No abstract

No abstract

The necessity of recognizing handwritten characters is increasing day by day because of its various applications. The objective of this paper is to provide a sophisticated, effective and efficient way to recognize and classify Bangla handwritten characters. Here an extended convolutional neural network (CNN) model has been proposed to recognize Bangla handwritten characters. Our CNN model is tested on "BanglalLekha-Isolated" dataset where there are 10 classes for digits, 11 classes for vowels and 39 classes for consonants. Our model shows accuracy of recognition as: 99.50% for Bangla digits, 93.18% for vowels, 90.00% for consonants and 92.25% for combined classes.

Objective of this paper is to identify a person taking fingerprint as a biometric parameter using wavelet packet transform. Here both conventional discrete wavelet transform (DWT) and discrete wavelet packet transform (WPT) are used considering special basis function/matrix to extract the coefficients of basis functions those convey the most of the energy of the signal or image. Here top 5% coefficients are chosen which actually convey the characteristics of an image. The outcome of the paper is to determine the set of energetic coefficients of basis functions which carry the features of an image hence storage required to preserve the template of images will be reduced considerably.

No abstract