Perbandingan Pengenalan Karakter Plat Kendaraan Bermotor dengan Metode City Block Distance, Euclidean Distance dan Canberra Distance

Abstract

An object has many variations of patterns that can be used as a basis for information to recognize an object. In the search of an image, in general the approach used is the distance approach. Approaches using this distance function to determine similarity or dissimilarity of two feature vector expressed by a score or rank, where the smaller the value ranking the more closely the similarities of the two vectors. Distance measurement methods that can be used in this case include city block distance, Euclidean distance, Canberra distance. This paper will discuss the comparison of the results of the motor vehicle license plate character recognition using distance measurement city block distance, Euclidean distance and Canberra distance. The result shows that for character recognition using distance measurement method, City Block Distance is better than Euclidean distance and Canberra distance. Keyword-Character recognition, distance measurement, City Block, Euclidean, Canberra