This paper presents an investigation of the entropy of the Telugu script. Since this script is syllabic, and not alphabetic, the computation of entropy is somewhat complicated.

An end-to-end method for multi-language scene text localization, recognition and script identification is proposed. The approach is based on a set of convolutional neural nets. The method, called E2E-MLT, achieves state-of-theart performance for both joint localization and script identification in natural images and in cropped word script identification. E2E-MLT is the first published multi-lan...

The focus of this paper is Web accessibility for disabled people. Much of the Web remains inaccessible or difficult to access by people across a spectrum of disabilities and this may have serious implications for the potential use of the Web for increasing social inclusion. The topic of disabled Web access is introduced through a consideration of four discourses: digital divide, social construc...