Balinese Script Handwriting Recognition Using Convolutional Neural Network

Handwritten Balinese script consists of letters that have unique and complex shapes and adapt to the writer's handwriting. The main challenge in recognizing these characters is the variation in letter shapes due to different writing styles, lighting conditions and varying image quality. The aim of this research is to develop a deep learning model using CNN to recognize handwritten Balinese script. This research has carried out several CNN model configurations with different learning rates. The research results show that the model developed can produce excellent performance with an accuracy rate of 94.4%.