Islamic Star Pattern Images Recognition by Central Moment Invariants

In this paper, a proposed system for classify Islamic geometric patterns with emphasis on representation and recognition stages is introduced. Finding a unique feature for classify IGPs is a hard work that hasn’t been done since today because of the diversity of their different structure. To implement this technique, we use shape based classification. Geometric central moments have been utilized as shape image descriptor. In classification stage, two different classifiers namely K-nearest neighbor rule, feed forward neural network has been used Set of different experiments on binary images of regular, translated, rotated and scaled Islamic geometric shape has been done and variety of results has been presented. The best result was 92.84% correct recognition demonstrating geometrical central moments and Nearest Neighbor are adequate for Islamic star pattern recognition. This research is a part of an application for analysis Islamic Geometric Patterns images.