Abstract—There are various kinds of pictures on postage stamps, which are very difficult for people and philatelists to identify them. Therefore, the objective of this research is to develop a computer system which is capable of recognizing some postage stamp images by using the digital image processing (DIP) technique. The developing system is called “Postage Stamp Recognition System” or “PSRS”. The PSRS consists of five sub-systems, which are: 1) image acquisition, 2) image preprocessing, 3) feature extraction, 4) image recognition, and 5) result presentation. The experiments were conducted on 3,593 postage stamp images, which are divided into 3,093 images for a training data set, and 500 images for testing the system. The precision of the system is 93.60 percent, with the average access time of 1.43 seconds per image.

Index Terms—postage stamp, content-based image retrieval, digital image processing, pattern recognition, euclidean distance

Cite: C. Pornpanomchai and P. Leerasakultham, "Postage Stamp Recognition System (PSRS)," International Journal of Signal Processing Systems, Vol. 2, No. 2, pp. 91-95, December 2014. doi: 10.12720/ijsps.2.2.91-95