Identification and Recognition of Printed Distorted Characters using DCR Method
In the past few years Technology advances in two areas and these are computers and communication networks.
This advancement has provided the opportunity to make literature and important information available to the people from all
around of the world. By developing image databases of these manuscripts and different documents and making them
available on the Internet, it will be easier for the people to access them. We can save the documents from further degradation
by preserving them and at the same time make them accessible in perpetuity in the digital library and it is another significant
advantage. For these reasons character recognition is an ever-ending research application in the real world and each character
recognisation should be accurate. So that it leads to understand the exact meaning and concept. Analyzing the distorted
character is quite complicated task. In some languages like Bengali, Hindi even English the distorted character may appear
like some other character so that important data may be missed. In this paper, we present an algorithm to improve the quality
of the distorted document images well as the recognition of the distorted texts. Experimental result shows that the accuracy
of our proposed technique is 97%.
Keywords - Distorted Document, Image Enhancement, Ocr, Dcr, Character Segmentation, Text Extraction.