Feature extraction of jabon (anthocephalus sp) leaf disease using discrete wavelet transform

摘要: Jabon (Anthocephalus cadamba (Roxb.) Miq) is one type of forest plants that have very rapid growth until the process harvest. One inhibitor a disease attacks leaves in form spots and blight can cause death during this tree. The purpose to detect object diseases attack jabon at time nursery. Images affected leaf segmented by reducing RGB color cylinders separate from background. Reduced channel G-R provides information areas contained image leaf. Furthermore, characteristics be detected well using DWT 3-level decomposition with SVM classification results both classes 84.672%.