Java Characters Recognition using Evolutionary Neural Network and Combination of Chi2 and Backpropagation Neural Network

摘要

Javanese language is the language used by the people on the island of Javaudand it has its own form of letters called Java characters. Recognition of Javaudcharacters is quite difficult because it consist of basic characters, numbers,udcomplementary characters, and so on. In this research we developed a systemudto recognize Java characters and compared two methods of neural networkudnamely evolutionary neural network and combination of Chi2 andudbackpropagation neural network. Input for the system is a digital image of Javaudcharacters. Before entering into the neural network, the digital image isudprocessed by reducing noise, segmentation and thinning and feature extraction.udFrom experimental result, evolutionary neural network has 60 averageudrecognition accuracy, while combination of Chi2 and backpropagation networkudhas 70 average recognition accuracy