Convolutional Neural Network (CNN) to determine the character of wayang kulit

Abstract

Indonesia is a country with diverse cultures. One of which is Wayang Kulit, which has been recognized by UNESCO. Wayang kulit has a variety of names and personalities, however most younger generations are not familiar with the characters of these shadow puppets. With today's rapid technological advancements, people could use this technology to detect objects using cameras. Convolutional Neural Network (CNN) is one method that can be used. CNN is a learning process that is included in the Deep Learning section and is used to find the best representation. The CNN is commonly used for object detection, would be used to classify good and bad characters. The data used consists of 100 black and white puppet images that were downloaded one at a time. The data was obtained through a training process that uses the CNN method and Google Colab to help speed up the training process. After that, a new model is created to test the puppet images. The result obtained a 92 percent accuracy rate, means that CNN can differentiate the Wayang Kulit character

Water quality identification based on remote sensing image in industrial waste disposal using convolutional neural networks

“…In addition to using the MNIST database, tests were conducted on the handwriting of 20 correspondents who wrote the numbers 0 to 9. [6] used a convolutional neural network (CNN) to distinguish and recognize good and bad characters in wayang kulit characters. In all scenarios, the accuracy level for classifying wayang kulit characters using the CNN method has an average value of 92 percent.…”

“…The FC layer functions to distinguish between classes and perform nonlinear transformations to obtain output values. The CONV and SUBS layers are part of feature learning, while FC is part of the classification [6], [13], [28], [29].…”