Buddhist amulet coin recognition by genetic algorithm

Abstract

The objective of this research is to develop a computer system, which can recognize some Thai Buddhist amulet coins. The system is called “Buddhist amulet coin recognition system” or BACRS. The system consists of five modules, namely: 1) image acquisition, 2) image preprocessing, 3) feature extraction, 4) image recognition and 5) result presentation. The BACRS applied the rule-based technique and genetic algorithm method to recognize Buddhist amulet coins. The BACRS used 216 amulet coin patterns with 1,080 images to train the system. The experiments conducted on 130 Buddhist amulet coins for recognition. The precision rate of the system is 91.53 percent and the average access time is 1.05 seconds per amulet image.