Design and Implementation of Offline Handwritten Gurmukhi Word Recognition System

Abstract

Handwritten word recognition system (HWRs) is one of the most emerging applications of pattern recognition which is a fully grown, advancing, energizing, and helpful field and is going about as an establishment stone for the emerging advanced applications. Handwritten Word recognition system in offline handwritten word is a technique that is used to convert the scanned images like handwritten text or pdf files into an editable format. Human reading can be replaced by HWRs and is one of the recent research topics. The conversion of digitized information into textual form becomes a challenging task but mandatory task to make it readable by any machine doing ventures of a million pages per second. When it comes to analyzing manuscripts that are not even sometimes in readable form and the text in those manuscripts has deteriorated, it becomes even more challenging to get that precious text and to make the masses of text available. newlineThe work carried out in this thesis mainly focuses on the various phases of the HWR with an inch of deep learning added for the recognition of Gurmukhi handwritten words related to tehsils and sub-tehsils of the Punjab. A lot of challenges faced in different phases of the HWR have also been discussed in detail. newline