Survey on Restoration of Degraded Document using Different Thresholding Techniques
Journal Title: International Journal of Engineering Sciences & Research Technology - Year 30, Vol 3, Issue 4
Abstract
Documents are degraded normally by the disturbance caused in the background . These are known as bleed-through effect, (ie) a pattern that interferes with the main text due to seeping of ink from the reverse side. These degradations must be removed to improve human or automatic readability. Degraded documents will have background noise ,changes in contrast and illumination ,so it is best to use local thresholding methods to restore foreground. Comparative study on some local adaptive thresholding methods such as niblack, sauvola, wolf and feng are done and analyse its quality over different images. The quality of the image is measured in terms of PSNR and correlation.
Authors and Affiliations
Jensiya Asirvatham*
Keywords
Related Articles
- EP ID EP95391
- DOI -
- Views 56
- Downloads 0
Static Structural Analysis of Rotary Weeding Blades using CAD Software
Rotary weeding blade is one of the important component for a tractor operated rotary weeder and need to be designed in such a way that they have enough strength in order to avoid the unbalancing force on the rotar...
EMOTIONAL INTELLIGENCE AND ITS IMPACT TO THE PERFORMANCE OF MARITIME DECK AND ENGINE CADETS IN THE SHIPBOARD TRAINING OF NAVAL STATE UNIVERSITY, NAVAL, BILIRAN PROVINCE
The study assessed on the Emotional Intelligence and its Impact to the Performance of Deck and Engine Cadets of the NSU-CME who underwent shipboard training. It examined the two programs: the Bachelor of Science in...
High Order Differential Covariance based Source Separation of Monkey’s fMRI Data
Blind source separation is a statistical technique in which an observed mixed data is decomposed in source signals and mixing channel. In BSS normally no, or little knowledge about the mixing channel is available...
FAULT DEVELOPMENT PROCESS IN ELECTRICAL DRIVES: A REVIEW
This work studied the use of neural networks for model based fault diagnostics of induction motors. The idea was to use data provided by accurate FEM simulations of motor operation under different fault and load c...
COMPARATIVE ANALYSIS OF FILTERS FOR TEXT EXTRACTION FROM NOISY IMAGES
With rapid growth of the internet the amount of image and video data is increasing exponentially. The text data present in images and videos is useful for automatic annotations, indexing and structuring of images....