No abstract

No abstract

Huge quantity of information is lying quiescent in historical manuscripts. This information would go wasted if it is not stored digitally. In keyword spotting, all occurrences of a query keyword image are retrieved from scanned document images. The problem of spotting words from handwritten documents is difficult due to its huge changeability in writing styles and its large vocabulary. Existing keyword spotting approach is mainly based on statistical depiction of word image. This paper presents an efficient structural depiction of word image, where the handwritten words are represented using graph based method for historical handwritten devanagari manuscripts. Experimentation is conducted on historical handwritten Shankaracharya's documents written in Devanagari. The results were promising in terms of accuracy and efficiency.

Abstract:Translating text form one language to another by extracting text from an image will enhance the knowledge of society without any language barrier. Text extraction from image involves detecting the text from given image, finding the presence of text location, extraction, enhancement and recognition of text from the given image. Machine translation technique is used to translate source language to target language. Machine translation plays an important role in benefiting linguists, sociologists, computer scientists. The demand of translation has become more in recent years due to increase in the exchange of information between various regions using different regional languages. A large number of techniques have been proposed to address this problem. The purpose of this paper is to review and classify various text extraction algorithms and translation techniques.