Handwritten character and word recognition using their geometrical features through neural networks

ABSTRACT In this research we are aiming to use geometrical features and evolutionary computational algorithm to automatically recognize (read) off-line handwritten character , two letter words using Sobel edge detection technique with an increased feature extraction. The nature of handwritten characters is difficult to identify and hence the problems that could be faced when automatically (optically) recognizing them. This Research concentrates on the feature extraction process, i.e. extraction of the main geometrical features of each of the extracted handwritten characters and then two letter words. A complete system able to recognize handwritten characters of only a single writer is proposed and discussed. A review of some of the previous trials in the field of off-line handwritten character recognition is included. The system first attempts to remove some of the variations found in the images that do not affect the identity of the handwritten word (slant correction, slope correction, and baseline estimation). Next, the system Codes the skeleton of the word so that feature information about the lines in the skeleton is extracted (segmentation and feature extraction). The features include locating endpoints, junctions, turning points, loops, generating frames (segmentation step) and detecting strokes. These features are then passed on to the recognition system for recognition. The character classification is achieved in this research using a feed-forward error back propagation neural. Similar way the two letter words from the same writer is also found and identified. In this method the data of the tow letter words is already stored in the offline dictionary for usage. Keywords: Neural network, feed forward back propagation, feature extraction.