A novel skew estimation approach using radon transform

Skew is the angle the image makes with the corner of the page. Skew estimation is a preprocessing step in document image analysis. It involves two steps skew detection and skew correction. Skew estimation is needed because while scanning the document image a little skew is unavoidable. If skew is present in the document then Optical Character Recognizer (OCR) will identify the script wrongly. In this paper an effective and reliable skew estimation technique using radon transform is proposed. Radon transform will take O(N²M) time to execute for a document image of size N X N and with the skew of M. The proposed technique is tested on 150 document images with skew varying from 1 to 25 degrees. The accuracy of identification is 100% and average time of execution is 2.23seconds.