A Novel Algorithm for Efficient Detection and Skew Correction of Document Images with Clustered Background

In this paper, a new approach is presented for the study and implementation of document image detection and skew estimation. Image of text document can be acquired via camera may cause sometime a skew angle or tilt in document image. Many techniques were developed for skew detection. Improved optics and smaller size of digital camera has resulted in lot more document images with diverse background and skew. To find skew angle of document image on different background, first background has to be detected. For that a novel histogram based adaptive thresholding method is used.  The skew angle used in database with the range of ±90º w.r.t x-axis.