Binarization of Badly Illuminated Document Images through Shading Estimation and Compensation

Authors:
S. J. Lu;C. L. Tan
Affiliations:
National University of Singapore, Kent Ridge, 117543, Singapore;National University of Singapore, Kent Ridge, 117543, Singapore
Venue:
ICDAR '07 Proceedings of the Ninth International Conference on Document Analysis and Recognition - Volume 01
Year:
2007

Citing 0
Cited 7

Illumination Compensation for Document Images Using Local-Global Block Analysis

IVIC '09 Proceedings of the 1st International Visual Informatics Conference on Visual Informatics: Bridging Research and Practice
Transition pixel: A concept for binarization based on edge detection and gray-intensity histograms

Pattern Recognition
Binarization of historical document images using the local maximum and minimum

DAS '10 Proceedings of the 9th IAPR International Workshop on Document Analysis Systems
Markov random field based binarization for hand-held devices captured document images

Proceedings of the Seventh Indian Conference on Computer Vision, Graphics and Image Processing
Combining statistical and geometrical classifiers for text extraction in multispectral document images

Proceedings of the 2011 Workshop on Historical Document Imaging and Processing
AdOtsu: An adaptive and parameterless generalization of Otsu's method for document image binarization

Pattern Recognition
Efficient illumination compensation techniques for text images

Digital Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a document image binarization tech- nique that segments text from badly illuminated document images. Based on the observations that text documents nor- mally lie over a planar or smoothly curved surface and have a uniformly colored background, badly illuminated docu- ment images are binarized by using a smoothing polynomial surface, which estimates the shading variation and com- pensates the shading degradation based on the estimated shading variation. Badly illuminated document images are accordingly binarized through the global thresholding of the compensated document images. Compared with the re- ported methods, the proposed technique is tolerant to the variations in text size and document contrast. At the same time, it is much faster and able to produce a binary text im- age with little background noise.