NISKE ELMY PAULINA
Program Studi Teknik Informatika, Politeknik Negeri Jember

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Klasifikasi Kerusakan Mutu Tomat Berdasarkan Seleksi Fitur Menggunakan K-Nearest Neighbor NISKE ELMY PAULINA; ZILVANHISNA EMKA FITRI; ABDUL MADJID; ARIZAL MUJIBTAMALA NANDA IMRON
MIND (Multimedia Artificial Intelligent Networking Database) Journal Vol 6, No 2 (2021): MIND Journal
Publisher : Institut Teknologi Nasional, Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/mindjournal.v6i2.144-154

Abstract

AbstrakTomat (Lycopersicum esculentum Mill.) merupakan satu komoditas unggulan pertanian karena penjualan jangka panjangnya baik. Menurunnya jumlah produktivitas dan mutu tomat disebabkan oleh curah hujan yang tinggi, cuaca dan budidaya yang tidak baik sehingga buah tomat menjadi busuk, retak, dan timbul bercak. Penyuluhan terkait peningkatan mutu tomat dinilai kurang efektif sehingga dibutuhkan sebuah sistem identifikasi kerusakan mutu buah tomat yang mampu memberikan edukasi kepada petani. Penelitian ini adalah pengembangan penelitian sebelumnya, untuk mendapatkan citra segmentasi dan ekstraksi fitur digunakan penggunaan contrast stretching dan deteksi tepi sobel. Namun kedua teknik tersebut diganti penggunaan operasi citra negatif. Didapatkan fitur yang optimal adalah gabungan fitur morfologi dan pada masing-masing sudut berdasarkan seleksi fitur. Persentasi akurasi metode KNN pada pelatihan sebesar 86.6% sedangkan akurasi pengujiannya sebesar 70%.Kata kunci: kerusakan mutu, tomat, seleksi fitur, K-Nearest NeighborAbstractTomato (Lycopersicum esculentum Mill.) is one of the leading agricultural commodities because of its good long-term sales. The decrease in the amount of productivity and quality of tomatoes is caused by high rainfall, bad weather and cultivation so that the tomatoes become rotten, cracked, and have spots. Counseling related to improving the quality of tomatoes is considered ineffective so that a system for identifying damage to the quality of tomatoes is needed that is able to provide education to farmers. This study is a development of previous research, to obtain segmented images and feature extraction using contrast stretching and sobel edge detection. However, both techniques were replaced by using negative image operations. The optimal feature is a combination of morphological features and correlations at each angle based on feature selection. The percentage of accuracy of the KNN method in training is 87%, while the accuracy in the testing is 70%.Keywords: quality damage, tomato, feature selection, K-Nearest Neighbo