Khmer Optical Character Recognition Using Zernike Moment

Chuor, Porchourng;Verleysen, Michel;Valy, Dona
(2015) 2015 Khmer Natural Language Processing annual conference (KNLP 2015) — Location: Phnom Penh (Cambodia) (4.December.2015)

Files

KhmerOpticalCharacterRecognitionusingZernikeMoment.pdf
  • Restricted Access
  • Adobe PDF
  • 332.05 KB
Request a copy

Details

Authors
  • Chuor, PorchourngInstitute of Technology of Cambodia
    Author
  • Verleysen, MichelUCLouvain
    Author
  • Valy, DonaInstitute of Technology of Cambodia
    Author
Abstract
In this paper, we focus on an Optical Character Recognition (OCR) system for printed text documents in Khmer language by using Zernike Moment. The Zernike Moment method is used as a feature extraction method to solve the recognition problem. We compute the moments from sub-characters to extract their feature vectors. The final recognition result is achieved by employing a classifier based on the Nearest Neighbor method. The method is experimented on 5 documents with font size of 12, 15, and 36 respectively.
Affiliations

Citations

  • APA
  • Chicago
  • FWB

Chuor, P., Verleysen, M., & Valy, D. (2015). Khmer Optical Character Recognition Using Zernike Moment. 2015 Khmer Natural Language Processing annual conference (KNLP 2015), Phnom Penh (Cambodia). https://hdl.handle.net/2078.5/253903