No abstract

No abstract

No abstract

Recognition of ancient Tamil hand written characters from inscriptions is difficult. Ancient Tamil characters are different from current century's Tamil character. This paper concentrates on the century identification of ancient Tamil characters and converting them into current century's form using MATLAB. In this paper, a method for recognizing Tamil characters from stone inscriptions, called the contour-let transform, which has been recently introduced, is adopted. From previous research works, it's noticed that Wavelet transforms are not capable of reconstructing curved images perfectly. The contour-let transform offers a solution to remedy to this insufficiency. Contour-let transform is a 3D approach technique whereas wavelet transform is a 2D technique. The characters from the input image are recognized through clustering mechanism. Further the noise present in the image is removed by fuzzy median filters. Neural networks are employed to train the image and compare the data with the current century's character. Hence a more accurate recognition of Ancient Tamil characters from stone inscriptions is obtained.

Liver cancer leads to more number of human deaths nowadays. Patient survival chances can be increased by early detection of the tumour. Texture analysis based on moment features for CT liver scan images is proposed here. The texture feature is extracted by local binary pattern and statistical features are extracted by Legendre moments. This communication presents a comparative analysis between these Legendre moments, local binary pattern and combined features. The classification accuracy of 96.17% is obtained for CT liver images. The experimental result shows that better texture classification is obtained using the proposed method.