RECOGNITION OF ALPHABET LETTERS THROUGH HANDWRITING PATTERNS USING CONVOLUTIONAL NEURAL NETWORK ALGORITHM

Gumilang, Muhammad Satrio (2024) RECOGNITION OF ALPHABET LETTERS THROUGH HANDWRITING PATTERNS USING CONVOLUTIONAL NEURAL NETWORK ALGORITHM. Tugas Akhir thesis, Informatics.

[img] Text
5200411155_Muhammad Satrio Gumilang_abstrakTA.pdf
Download (15kB)

Abstract

ABSTRACT The development of information technology has had a significant impact in solving problems faced by individuals, organizations and society. However, this speed of development also poses challenges for those with physical disabilities, such as vision loss. Losing their eyesight makes them face difficulties in communicating and writing on computers. In this research, researchers developed an automatic program that aims to overcome these obstacles by using handwriting pattern recognition techniques. This program allows individuals with vision loss to write on a computer using their handwriting as input, replacing the need for a special braille keyboard. To train a handwriting pattern recognition system, researchers used a dataset of 1,040 handwriting images, consisting of 520 images of uppercase letters and 520 images of lowercase letters. Researchers utilized the Convolutional Neural Network (CNN) method in modeling to recognize handwriting patterns with accuracy that had been carried out in several training scenarios to obtain a training accuracy of 89.82% and testing accuracy of 87%. In addition, researchers used the Streamlit framework as an interactive user interface, allowing users to easily access and use the handwriting pattern recognition program that we have developed. The system developed has the potential to become a reference basis in the development of more complex handwriting pattern recognition models, and can be applied in pattern recognition in other domains. Keywords: Braille Keyboard, Handwriting Patterns, Convolutional Neural Network.

Item Type: Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir)
Subjects: T Technology > T Technology (General)
Divisions: Fakultas Sains Dan Teknologi > S1 Informatika
Depositing User: Kaprodi S1 Informatika UTY
Date Deposited: 12 Aug 2024 04:39
Last Modified: 12 Aug 2024 04:39
URI: http://eprints.uty.ac.id/id/eprint/16027

Actions (login required)

View Item View Item
UTY Open Access is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.