Article Per Year (5 Year)
p-Index From 2019 - 2024
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Qurrotul A'yun
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

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

Found 1 Documents
Search

 1 
Rancang Bangun Deteksi Kemanisan Buah Semangka menggunakan Metode Gray Level Co-Occurrence Matrix dan Backpropagation Neural Network berbasis Raspberry Pi Qurrotul A'yun; Fitri Utaminingrum
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 6 No 2 (2022): Februari 2022
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Watermelon in Latin Citrullus lanatus is a fruit with green skin characteristics with red or yellow arrays and flesh color. Watermelon is a type of fruit favored by the people of Indonesia. The high market demand for watermelon must be balanced with watermelon production. which continues to increase as well. Watermelon that has been harvested needs handling to classify the quality of watermelon on the market, one of which is the classification of watermelon sweetness. Measuring the size of the sweetness of watermelons can be destructively using the Brix refractometer, but this is not practical because you have to split the fruit first. Therefore, an innovation is needed to detect watermelon sweetness with digital image processing. In this study the sweetness of watermelon will be divided into 3 classes, namely low, average and high. This study uses the Gray Level Co-Occurrence Matrix method for feature extraction using 6 features, namely Dissimilarity, Homogeneity, Contrast, Correlation, Energy, and Angular Second Moment (ASM) with values ​​of d= 1, 2, 3 and angle = 0°. , 45°, 90°, and 135°. For sweetness class classification using Backpropagation Neural Network method. In Epoch and learning rate testing, the best training accuracy is 86% at Epoch 12,000 and learning rate is 0.01 with values ​​d=1 and = 0°. Then the best value is used in the system integration test and the highest accuracy is obtained, namely 85.7% at a distance of 15cm and the average computation time required is 10.05997 seconds.
Co-Authors Utaminingrum, Fitri