Identifikasi Huruf Kapital Tulisan Tangan Menggunakan Linear Discriminant Analysis dan Euclidean Distance

DOI: https://doi.org/10.21456/vol8iss1pp57-67

Article Metrics: (Click on the Metric tab below to see the detail)

Article Info
Published: 30-04-2018
Section: Research Articles
Fulltext PDF Tell your colleagues Email the author

The human ability to recognize a variety of objects, however complex the object, is the special ability that humans possess. Any normal human will have no difficulty in recognizing handwriting objects between an author and another author. With the rapid development of digital technology, the human ability to recognize handwriting objects has been applied in a program known as Computer Vision. This study aims to create identification system different types of handwriting capital letters that have different sizes, thickness, shape, and tilt (distinctive features in handwriting) using Linear Discriminant Analysis (LDA) and Euclidean Distance methods. LDA is used to obtain characteristic characteristics of the image and provide the distance between the classes becomes larger, while the distance between training data in one class becomes smaller, so that the introduction time of digital image of handwritten capital letter using Euclidean Distance becomes faster computation time (by searching closest distance between training data and data testing). The results of testing the sample data showed that the image resolution of 50x50 pixels is the exact image resolution used for data as much as 1560 handwritten capital letter data compared to image resolution 25x25 pixels and 40x40 pixels. While the test data and training data testing using the method of 10-fold cross validation where 1404 for training data and 156 for data testing showed identification of digital image handwriting capital letter has an average effectiveness of the accuracy rate of 75.39% with the average time computing of 0.4199 seconds.

Keywords

Computer Vision; Euclidean Distance; Linear Discriminant Analysis; 10-Fold Cross Validation.

  1. Septa Cahyani 
    Universitas Indo Global Mandiri, Indonesia
  2. Rita Wiryasaputra 
    Universitas Indo Global Mandiri, Indonesia
  3. Rendra Gustriansyah 
    Universitas Indo Global Mandiri, Indonesia
  1. Achmad, Balza. 2006. Diktat Kecerdasan Buatan. Yogyakarta: Fakultas Teknik Universitas Gadjah Mada.
  2. Falanda, Filian, Rendra Gustriansyah, and Hartini. 2016. “Penentuan Objek Wisata, Objek Kuliner Serta Akomodasi Disekitar Pengguna Dikota Palembang Dengan Menggunakan Algoritma Euclidean Distance.” Ilmiah Informatika Global 7(1):17–24.
  3. Han, J., M. Kamber, and J. Pei. 2012. Data Mining: Concepts and Techniques. 3rd ed. Massachusetts (US): Morgan Kaufmann. Retrieved (http://eprints.uny.ac.id/41356/2/BAB II KAJIAN TEORI.pdf).
  4. Hastiana, Rumi. 2010. Segementasi Citra Digital Pembuluh Darah Mata Untuk Mendeteksi Tingkat Keparahan Diabetic Retinopathy. Malang.
  5. Kasih, Patmi and Yuliana Melita Pranoto. 2015. “Handwritten Character Recognition Untuk Evaluasi Perkembangan Kemampuan Menulis Anak Paud.” Pp. 273–81 in Seminar Nasional “Inovasi dalam Desain dan Teknologi.” IDeaTech2015.
  6. Muntasa, Arif. 2015. Pengenalan Pola; Aplikasi Untuk Pengenalan Wajah, Analisis Terstruktur Objek, Pengenalan Plat Nomor Kendaraan Dan Segmentasi Pembuluh Darah. Yogyakarta: Graha Ilmu.
  7. Pratiwi. 2014. “Metode Ekstraksi Ciri 2dpca Pada Pengenalan Citra Wajah Dengan Matlab.” Teknologi 7:1–5.
  8. Sholahuddin, Asep, Rustam E. Siregar, Iping Supriana, and Setiawan Hadi. 2010. “Penerapan Metode Linear Discriminant Analysis Pada Pengenalan Wajah.” Konferensi Nasional Matematika, Ke-15 Di UNIMA (December 2013).
  9. Simangunsong, VFR. 2015. “Klasifikasi Fragmen Metagenom Menggunakan Principal Component Analysis Dan K-Nearest Neighbor.” Institut Pertanian Bogor.
  10. Supardi. 2010. “Pemrograman Komputer Pendidikan Fisika Smt II Tahun 2009/2010.” Retrieved (https://supardi.files.wordpress.com/2010/03/praktikum-4.pdf ).
  11. Sutojo, T., Edy Mulyanto, and Suhartono Vincent. 2011. Kecerdasan Buatan. Yogyakarta: Andi.
  12. Sutoyo, T., Edy Mulyanto, Vincent Suhartono, Oky Dwi Nurhayati, and Wijanarto. 2009. Teori Pengolahan Citra. 1 st Publi. Yogyakarta: Andi.
  13. Wirayuda, Tjokorda Agung Budi, Syilvia Vaulin, and Retno Novi Dayawati. 2009. “Pengenalan Huruf Komputer Menggunakan Algoritma Berbasis Chain Code Dan Algoritma Sequence Alignment.” 19–24.
  14. Wiryasaputra, Rita, Rendra Gustriansyah, and Wawan Nurmansyah. 2013. “Pembangunan M-Konseling Psikologi Klinis.” Seminar Nasional Teknologi Informasi 74–78.
  15. Witten, H.Ian, Eibe Frank, and A.Mark Hall. 2011. Data Mining Practical Mechine Learning Tools and Techniques Third Edition (3rd Ed.). Elsevier Inc.