Introduction of Indonesian Significant Alphabet Images (BISINDO) using The Convolutional Neural Network Algorithm

Kabut, Stefanus
Kelen, Yoseph Pius Kurniawan
Baso, Budiman
Chrisinta, Debora

Jurnal Sisfotek Global
Vol 15, No 1 (2025): JURNAL SISFOTEK GLOBAL

Bisindo alphabet recognition is the process by which a computer system or software recognizes and recognizes the letters of the Bisindo alphabet. The Bisindo alphabet is a special alphabet used to communicate with people who are hearing or speech impaired. This process uses image processing and machine learning techniques to identify and classify each letter based on its shape and visual characteristics. This study used a dataset consisting of 520 Kaggle images divided into 26 categories. These images are resized, normalized and scaled up to improve model performance. A Convolutional Neural Network (CNN) model was developed and achieved 99.12587% accuracy after training. After the model was developed, the API was implemented using Flask. API functionality is tested using online interactions, ensuring accurate responses to image classification before implementation in mobile applications.