International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 160 - Number 4 |
Year of Publication: 2017 |
Authors: Anindya Halder, Amit Kumar Upadhyay, Sujit Kumar Das |
10.5120/ijca2017913044 |
Anindya Halder, Amit Kumar Upadhyay, Sujit Kumar Das . Numismatic Image Segmentation: An Empirical Study. International Journal of Computer Applications. 160, 4 ( Feb 2017), 36-39. DOI=10.5120/ijca2017913044
Numismatic coins are one of the important elements to know history of economic, culture and society in ancient as well as present time. Extraction of information from numismatic coins in traditional manual method sometimes contains human error and in the same time it is time consuming. Engineering in this aspects deal with problems of traditional method. Engineering tasks namely pattern recognition, computer vision, and object authentication are high level computational tasks where machine deal with image data for processing. To do these high level tasks image segmentation is a basic process. Its purpose is to separate the targets from the background in an image in order to provide the basis for the subsequent sorting, recognition and indexing. Again segmentation methods depends on types of image data, we are dealing with. In this paper we discussed some existing methods of image segmentation and tried to find out suitable method for coin images. Maximum Entropy Thresholding(MET) based on normalised histogram method giving us more suitable segmented image.