ON-LINE HANDWRITTEN ARABIC CHARACTER RECOGNITION BASED ON GENETIC ALGORITHM

Abstract

On-line Arabic handwritten character recognition is one of the most challenging problems in pattern recognition field. By now, printed Arabic character recognition and on-line Arabic handwritten recognition has been gradually practical, while offline Arabic handwritten character recognition is still considered as "The hardest problem to conquer" in this field due to its own complexity. Recently, it becomes a hot topic with the release of database, which is the first text-level database and is concerned about the area of realistic Arabic handwritten character recognition.

At the realistic Arabic handwritten text recognition and explore two aspects of the problem. Firstly, a system based on segmentation-recognition integrated framework was developed for Arabic handwriting recognition. Secondly, the parameters of embedded classifier initialed at character-level training were discriminatively re-trained at string level.

            The segmentation-recognition integrated framework runs as follows: the written character is first over-segmented into primitive segments, and then the consecutive segments are combined into candidate patterns. The embedded classifier is used to classify all the candidate patterns in segmentation lattice. According to Genetic Algorithm (Crossover, mutation, and population), the system outputs the optimal path in segmentation-recognition lattice, which is the final recognition result. The embedded classifier is first trained at character level on isolated character and then the parameters are updated at string level on string samples.