Skip to main content
Conference ProceedingsOPEN ACCESS

Image thresholding of historical documents using entropy and ROC curves

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (2005) 3773 LNCS 905-916
DOI: 10.1007/11578079_93
5Citations
Citations of this article
10Readers
Mendeley users who have this article in their library.

Abstract

It is presented herein a new thresholding algorithm for images of historical documents. The algorithm provides high quality binary images using entropy information of the images to define a primary threshold value which is adjusted with the use of ROC curves. © Springer-Verlag Berlin Heidelberg 2005.

Cite

CITATION STYLE

APA

Mello, C. A. B., & Costa, A. H. M. (2005). Image thresholding of historical documents using entropy and ROC curves. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 3773 LNCS, pp. 905–916). https://doi.org/10.1007/11578079_93

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free