• R K SHARMA

      Articles written in Sadhana

    • Statistical feature extraction based iris recognition system

      ATUL BANSAL RAVINDER AGARWAL R K SHARMA

      More Details Abstract Fulltext PDF

      Iris recognition systems have been proposed by numerous researchers using different feature extraction techniques for accurate and reliable biometric authentication. In this paper, a statistical feature extraction technique based on correlation between adjacent pixels has been proposed and implemented. Hamming distance based metric has been used for matching. Performance of the proposed iris recognition system (IRS) has been measured by recording false acceptance rate (FAR) and false rejection rate (FRR) at differentthresholds in the distance metric. System performance has been evaluated by computing statistical features along two directions, namely, radial direction of circular iris region and angular direction extending from pupil tosclera. Experiments have also been conducted to study the effect of number of statistical parameters on FAR and FRR. Results obtained from the experiments based on different set of statistical features of iris images show thatthere is a significant improvement in equal error rate (EER) when number of statistical parameters for feature extraction is increased from three to six. Further, it has also been found that increasing radial/angular resolution,with normalization in place, improves EER for proposed iris recognition system

    • Recognition of online handwritten Gurmukhi characters based on zone and stroke identification

      KARUN VERMA R K SHARMA

      More Details Abstract Fulltext PDF

      Handwriting recognition is a technique that converts handwritten characters into a machine-processable format. Handwritten characters can either be presented to machine online or offline. A good amount of research in this area has been carried out for English, Chinese, Japanese and Korean languages. Research is also going on for Indian languages on developing online handwriting recognition systems. Headline and baseline are common features in most Indic languages which divide a character into three zones, namely, upper, middle andlower zones. Identification of headline and baseline is a major task for classification of strokes located in these three zones. A zone identification algorithm is proposed and tested in this text for online handwriting recognitionof Gurmukhi script. The strokes are grouped into these separate zones and are recognized based on respective support vector machine model for each zone. A rule-based approach has also been applied and tested for generation of characters from the set of recognized strokes. In this work, an accuracy of 95.3% has been achieved for zone identification and an accuracy of 74.8% has been achieved for character identification for Gurmukhi script. This accuracy has been achieved when the recognition engines of three zones were tested onthe dataset of 428 characters each written by 10 users.

    • Recognition of online unconstrained handwritten Gurmukhi characters based on Finite State Automata

      HARJEET SINGH R K SHARMA V P SINGH

      More Details Abstract Fulltext PDF

      This paper presents a language-based efficient post-processing algorithm for the recognition of online unconstrained handwritten Gurmukhi characters. A total of 93 stroke classes have been identified to recognize the Gurmukhi character set in this work. Support Vector Machine (SVM) classifier has been employedfor stroke classification. The main objective of this paper is to improve the character level recognition accuracy using an efficient Finite State Automata (FSA)-based formation of Gurmukhi characters algorithm. A databaseof 21,945 online handwritten Gurmukhi words is primarily collected in this experiment. After analysing the collected database, we have observed that a character can be written using one or more strokes. Therefore, a totalof 65,946 strokes have been annotated using the 93 identified stroke classes. Among these strokes, 15,069 stroke samples are considered for training the classifier. The proposed system achieved promising recognition accuracyof 97.3% for Gurmukhi characters, when tested with a new database of 8,200 characters, written by 20 different writers.

    • A novel framework for writer identification based on pre-segmented Gurmukhi characters

      MUNISH KUMAR M K JINDAL R K SHARMA SIMPEL RANI JINDAL

      More Details Abstract Fulltext PDF

      Handwriting is an obtained apparatus utilized for correspondence of one’s recognition or sentiments. Components that judge a person’s handwriting is not merely subject to the individual’s handwriting depends on the background, additionally considers like nervousness, inspiration and the reason for the handwriting. In spite of the high variation, in a man’s handwriting, recent outcomes from various writers have demonstrated that it has adequate individual quality to be utilized as an identification strategy. In this paper, theauthors are the pact with a novel approach to text dependent writer identification in view of pre-segmented Gurmukhi characters. The text dependent writer identification framework proposed in this paper includes distinctive stages like preprocessing, feature extraction, classification or identification. The feature extraction stage incorporates four schemes, zoning, diagonal, transitions and peak extent based features. To analyze the proposed framework execution, experiments are performed with two classifiers, namely, k-NN and SVM. SVM is also considered with linear-kernel in the present work. For experimental results, we have collected 31,500 samples from 90 different writers for 35 class problem. Maximum writer identification accuracy of 89.85% has been achieved by using a combination of zoning, transition and peak extent based features with Linear-SVMclassifier when we have taken 70% data as the training set and remaining 30% data as the testing set. Using 10-fold cross validation, we have achieved an accuracy of 94.76% with a combination of zoning, transition andpeak extent based features and Linear-SVM classifier.

  • Sadhana | News

    • Editorial Note on Continuous Article Publication

      Posted on July 25, 2019

      Click here for Editorial Note on CAP Mode

© 2017-2019 Indian Academy of Sciences, Bengaluru.