Pemutuan belimbing manis dengan citra pelatihan tunggal menggunakan algoritme VF15 berbasis histogen warna

Abstract

Information on starfruit grading is necessary to professional distributor to give the best service. One of modern technique is digital image processing which using image and manipulating it to get one or more information. RGB, HSV, and YCbCr are standard models used in various color imaging applications. This research uses histogram of RGB, HSV, and YCbCr models and to show that not all of information is effective to classify starfruit. Image histogram is the basic for numerous spatial domain processing technique. Image histogram is used as input to Voting Feature Interval 5 (VFI5) algorithm. Histogram will be divided into several intervals and On training stage, it will be determined vote value from single training image per class. The vote value is used to classify testing data. This research shows that R, H, and Cr give the best accuracy to classify testing data with accuracy reaches 90.20%.