Machine Vision Algorithms in Java: Techniques and Implementation
Machine Vision Algorithms in Java: Techniques and Implementation
Artificial Intelligence: A Guide to Intelligent Systems
Artificial Intelligence: A Guide to Intelligent Systems
Digital Image Processing
| Hi-index | 0.00 |
Nowadays, the history of Batu Aceh has been forgotten through the centuries. If this happens, future generations will not know the absence of this creative heritage. Even to an expert, it takes time for them to recognize and memorize each type easily. To solve this difficulty, a prototype is devised to guide future generation to appreciate these precious cultural heritage artifacts in the Islamic-Malay civilization. A neural network approach is employed for supervised classification of this Batu Aceh object images. In this research, back propagation algorithm is applied. In order to classify the type, several images of each type of Batu Aceh are used as training samples. The image samples would be processed to extract useful information to be fed into each type of Batu Aceh. The network will be trained first with data samples that have been converted into binary forms. Then, network parameters such as momentum value, learning rate and number of hidden neuron will be set to ensure the performance of the system. After several experiments were conducted, 0.04 learning rate value with 40 hidden neurons in the hidden layer was found to be the optimal parameter values for the neural network. The learning curve is smooth and the performance goal is met.