Adaptive thresholding methods for documents image binarization

  • Authors:
  • Bilal Bataineh;Siti N. H. S. Abdullah;K. Omar;M. Faidzul

  • Affiliations:
  • Center for Artificial Intelligence Technology, Universiti Kebangsaan Malaysia, Bangi, Selangor, Malaysia;Center for Artificial Intelligence Technology, Universiti Kebangsaan Malaysia, Bangi, Selangor, Malaysia;Center for Artificial Intelligence Technology, Universiti Kebangsaan Malaysia, Bangi, Selangor, Malaysia;Center for Artificial Intelligence Technology, Universiti Kebangsaan Malaysia, Bangi, Selangor, Malaysia

  • Venue:
  • MCPR'11 Proceedings of the Third Mexican conference on Pattern recognition
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Binarization process is easy when applying simple thresholding method onto good quality image. However, this task becomes difficult when it deals with degraded image. Most current binarization methods involve complex algorithm and less ability to recover important information from a degradation image. We introduce an adaptive binarization method to overcome the state of the art. This method also aims to solve the problem of the low contrast images and thin pen stroke problems. It can also enhance the effectiveness of solving all other problems. As well as, it does not need to specify the values of the factors manually. We compare the proposed method with known thresholding methods, which are Niblack, Sauvola, and NICK methods. The results show that the proposed method gave higher performance than previous methods.