Penerapan Seleksi Fitur Pada Metode K-Nearest Neighbor Untuk Identifikasi Kerusakan Buah Tomat (Lycopersicon Esculentum L.)
Total View This Week0
Total View This Week0
Institusion
Politeknik Negeri Jember
Author
Paulina, Niske Elmy
Subject
458 - Teknik Informatika
Datestamp
2021-09-16 08:32:34
Abstract :
Tomato (Lycopersicon esculentum L.) is one of the most horticultural crops that have economically high value in Indonesia. The researchers applied feature selection to the K-Nearest Neighbor method to identify damage to tomatoes (Lycopersicon Esculentum L.) to increase government efforts through counseling to make it easier for farmers to recognize the damage of tomat - feature extraction, namely: perimeter, area and shape factor. The use of digital images is an effective and efficient way to identify damaged tomatoes without damaging the fruit. By using the k-nearest neighbor method, from the comparison of k values, the highest percentage of accuracy is K=3 with training results of 87% and testing of 70%. The results of the feature selection obtained a comparison with a suitable level of accuracy for k-nearest neighbor, namely morphology and correlation GLCM features have better accuracy compared to other morphological and GLCM features. Keywords : Tomato, Morphological Feature, GLCM, K-Nearest Neighbor
Tomato (Lycopersicon esculentum L.) is one of the most horticultural crops that have economically high value in Indonesia. The researchers applied feature selection to the K-Nearest Neighbor method to identify damage to tomatoes (Lycopersicon Esculentum L.) to increase government efforts through counseling to make it easier for farmers to recognize the damage of tomat - feature extraction, namely: perimeter, area and shape factor. The use of digital images is an effective and efficient way to identify damaged tomatoes without damaging the fruit. By using the k-nearest neighbor method, from the comparison of k values, the highest percentage of accuracy is K=3 with training results of 87% and testing of 70%. The results of the feature selection obtained a comparison with a suitable level of accuracy for k-nearest neighbor, namely morphology and correlation GLCM features have better accuracy compared to other morphological and GLCM features. Keywords : Tomato, Morphological Feature, GLCM, K-Nearest Neighbor
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
