Detection of the Existence Rhodamine B in Chili Paste with Digital Image Processing

DOI

10.2991/absr.k.220305.042How to use a DOI?

Keywords

Artificial Neural Network; Chili Paste; Digital Image Processing; Matlab; Rhodamine B

Abstract

The need for chili paste is increasing in line with the increasing variety of types and menus of dishes that use chili paste. To improve the colour of chili paste some traders use the textile dye Rhodamine B which causes significant health risks. The purpose of this study is to develop an image processing system that can detect the existence of Rhodamine B in chili paste quickly and practically using image processing and artificial neural networks (ANN). Image data retrieval using digital image acquisition techniques with the help of a webcam which is placed on the image capture toolbox and a computer. The number of samples used were 172 training data, 16 validation data, and 12 test data consisting of chili paste with the addition of various variations in the amount of Rhodamine B, namely 0 gram; 0.25 gram; 0.5 gram; 1 gram of Rhodamine B in 1 kg of chili paste. The data that has been extracted from the colour and texture features are then further processed and trained with an artificial neural network using the backpropagation method in order to detect the existence of Rhodamine B in chili paste properly. The resulting test system was then tested on 20 market samples taken from 4 different markets. The results of this study indicate that the parameters used to detect the existence of Rhodamine B in chili paste are parameters a, green, L, and b. The structure of the artificial neural network system consists of 4 input layers, 24 hidden layers, and 2 output layers. The GUI (Graphical User Interface) can detect the existence of Rhodamine B in chili paste with variations in the amount of Rhodamine B with the accuracy validation of the ANN system in detecting Rhodamine B in chili paste samples by 100% and testing the ANN system by 83%. Testing market samples with the ANN system gets an accuracy of 50%, the accuracy of the market samples is because the materials used are not controlled or heterogeneous.

Copyright

© 2022 The Authors. Published by Atlantis Press International B.V.

Open Access

This is an open access article under the CC BY-NC license.

Download article (PDF)

TY  - CONF
AU  - Ihsan Ali Suwarno
AU  - Nafis Khuriyati
AU  - Makhmudun Ainuri
PY  - 2022
DA  - 2022/03/10
TI  - Detection of the Existence Rhodamine B in Chili Paste with Digital Image Processing
BT  - Proceedings of the 2nd International Conference on Smart and Innovative Agriculture (ICoSIA 2021)
PB  - Atlantis Press
SP  - 278
EP  - 283
SN  - 2468-5747
UR  - https://doi.org/10.2991/absr.k.220305.042
DO  - 10.2991/absr.k.220305.042
ID  - Suwarno2022
ER  -