Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vol 6 No 6 (2022): Juni 2022

Prediksi Bobot Segar pada Tanaman Hidroponik berdasarkan Kondisi Daun menggunakan Metode Pengolahan Citra Digital dan Jaringan Syaraf Tiruan

Mohamad Abyan Naufal Fachly (Fakultas Ilmu Komputer, Universitas Brawijaya)
Hurriyatul Fitriyah (Fakultas Ilmu Komputer, Universitas Brawijaya)
Rizal Maulana (Fakultas Ilmu Komputer, Universitas Brawijaya)



Article Info

Publish Date
21 Jul 2022

Abstract

Based on data from the Central Statistics Agency, the production of mustard plants in Indonesia, especially in East Java Province in 2020 reached 77,176 tons. Where this production increased compared to the production in 2019 which reached 74,395 tons . Mustard plants in Indonesia are cultivated using several techniques, one of the techniques is hydroponics. Hydroponics is a system of cultivating plants using water that contains nutrients and minerals without soil. Plants produced from hydroponics have better quality than conventionally grown plants and also hydroponic plants can prevent plants from pests. Besides having advantages, hydroponic cultivation of mustard plants also has disadvantages such as the amount of electricity used, water discharge, solution concentration, and moss which can affect plant conditions like fresh weight. Fresh weight is one indicator of plant growth, where if the plant growth is good, the fresh weight produced will be high. Therefore, researchers want to predict the condition of hydroponic plants based on their fresh weight using digital image processing methods and artificial neural networks. This system works using a RaspberryPi as hardware that is connected to a webcam. Later the camera will capture the image of the leaves from the hydroponic mustard plant, then provide output in the form of fresh weight information on a 16x2 LCD. Based on the tests carried out on the system using 10 mustard plants to predict the fresh weight of the plant, it resulted in an an average MAPE of 0.67%. In addition , two computational time tests were also carried out , which consisted of the computation time on the system and the computation time for the Artificial Neural Network method . Where the computational time on the system produces an average time of 4.884 seconds and the computation time of the artificial neural network method with an average time of 0.192 seconds.

Copyrights © 2022






Journal Info

Abbrev

j-ptiik

Publisher

Subject

Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Description

Jurnal Pengembangan Teknlogi Informasi dan Ilmu Komputer (J-PTIIK) Universitas Brawijaya merupakan jurnal keilmuan dibidang komputer yang memuat tulisan ilmiah hasil dari penelitian mahasiswa-mahasiswa Fakultas Ilmu Komputer Universitas Brawijaya. Jurnal ini diharapkan dapat mengembangkan penelitian ...