PENGARUH PEMAKAIAN EDGE DETECTION PADA SISTEM PENGENALAN HURUF KAPITAL TULISAN TANGAN

Abstract

This paper investigates the influence of edge detection as a pre processing method for a capital letter handwriting recognition system on the accuracy of nearest neigbor classification system. The experimental results show that the use of edge detection did not  significantly enhance the recognition accuracy in comparison to the selection of large structuring element size in the dilation operation.

 

Keywords:     Sobel edge detection, nearest neighbor, dilation morphology, binary image, handwriting recognition