EFFICIENT COMPRESSION OF BINARIZED TAINTED DOCUMENTS
Dept. of CSE RGUKT, IIIT-AP, RK Valley, AP, India, Ratnakumari Challa · International Journal of Advanced Research in Computer Science · 2018
Tainted documents are degraded or ruined documents of low quality and of worn out look.The taintations are like disparity variation, smear, ooze, uneven illumination.To enhance the visual quality of the tainted document binarization technique is applicable.Binarization can binarize all the tainted documents but performing binarization to faultily tainted documents is a complicated task, the complication is observed in the identification of variations between the document background and text foreground.The system uses OTSU binarization that can binarize any kind of taintations.The proposed technique addresses the variations between background and foreground text of the document and calculates the optimum threshold separating the two classes so that their combined spread is minimal or equivalent hence the vision quality increases.Enhancement of vision quality also results in the enhancement of document size.Compression is performed on the binarized tainted document to reduce the tainted document size.The compression technique projected to use in this paper is Run Length coding which helps to reduce the size of the tainted document.Run Length coding is lossless compression technique which is very successful in dealing with binary images.