Foreground Estimation in a Degraded Text document
Ilavarasn Chandra, K Nagarjunavarma, G. Veera Senthil Kumar · International Journal of Computer Applications · 2012
In this paper an attempt is made to retrieve the text region alone from a degraded text document. For doing that, four different filters are used for noise removal in the text document. Later document binarization is done using thresholding. Three different thresholding techniques are implemented for foreground-background separation.Then candidate region is selected and features are extracted. The features are then fed to an SVM to classify text and non-text regions. The proposed approach is implemented and tested on various hand written and machine printed degraded text documents.