Pengenalan Huruf Braille Menggunakan Radially Average Power Spectrum Dan Geometri

Soffiana Agustin(1*), Anita Sari(2), Ernawati Ernawati(3)

(1) Universitas Muhammadiyah Gresik
(2) Universitas Muhammadiyah Gresik
(3) Universitas Muhammadiyah Gresik
(*) Corresponding Author

Abstract


Education is very important in the growth of children, including children with special needs such as people with visual impairments. Not only at school children can also develop their potential in learning anywhere and anytime. One of the most important learning processes is reading and writing. People with visual impairments carry out reading and writing activities using Braille. The problem is the lack of family knowledge about braille, so the activities accompanying blind children cannot be conducted. This study aims to convert braille into text, so it can make it easier for families to understand braille. The introduction of braille letters is done by extraction of frequency and spatial features. The methods proposed in this study are fast fourier transform (fft), radially average power spectrum (rapsv) and wavelets as well as several spatial features, namely local binary pattern (lbp), segmentation based on fractal analysis (sfta), first order statistics, gray level co-occurrence matrix (glcm), moment invariant and geometric features. In this study, the classification process was carried out using Bayes Net, Naïve Bayes, SVM, KNN and Random Forest. From several experiments it was found that the Random Forest classification method gave the best results. The rapsv method provides an accuracy of 93.91%, frequency feature extraction produces the same accuracy as the combination of rapsv with geometry, which is 94.04% and of all features, an accuracy of 97.18% is obtained.

Article metrics

Abstract views : 166 | views : 29

Full Text:

PDF (Bahasa Indonesia)

References


N. Muda, N. K. Nik Ismail, S. A. Abu Bakar and J. M. Zain, "Optical Character Recognition By Using Template Matching (Alphabet)," UMP Institutional Repository, Malaysia, 2018.

M. Ryan and N. Hanafiah, "An Examination of Character Recognition on ID card using Template Matching Approach," Procedia Computer Science, vol. 59, pp. 520-529, 2015P.

R. Trianto, N. N. D. Merdekawati, R. P. S. Nugraha, D. N. Y. Astiti and H. G. Tri Atmojo, "KLASIFIKASI HURUF KATAKANA DENGAN METODE," Universitas Brawijaya, Malang, Indonesia, 2014.

S. Muharom, "Pengenalan Nomor Ruangan Menggunakan Kamera Berbasis OCR Dan Template Matching," Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi, vol. 4, no. 1, pp. 27-32, 2019.

I. Ozdemira and A. Karnielib, "Predicting forest structural parameters using the image texture derived from WorldView-2 multispectral imagery in a dryland forest, Israel," Science Direct - International Journal of Applied Earth Observation and Geoinformation, vol. 13, p. 701–710, 2011.

Z. Szantoia, F. Escobedob, A. Abd-Elrahmana, S. Smith and L. Pearlstinec, "Analyzing fine-scale wetland composition using high resolution imagery and texture features," Science direct - International Journal of Applied Earth Observation and Geoinformation, vol. 23 , p. 204–212, 2013.

B. Beguet, D. Guyon, S. Boukir and N. Chehata, "Automated retrieval of forest structure variables based on multi-scale texture analysis of VHR satellite imagery," Science Direct - ISPRS Journal of Photogrammetry and Remote Sensing, vol. 96, p. 164–178, 2014.

B. Sırmacek and C. Unsalan, "Urban Area Detection Using Local Feature Points and Spatial Voting".

M. Laba, B. Blair, R. Downs, B. Monger, W. Philpot, S. Smith, P. Sullivan and P. C. Baveye, "Use of textural measurements to map invasive wetland plants in the Hudson River National Estuarine Research Reserve with IKONOS satellite imagery," Science Direct - Remote Sensing of Environment , vol. 114, p. 876–886, 2010.

H. Fadaei, T. Sakai, T. Yoshimura and M. Kazuyuki, "ESTIMATION OF TREE DENSITY WITH HIGH-RESOLUTION IMAGERY IN THE ZARBIN FOREST OF NORTH IRAN (CUPRESSUS SEMPERVIRENCE VAR. ORZONTALIS),," International Archives of the Photogrammetry, Remote Sensing and Spatial Information Science, Kyoto Japan, vol. XXXVIII, no. 8, pp. 679 - 684, 2010.

D. Neves and C. Carneiro, "Semi-automatic Use of High Resolution Images and Digital Elevation Models for Counting and Identification of Forest Trees," in Geocomputation 2007 Conference, Maynooth, Ireland, 2007.

J. Li, X. Yan and D. Zhang, "Optical Braille recognition with Haar wavelet features and Support-Vector Machine," Changchun, 2010.

Y. M. Florestiyanto and H. Prapcoyo, "Braille Detection Application Using Gabor Wavelet and Support Vector Machine," RSF Conference Series: Engineering and Technology, vol. 1, no. 1, pp. 160-169, 2021.

S. Agustin, H. Ginardi and H. Tjandrasa, "Identification of Oil Palm Plantation in Ikonos Images using Radially Averaged Power Spectrum Values," in International Conference on Information, Communication Technology and System, Surabaya, Indonesia, 2015.

Y. Sahaduta and C. Lubis, "Gray Level Co-ocurence Matrix sebagai Pengekstraksi Ciri pada Pengenalan Naskah Braille," in Seminar Nasional TeknologiInformasi dan Multimedia STMIK AMIKOM Yogyakarta, Yogyakarta, Indonesia, 2013.

E. Ronando and A. Sudaryanto, "Sistem Pengenalan Pola Huruf Brailler Berbasis Audio menggunakan Metode Naive Bayes," Jurnal Ilmu Komputer dan Desain Komunikasi Visual, pp. 43 - 52, 2018.

R. E. S.-Y. Jonathan Cepeda-Negrete, "Automatic selection of color constancy algorithms for dark image enhancement by fuzzy rule-based reasoning," Elsevier - Science Direct, vol. 28, no. Soft Computing, pp. 1-10, 2015.

A. Kadir and A. Susanto, Teori dan Aplikasi Pengolahan Citra, Yogyakarta: AndiPublisher, 2013.




DOI: https://doi.org/10.35314/isi.v8i1.2926

Refbacks

  • There are currently no refbacks.




Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.


This Journal has been listed and indexed in :

Crossref logo Find in a library with WorldCat

Copyright of Jurnal Inovtek Polbeng - Seri Informatika (ISSN: 2527-9866)

Creative Commons License
ISI: Inovtek Polbeng Seri Informatikan is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Editorial Office :
Pusat Penelitian dan Pengabdian kepada Masyarakat
 Politeknik Negeri Bengkalis 
Jl. Bathin alam, Sungai Alam Bengkalis-Riau 28711 
E-mail: jurnalinformatika@polbeng.ac.id
www.polbeng.ac.id

Web
Analytics
View My Stats