Ensiklopedia Digital Varietas Ubi Jalar Berdasarkan Klasifikasi Citra Daun Menggunakan KNearest Neighbor

Prasetya, Bahtiar Adi and Fitri, Zilvanhisna Emka and Madjid, Abdul and Imron, Arizal Mujibtamala Nanda (2022) Ensiklopedia Digital Varietas Ubi Jalar Berdasarkan Klasifikasi Citra Daun Menggunakan KNearest Neighbor. Elektrika, 14 (1). pp. 1-6. ISSN 2085-0565

	Text (cek turnitin) Elektrika_Ensiklopedia Digital Varietas Ubi Jalar Berdasarkan Klasifikasi Citra Daun Menggunakan KNearest Neighbor.pdf - Supplemental Material Available under License Creative Commons Attribution Share Alike. Download (2MB)
	Text (full paper) 14.Elektrika_2022.pdf - Published Version Restricted to Repository staff only Available under License Creative Commons Attribution Share Alike. Download (1MB)

Abstract

Sweet potato is a source of carbohydrates which is an alternative food in order to accelerate food diversification. This is due to the high productivity of sweet potato so it is very profitable to cultivate. Sweet potato has many varieties, one of the differences is observed based on leaf shape which has four kinds of leaf shape, namely cordate, lobed, triangular and almost divided. The problem that often occurs is that many varieties have similarities, causing difficulties in distinguishing sweet potato varieties, especially for novice farmers. To overcome this problem, the researchers created a digital encyclopedia of sweet potato varieties based on leaf shape using computer vision. The parameters used are area, perimeter, metric, length, diameter, ASM, IDM, entropy, contrast and correlation at angles of 0°, 45°, 90° and 135°. The amount of data used is 256 training data and 40 testing data. The K-Nearest Neighbor method is able to classify sweet potato leaf images for digital encyclopedias with an accuracy of 95% with variations in the values of K = 23 and K = 25.

Actions (login required)

View Item