Title :
An original multi-scale algorithm to binarize images
Author :
Tabbone, S. ; Wendling, L.
Author_Institution :
LORIA-UMR, Univ. of Nancy, Vandoeuvre-les-Nancy, France
Abstract :
An original approach to binarize gray level images is proposed. A multi-scale algorithm based on a statistical test of homogeneity decides if a region belongs to the background or not. At each iteration the image is smoothed with a nonlinear filter in order to remove the noise. Stable regions in scale space are used to automatically find a threshold from the intensity histogram.
Keywords :
document image processing; image recognition; image segmentation; nonlinear filters; smoothing methods; bi-level thresholding; gray level image binarization; image smoothing; intensity histogram; multi-scale algorithm; noise removal; nonlinear filter; stable scale space regions; statistical test of homogeneity; threshold; Automatic testing; Classification algorithms; Focusing; Histograms; Image edge detection; Image segmentation; Laplace equations; Nonlinear filters; Signal processing algorithms; Statistical analysis;
Conference_Titel :
Pattern Recognition, 2002. Proceedings. 16th International Conference on
Print_ISBN :
0-7695-1695-X
DOI :
10.1109/ICPR.2002.1047461
