Muhammad Fadhil Sadeli
Fakultas Ilmu Komputer, Universitas Brawijaya

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

Found 1 Documents
Search

Rancang Bangun Sistem Portabel untuk Klasifikasi Cendol Merah Mengandung Rhodamin B menggunakan Metode Jaringan Syaraf Tiruan Muhammad Fadhil Sadeli; Dahnial Syauqy; Edita Rosana Widasari
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 6 No 6 (2022): Juni 2022
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

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

Abstract

Cendol is a traditional West Javanese drink made from hunkwe flour or mung bean flour. Cendol that is often found is green cendol. However, there are also cendol sellers who use red cendol derived from food coloring or agar powder. However, there are still cendol sellers who add synthetic dyes containing Rhodamin B to cendol as a dye. Because the price is relatively cheap and makes the color more striking so that buyers become interested in buying the cendol. Rhodamine B is a synthetic dye in the form of a crystalline powder, green in color, odorless, and fluoresces in a bright red solution. Rhodamine B is very dangerous if consumed and inhaled which can cause liver function disorders, cancer, irritation of the respiratory tract, skin, and eyes. The misuse of these dyes occurs due to a lack of public knowledge about how to distinguish food coloring from Rhodamine B dye and the dangers of its use. Therefore, the researchers designed a system that can classify cendol containing Rhodamine B based on color. The system is built with a portable design for efficiency and portability. This system uses a power bank as a resource, then uses a TCS3200 sensor to determine the RGB value of the cendol color, a 16x2 LCD with I2C to display the output and classification results of the system, Arduino uno as a microcontroller to process data and calculate classifications, and an Artificial Neural Network (ANN). as a classification method. This system utilizes 50 sets of training data, 25 sets of test data for the ANN method, and 15 sets of test data for the whole system. Based on the results of the tests carried out, the accuracy of the Artificial Neural Network method was 96.08% with an average computation time of 58 ms and an overall system accuracy of 93.34%.