| dc.description.abstract |
Guava leaves that are consumed have good properties for humans, but not all guava leaves can be consumed, this is due to the many varieties and types of guava currently available, so that some people do not recognize the type of guava based on the shape of the leaves. Therefore, to avoid the problem of identifying types of guava based on the shape of the guava leaf, a study was carried out on the classification of guava types based on the shape of the guava leaf. In this study, the classification of guava types used for classification based on leaf shape using the Principle Component Analysis (PCA) and K-Nearest Neighbor (K-NN) methods is the types of guava leaves, guava leaves, cashew leaves and water guava leaves.
. The PCA method functions to extract guava leaf images by obtaining eigenvector values and eigenvalues, while K-NN is used for the process of recognizing guava species based on the proximity of the eigenvector values and eigenvalues of each leaf image, so as to produce output in the form of a classification of guava species. From the research results obtained as many as 16 test images, obtained an accuracy of 81.25%.
Keywords : Classification, Guava, Image, PCA, K-NN.
| en_US |
