26-06-2013, 12:09 PM
Binarization of Degraded Document images
Binarization of Degraded.pptx (Size: 802.51 KB / Downloads: 73)
Introduction
Separate the pixel values of the input grayscale image into two classes like background and foreground
Segmentation for object regions from background
It is usually performed in the preprocessing stage of different document image processing applications such as OCR, Text mining , word spotting etc
Various techniques
Global Thresholding
Local Thresholding
Adaptive Thresholding
Morphological Operations
Background Study
Moghaddam et al., Patt. Rec., vol. 43, pp. 2186–2198, 2010
It is a grid based multi-scale framework
It has the ability to improve the binarization results and to restore weak connections and strokes
Tabatabaei and Bohlool, ICIP, pp. 3573–3576, 2010
This method is based on background estimation by using morphological closing operation
Morphological closing operation is used to compensate for uneven background illumination
This method is independent of parameters