Pengenalan Karakter Jawi Tulisan Tangan Menggunakan Fitur Sudut
Abstract
Abstrak— Karakter Jawi salah satu varian karakter Arab adalah karakter Jawi. Karakter tersebut mengakomodasi penulisan bahasa Melayu. Karakter Jawi banyak digunakan oleh kerajaan Islam di Nusantara sehingga banyak dokumen menggunakan karakter tersebut. Untuk menjaga kelestarian dokumen tersebut maka karakter Jawi tersebut harus di rubah kedalam format digital, salah satu caranya adalah dengan metode Optical Character Recognition (OCR). Salah satu metode OCR dalam mengekstraksi fitur adalah menggunakan sudut yang dibentuk dari bagian utama karakter. Fitur yang dihasilkan dari sudut tersebut digunakan oleh SVM untuk di klasifikasi kedalam karakter Jawi. Dengan menggunakan fitur tersebut, SVM mampu mengklasifikasi karakter Jawi dengan tingkat akurasi rata-rata sebesar 78,86%.
Abstract— The Jawi character is one variant of Arabic characters that adopts Malay language writing. The character of Jawi is widely used by the Islamic kingdom in the Archipelago. Many document written using these characters. To preserve the history and the book, the Jawi character need to be converted into digital form, one way is by Optical Character Recognition (OCR) method. One method of OCR in extracting features is to use angles formed by main part of the character. The features generated are used by SVM to be classified into Jawi characters. By using these features, SVM is able to classify Jawi characters with an average level of accuracy of 78.86%.
Full Text:
PDFReferences
H. Modi and M. C., “A Review on Optical Character Recognition Techniques,” Int. J. Comput. Appl., vol. 160, no. 6, pp. 20–24, 2017.
L. M. Lorigo and V. Govindaraju, “Offline Arabic Handwriting Recognition :,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 28, no. 5, pp. 712–724, 2006.
M. F. Nasrudin, K. Omar, M. S. Zakaria, and L. C. Yeun, “Handwritten Cursive Jawi Character Recognition : A Survey,” pp. 247–256, 2008.
Safrizal, F. Arnia, dan R. Muharar. “Pengenalan Aksara Jawi Tulisan Tangan Menggunakan Freeman Chain Code (FCC), Support Vector Machines, dan Aturan Pengambilan Keputusan,” Jurnal Nasional Teknik Elektro, vol. 5(1), pp. 45–55, 2016
R. Redika, F. Teknologi, K. Omar, F. Teknologi, and F. Teknologi, “Handwritten Jawi Words Recognition Using Hidden Markov Models,” pp. 4–8, 2008.
M. F. Nasrudin, M. Petrou, and L. Kotoulas, “Jawi Character Recognition using the Trace Transform,” 2010.
A. Heryanto, F. Teknologi, S. Maklumat, F. Teknologi, K. Omar, and F. Teknologi, “Offline Jawi Handwritten Recognizer Using Hybrid Artificial Neural Networks and Dynamic Programming,” 2008.
T. R. Singh, S. Roy, O. I. Singh, T. Sinam, and K. M. Singh, “A New Local Adaptive Thresholding Technique in Binarization,” vol. 8, no. 6, pp. 271–277, 2012.
N. K. Garg, “Binarization Techniques used for Grey Scale Images,” vol. 71, no. 1, pp. 8–11, 2013.
N. Otsu, “A Threshold Selection Method from Gray-Level Histograms,” IEEE Trans. Syst. Man. Cybern., 2008.
T. Y. Zhang and C. Y. Suen, “A fast parallel algorithm for thinning digital patterns,” Commun. ACM, 1984.
L. Lam and S. W. Lee, “Thinning methodologies—a comprehensive survey,” IEEE Trans. Pattern Anal. Mach. Intell., 1992.
DOI: https://doi.org/10.38038/vocatech.v1i0.1
Refbacks
- There are currently no refbacks.
VOCATECH : Vocational and Technology Journal
Unit Penelitian dan Pengabdian Masyarakat & Penjaminan Mutu
Akademi Komunitas Negeri Aceh Barat
Komplek STTU Alue Peunyareng, Ujong Tanoh Darat, Meureubo, Kabupaten Aceh Barat, Aceh 23615
Telp. (0655) 7110271
Email: vocatech@aknacehbarat.ac.id
VOCATECH: Vocational Education and Technology Journal Published by:
Unit Penelitian dan Pengabdian Masyarakat & Penjaminan Mutu
Akademi Komunitas Negeri Aceh Barat
Indexed by:

VOCATECH: Vocational Education and Technology Journal Creative Commons Attribution-ShareAlike 4.0 International License.