Classification of Calligraphy Writing Types Using Convolutional Neural Network Method (CNN)

  • Putra O
  • Musthafa A
  • Nur M
  • et al.
N/ACitations
Citations of this article
28Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Calligraphy is the art of beautiful Arabic writing in which a series of letters are formed in appropriate proportions, maintaining distance and accuracy containing verses from the Qur'an or Hadith. There is a challenge to recognize the type of calligraphy using machine learning. This study aims to classify the types of calligraphy writing for ordinary people who do not understand the differences between each type of calligraphy writing. This study builds a model using the Convolutional Neural Network (CNN) method. The image used will go through a noise cleaning, resizing, and cropping process. This method is to carry out the process of classifying the type of calligraphy using a dataset consisting of 230 of 2 different types of calligraphy, namely the Naskhi and Riq'ah types. 80% is used as training data and 20% for test data. In the modeling process there are two convolutional layers and two MaxPooling layers followed by a Fully connected layer. The CNN modeling results used to test the built data have an average percentage result of 89% accuracy from the training data used. For further research, it can be developed with other types of calligraphy.

Cite

CITATION STYLE

APA

Putra, O. V., Musthafa, A., Nur, M., & Rido, M. (2021). Classification of Calligraphy Writing Types Using Convolutional Neural Network Method (CNN). Procedia of Engineering and Life Science, 2. https://doi.org/10.21070/pels.v2i0.1136

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free