Page 74 - IJEEE-2023-Vol19-ISSUE-1
P. 74
Received: 09 September 2022 Revised: 02 October 2022 Accepted: 02 October 2022
DOI: 10.37917/ijeee.19.1.9
Vol. 19| Issue 1| June 2023
Ð Open Access
Iraqi Journal for Electrical and Electronic Engineering
Original Article
Deep learning and IoT for Monitoring Tomato Plant
Marwa Abdulla*1, Ali Marhoon,2
1 Computer Science, College of Education, University of Basrah, Basrah, Iraq
2 Electrical Engineering Department, College of Engineering, University of Basrah, Basrah, Iraq
Correspondence
* Marwa Abdulla
Computer Science, College of Education,
University of Basrah, Basrah, Iraq
Email: cepsm510006@avicenna.uobasrah.edu.iq
Abstract
Agriculture is the primary food source for humans and livestock in the world and the primary source for the economy of many
countries. The majority of the country's population and the world depend on agriculture. Still, at present, farmers are facing
difficulty in dealing with the requirements of agriculture. Due to many reasons, including different and extreme weather
conditions, the abundance of water quality, etc. This paper applied the Internet of Things and deep learning system to establish
a smart farming system to monitor the environmental conditions that affect tomato plants using a mobile phone. Through deep
learning networks, trained the dataset taken from PlantVillage and collected from google images to classify tomato diseases,
and obtained a test accuracy of 97%, which led to the publication of the model to the mobile application for classification for
its high accuracy. Using the IoT, a monitoring system and automatic irrigation were built that were controlled through the
mobile remote to monitor the environmental conditions surrounding the plant, such as air temperature and humidity, soil
moisture, water quality, and carbon dioxide gas percentage. The designed system has proven its efficiency when tested in terms
of disease classification, remote irrigation, and monitoring of the environmental conditions surrounding the plant. And giving
alerts when the values of the sensors exceed the minimum or higher values causing damage to the plant. The farmer can take
the appropriate action at the right time to prevent any damage to the plant and thus obtain a high-quality product.
KEYWORDS: Deep learning, IoT, Smart farming system, mobile application, Remote Irrigation, plant monitoring, tomato
disease.
I. INTRODUCTION humidity, precipitation, wind speed, soil content, and others,
by using sensors [4]. Thus, this data is used to automate
Agriculture considered the most critical development in farming techniques to make sound decisions, reduce waste,
human civilization, as people began farming thousands of increase yields, and reduce the effort to manage crops.
years ago. In the ancient culture of Mesopotamia, agriculture Therefore, the IoT in agriculture is a research topic that needs
was the country's main activity, and the reason for its to be developed, especially in establishing systems to
prosperity was the abundance of fresh water. And agriculture monitor climatic conditions that affect plant health,
continued to flourish like this. In 1979, Iraq was self- automating farming techniques, and reducing the challenges
sufficient in many crops, including wheat [1]. But at the facing this process [5]. Several methods have been proposed
moment our country, Iraq, suffers from a decline in for using sensors to measure and monitor environmental
agricultural production of all kinds in general due to several factors, in addition to using DL algorithms in predicting,
reasons, including the little knowledge of the farmer to the classifying, and counting crops in agricultural fields [6], [7].
plant diseases, accommodation on agricultural land, lack of So IoT is involved in the agricultural field in many areas,
seeds, fertilizers, water, modern agricultural machinery, etc. including Monitoring the Climatic [8], [9], automatic
[2]. We can provide farmers with a system based on the irrigation of the soil to reduce water waste [10], Soil
Internet of Things (IoT) and Deep Learning (DL) to help chemical properties [11]. Drones with Sensors and Cameras
them protect their farms. Introducing DL in the agricultural It is possible to draw maps, photograph, and survey
field is an essential modern research field. And this field agricultural lands [12], As for deep learning, it has entered
includes sub-fields that still need research and development, into many fields, including classification of plant diseases,
such as classifying viral and fungal plant diseases, counting pest identification, fruit counting, etc. [13], [14]. So in this
fruits of all kinds, predicting the date of harvesting fruits, paper we focus on the tomato crop is one of the most
knowing the type of plant, and others [3]. The applications important crops used in abundance locally and globally, and
of IoT in the agricultural field revolve around collecting we have recently noticed the lack and poor quality of its local
environmental data that affect plants, including temperature,
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and
reproduction in any medium, provided the original work is properly cited.
© 2023 The Authors. Published by Iraqi Journal for Electrical and Electronic Engineering by College of Engineering, University of Basrah.
https://doi.org/10.37917/ijeee.19.1.9 https://www.ijeee.edu.iq 70