Optical Character Recognition from Degraded Document Images

Segmentation of the text from badly degraded document images is very challenging tasks due to the high inter/intra variation between the document background and the foreground text of different types of document images. In this paper, a novel document image binarization technique is used to addresses the issues in the degraded document images by using adaptive image contrast. The adaptive image contrast is a combination of the local image contrast and the local image gradient that is tolerant to text and background variations caused by different types of document degradations. The adaptive contrast map is first constructed for an input degraded document image. Then the contrast map is then binarized and combined with the Canny’s edge map to identify the text stroke edge pixels. The document text is further segmented by a local threshold that is estimated based on the intensities of detected text stroke edge pixels within a local window. Then apply the vertical scanning to find how many lines in the binary document image. After then apply the horizontal scanning to find how many characters in the image. Then the character is recognized using discrete wavelet transform.