Klasifikasi Tingkat Kematangan, Kualitas dan Jenis Buah Pisang Berdasarkan Ciri Warna dan Bentuk Menggunakan Artificial Neural Networks

Dwi Putro, Aditya (2023) Klasifikasi Tingkat Kematangan, Kualitas dan Jenis Buah Pisang Berdasarkan Ciri Warna dan Bentuk Menggunakan Artificial Neural Networks. Klasifikasi Tingkat Kematangan, Kualitas dan Jenis Buah Pisang Berdasarkan Ciri Warna dan Bentuk Menggunakan Artificial Neural Networks, 7 (2). pp. 91-98. ISSN 2502-1613

	Text Klasifikasi Tingkat Kematangan, Kualitas dan Jenis Buah Pisang Berdasarkan Ciri Warna dan Bentuk Menggunakan Artificial Neural Networks fix.pdf Download (383kB)
	Text Plagiasi.pdf Download (1MB)
	Text Bukti Koresponden JTII fix.pdf Download (259kB)

Abstract

Banana fruit is a commodity that makes a major contribution to national and international fruit production figures. The government through the National Standardization Agency establishes standards for bananas, maintaining the quality of bananas. The purpose of this study was to classify the level of maturity, quality and type of banana based on color, size and shape characteristics in the Cavendish Banana Garden, Banyumas Regency, Central Java in accordance with SNI 7422: 2009. The bananas found in the Cavendish Banana Garden have various qualities, as a local fruit that has high economic value and has a market potential that is still wide open, bananas are one of the most reliable fruit commodities. The problem that is often found is the lack of accuracy and lack of knowledge of employees in distinguishing the types, quality and ripeness of bananas, especially new employees. Jaringa Saraf Tiruan (Neural Network) are used as a method in the classification process. The dataset in this study is a picture of bananas with 9 types, namely Ambon banana, plantain, Cavendish banana, Kirana banana, Barangan banana, jackfruit banana, gold banana and kapok banana. The ripeness of bananas in this study were the raw, ripe and overripe levels. The program is created using Tensorflow Python, the test results produce an accuracy level of 98,7 %

Actions (login required)

View Item