Pengenalan Citra Penyakit Daun Jagung Berbasis K-Nn (K-Nearest Neighbors)

FAOT, Nunung Ardyani (2023) Pengenalan Citra Penyakit Daun Jagung Berbasis K-Nn (K-Nearest Neighbors). Undergraduate thesis, Universitas Katolik Widya Mandira Kupang.

	Text ABSTRAK.pdf Download (246kB)
	Text BAB I.pdf Download (23kB)
	Text BAB II.pdf Restricted to Repository staff only Download (197kB)
	Text BAB III.pdf Restricted to Repository staff only Download (142kB)
	Text BAB IV.pdf Restricted to Repository staff only Download (336kB)
	Text BAB V.pdf Download (76kB)

Abstract

Corn is one of the staple foods of the world in general. Corn is one of the most important carbohydrate-producing feed crops in the world, in addition to wheat and rice. The average source of productivity at the farm level is still very low, ranging from 1.5-2.0 tons/ha 2007. There are still many corn farmers who do not know how or techniques to control diseases that attack corn leaves, in general diseases that attack corn leaves detected manually by the human eye. In this increasingly sophisticated era, technology is needed that can assist farmers in predicting corn leafdisease using a classification model. There have been previous studies in identifying corn leaf disease using KNN, but the accuracy rate is only 73.3%. This study aims to improve the accuracy of image recognition of sick corn leaves using the k-Nearest Neighbors (KNN) classifier. This study uses KNN with a number of neighbors 10, 5 and 20. The results of the calculation of the performance obtained by KNN k=10 is 87,8%, KNN k=5 is 86,9%, and KNN k=20 is 87,1%.

Actions (login required)

View Item