Binarization and Character Recognition of Degraded Printed Documents
Anju Gopi, Manju Thomas · 2015
Abstract-In the process of improving degraded printed document images, segmentation is one of the challenging task due to various document degradations. Here we proposes a new approach for the enhancement of degraded document image. It consists of an adaptive image contrast for document image binarization .Adaptive image Contrast can be constructed using local image gradient and local image contrast. Further the text stroke edge pixels are detected through the combination of binarized adaptive contrast map and canny's edge map .The text within the document is then segmented by a local thresholding technique. Then a post processing is applied to improve the quality of degraded images. It works for different types of degraded document images. Then an artificial neural network is used for optical character recognition purpose.