Feature Extraction for Java Character Recognition

Feature extraction is very important in the process of character recognition. A good feature of the character will increase the level of accuracy for the character recognition. In this research, the feature extraction experiment of Java characters is done, where those features could be used for Java character recognition later. Before performing the process of feature extraction, segmentation is performed to get each Java character in an image, and followed by skeletonization process. After skeletonization process, feature extraction is done including simple closed curve, straight lines and curve. Several experiments was done using various parameters and Java characters in order to obtain the optimal parameters. The experiment results of simple closed curve and straight line feature extraction are quite good, respectively reached 82.85% and 74.28%. However, the result of the curve detection is still not good, only reached 51.42%