A Threshlod Selection Method Based on Multiscale and Graylevel Co-occurrence Matrix Analysis

Authors:
Yun Li;Ching Y. Suen;Mohamed Cheriet
Affiliations:
Concordia University,Montreal,Canada;Concordia University,Montreal,Canada;University of Quebec,Canada
Venue:
ICDAR '05 Proceedings of the Eighth International Conference on Document Analysis and Recognition
Year:
2005

Citing 8
Cited 2

Uniqueness of the Gaussian Kernel for Scale-Space Filtering

IEEE Transactions on Pattern Analysis and Machine Intelligence
Minimum error thresholding

Pattern Recognition
A note on the use of graylevel co-occurence matrix in threshold selection

Signal Processing
A new method for image segmentation

Computer Vision, Graphics, and Image Processing
Gray Level Thresholding in Badly Illuminated Images

IEEE Transactions on Pattern Analysis and Machine Intelligence
Evaluation of Binarization Methods for Document Images

IEEE Transactions on Pattern Analysis and Machine Intelligence
An Introduction to Digital Image Processing

An Introduction to Digital Image Processing
Goal-Directed Evaluation of Binarization Methods

IEEE Transactions on Pattern Analysis and Machine Intelligence

Transition pixel: A concept for binarization based on edge detection and gray-intensity histograms

Pattern Recognition
A new binarization method for non-uniform illuminated document images

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

Noises and complex backgrounds often make the thresholding of degraded document images a difficult job. In this paper, we propose a new threshold selection method to handle severely degraded document images. First, multiscale image description is adopted to analyze the image edge. From it, edge pixel pair information is derived and recorded by a graylevel co-occurrence matrix. An appropriate threshold value is chosen by measuring the edge pixel pair co-occurrence matrix. The new method is tested with degraded document images. The experimental results show it is resistant to noises and complex backgrounds.