Human being or legitimate users are cheated by false requests disturbing their social life. This paper aims to provide secured authentification by verifying strength of CAPTCHA. The strength of CAPTCHA can be verified by solving it. The CAPTCHA solving has mainly three steps such as preprocessing, segmentation and character recognition. The input distorted image of CAPTCHA is passed through preprocessing module which removes noise and gives raw image which is then segmented. These segmented characters are recognized using pattern-matching technique achieving 88% accuracy. The result shows that this algorithm provides secured social relevance to avail free Internet services and in turn improve usability of well-claimed CAPTCHA. © 2010 IEEE.

Proceedings - 3rd International Conference on Emerging Trends in Engineering and Technology, ICETET 2010

Algorithm for secured online authentication using CAPTCHA

Digital Imaging and Communications in Medicine is a standard for handling, storing, printing, and transmitting information in medical imaging. The National Electrical Manufacturers Association holds the copyright to this standard. It was developed by the DICOM Standards committee. The other image viewers cannot collectively store the image details as well as the patient's information. So the image may get separated from the details, but DICOM file format stores the patient's information and the image details. Main objective is to develop a DICOM image viewer. The image viewer will open.dcm i.e. DICOM image file and also will have additional features such as zoom in, zoom out, black and white inverter, magnifier, blur, B/W inverter, horizontal and vertical flipping, sharpening, contrast, brightness and .gif converter are incorporated. © 2010 American Institute of Physics.

AIP Conference Proceedings

Image viewer using digital Imaging and Communications in Medicine (DICOM)

Data stored in databases keep growing as a result of businesses requirements for more information. A big portion of the cost of keeping large amounts of data is in the cost of disk systems, and the resources utilized in managing that data. This paper introduces various compression techniques for data stored in row oriented as well as column-oriented databases. Keeping data in this compressed format as it is operated upon has been shown to improve query performance by up to an order of magnitude. Intuitively, data stored in columns is more Compressible than data stored in rows. Compression algorithms perform better on data with low information entropy (high data value locality) i.e are used for optimization purpose. © 2010 IEEE.

ICCET 2010 - 2010 International Conference on Computer Engineering and Technology, Proceedings

Database compression techniques for performance optimization

This paper gives the improved method of generating a binary image IFS using Genetic Algorithm. To find the maps of IFSs that can encode black and white (BW) images, the Genetic Algorithm uses a variable-length genotype representation, i.e., each IFS is represented as a list of maps, and a map is represented as a set of real parameters. Special genetic operators that maintain and control the feasibility of the individuals in the population are adopted. A fitness function is defined that measures the similarity between the attractor and the image that penalizes a large number of maps and high contractivity factors. © 2010 IEEE.

ACE 2010 - 2010 International Conference on Advances in Computer Engineering

Fractal image compression using genetic algorithm

Voice Web, as the name suggests, accesses web resources via voice. VoiceXML is the markup language used to develop speech applications. VoiceXML is interactive and allows voice input to be received and processed by voice browser. Unfortunately, the navigation of VoiceXML document is completely controlled by application developer. Also, the user does not have flexibility to utter random word from currently executing dialog. The aim of the paper is to address the weakness of VoiceXML and develop an application, which dynamically detects recognition candidates in user content, in contrast with recognition candidates of the existing voice web, which depend on the application developer. In this application, a news service is implemented along with dictionary of IT-specific terms and dictionary of words from currently executing news. © 2010 American Institute of Physics.

A voice web application based on dynamic navigation of VXML

In this paper, we present a novel method for automatic recognition of isolated Marathi handwritten numerals. Chain code and Fourier Descriptors that capture the information about the shape of the numeral are used as features. After preprocessing the numeral image, the normalized chain code and the Fourier descriptors of the contour of the numeral are extracted. These features are then fed in the Support Vector Machine (SVM) for classification. The proposed method is experimented on a database of 12690 samples of Marathi handwritten numeral using fivefold cross validation technique. We have obtained recognition accuracy of 98.15%.

IJCA Special Issue on “Recent Trends in Image Processing and Pattern Recognition”

Marathi Handwritten Numeral Recognition using Fourier Descriptors and Normalized Chain Code

Recent research indicates that multimodal biometrics is the way forward for a highly reliable adoption of biometric identification systems in various applications, such as banks, businesses, government and even home environments. However, such systems would require large distributed datasets with multiple computational realms spanning organisational boundaries and individual privacies. In this paper, we propose a novel approach and architecture for multimodal biometrics that leverages the emerging grid information services and harnesses the capabilities of neural network as well. We describe how such a neuro-grid architecture is modelled with the prime objective of overcoming the barriers of biometric risks and privacy issues through flexible and adaptable multimodal biometric fusion schemes. On one hand, the model uses grid services to promote and simplify the shared and distributed resource management of multimodal biometrics, and on the other hand, it adopts a feed-forward neural network to provide reliability and risk-based flexibility in feature extraction and multimodal fusion, that are warranted for different real-life applications. With individual autonomy, scalability, risk-based deployment and interoperability serving the backbone of the neuro-grid information service, our novel architecture would deliver seamless and robust access to geographically distributed biometric data centres that cater to the current and future diverse multimodal requirements of various day-to-day biometric transactions. © Springer Science+Business Media B.V. 2010.

Innovations in Computing Sciences and Software Engineering

Risk-based neuro-grid architecture for multimodal biometrics

Learning relies on the acquisition of different types of knowledge supported by perceived information. It leads to the development of new capacities, skills, values, understanding, and preferences. Its goal is the increasing of individual and group experience. The paper starts with introduction of learning. Semi-supervised learning addresses the problem of labeled data by using large amount of unlabeled data, together with the labeled data, to build better classifiers. Because semi-supervised learning requires less human effort and gives higher accuracy, it is of great interest both in theory and in practice. The paper discusses various important approaches to semi-supervised learning such as self-training, co-training(CO). expectation maximization (EM) .CO-EM. Then the use of graph-based methods is explained. Finally the use of combination of support vector machines (SVM) and K-means clustering method is explained. The paper concludes with the proposed adaptive learning methodology. Copyright © 2009 by IICAI.