Digital Image Processing

S. Raja Kumar ¹, V. Subbiah Bharathi ²

Affiliations
1 Sathyabama University, Chennai, IN
2 DMI College of Engineering, Chennai, IN

Abstract

Recognition of similar shaped ancient Tamil character is a difficult problem and in character recognition system most of the errors occur from similar shaped characters. Researchers have used many methods of feature extraction for ancient Tamil character recognition. In this paper we proposed a novel feature extraction technique to improve the recognition results of two similar shaped ancient Tamil characters. In this work, we have taken the samples of offline ancient Tamil characters from six different centuries. The technique is based on F-ratio (Fisher Ratio), a statistical measure defined by the ratio between-class variance and within-class variance. F-ratio modifies the feature vector of two similar shape characters by weighting the feature elements. This weighting scheme enhances the feature elements that belongs to the distinguishable portions of the similar shaped characters and reduces the feature elements of the common portion of the characters, so that similar shaped characters can be identified easily. We considered pair of similar shape ancient characters of different stone inscriptions and we noted that f-ratio based feature weighting shows better recognition results. The proposed system achieves a maximum recognition accuracy of 94% using gradient features and template matching method.

Keywords

Similar Shape Ancient Tamil Characters, Roberts Filter, F-Ratio.

Full Text

     

M. A. Leo Vijilious ¹, V. Subbiah Bharathi ²

Affiliations
1 Sathyabama University, Chennai, IN
2 DMI College of Engineering, Chennai, IN

Abstract

Biometrics plays an important role in personal identification and it is becoming increasingly popular today. Palmprint matching is considered in this paper for effective identification of persons. Palmprint matching is performed based on Zernike moments feature descriptors and the classification using Modified Adaboost classifiers. Since, Zernike moments have the property of geometrical invariance, they are superior in image representative capability. From the experimental results, it has been observed that Zernike moments achieve superior performance than the other well-known moments.

Keywords

Biometrics, Palmprint, Feature Extraction, Zernike Moments, Adaboost.

Full Text

     

J. P. Ananth ¹, V. Subbiah Bharathi ²

Affiliations
1 Sathyabama University, IN
2 DMI College of Engineering, IN

Abstract

Image Retrieval is a field of study concerned with searching and retrieving images from a collection of database. The user participation in image retrieval system gains attention in the recent research in order to reduce the impact of completely depending on discrimination power of image features. In this proposed work, interactive genetic algorithm is employed where user selects one of the retrieved images for the next stage of mutation. Moreover, dual Hahn moments employed in this work, which are orthogonal and rotation invariants are effective image descriptors. Experiments were carried out on COREL images and the average retrieval rate of 88% reveals the efficacy of the proposed work.

Keywords

Moment Features, Image Retrieval, Genetic Algorithm.

Full Text

     

J. P. Ananth ¹, V. Subbiah Bharathi ²

Affiliations
1 Sathyabama University, IN
2 DMI College of Engineering, IN

Abstract

Retrieval of images from large databases is an emerging application particularly in medical and forensic departments. Face image retrieval is still a challenging task since face images can vary considerably in terms of facial expressions, lighting conditions etc. In feature based image retrieval methods, the accuracy depends on the discrimination power of the features. In this work, orthogonal moments were employed as features for the retrieval task. Due to the orthogonal property, these moments are inherently non-redundant exhibiting good image representation capability. Racah moments defined in a non-uniform lattice are proved to be better than other orthogonal moments in terms of reconstruction error. Face image retrieval using Racah moment features has been extensively experimented with YALE face database and FERET Database. The results reveal the efficacy of orthogonal moment descriptors.

Keywords

Feature Selection, Image Retrieval, Moment Feature, Racah Moment.

Full Text