EAI Endorsed Transactions
on Pervasive Health and Technology Research Article
1
Optimized Deep Learning Model for Disease Prediction
in Potato Leaves
Virendra Kumar Shrivastava
1,*
, Chetan J Shelke
2
, Aastik Shrivastava
3
, Sachi Nandan Mohanty
4
and
Nonita Sharma
5
1, 2
Department of Computer Science and Engineering, Alliance College of Engineering and Design, Alliance University,
Bangalore, India
3
Siemens Healthneers, Bengaluru, India
4
School of Computer Science &Engineering, VIT-AP University Amaravati, India
5
Department of Information Technology, Indira Gandhi Delhi Technological University, New Delhi, India
Abstract
Food crops are important for nations and human survival. Potatoes are one of the most widely used foods globally. But there
are several diseases hampering potato growth and production as well. Traditional methods for diagnosing disease in potato
leaves are based on human observations and laboratory tests which is a cumbersome and time-consuming task. The new age
technologies such as artificial intelligence and deep learning can play a vital role in disease detection. This research proposed
an optimized deep learning model to predict potato leaf diseases. The model is trained on a collection of potato leaf image
datasets. The model is based on a deep convolutional neural network architecture which includes data augmentation, transfer
learning, and hyper-parameter tweaking used to optimize the proposed model. Results indicate that the optimized deep
convolutional neural network model has produced 99.22% prediction accuracy on Potato Disease Leaf Dataset.
Keywords: Deep Learning, Artificial Intelligence, Machine Learning, Deep Convolutional Neural Network, Optimized Deep
Convolutional Neural Network Model, Disease Prediction
Received on 19 June 2023, accepted on 02 September 2023, published on 27 September 2023
Copyright © 2023 V. K. Shrivastava et al., licensed to EAI. This is an open access article distributed under the terms of the CC BY-
NC-SA 4.0, which permits copying, redistributing, remixing, transformation, and building upon the material in any medium so long as
the original work is properly cited.
doi: 10.4108/eetpht.9.4001
1. Introduction
In recent years, food production has been heavily impacted
due to plant diseases. Plant diseases are caused by climate
change, adverse impact on the environment, heavy usage of
fertilizers and so on. Climate change has severely impacted
potato yield due to a variety of diseases. The most destructive
diseases in potato leaves are late blight and early blight.
These diseases have largely emerged in the last few years [1]
due to many reasons including climate change. The infections
that damage plants, starting in the leaves before spreading to
the entire plant, are the major causes of the yield decline in
*
Corresponding author. Email: vk_shrivastava@yahoo.com
potato production. Potatoes are a largely consumed food item
in the world. According to a report published in Statista, over
376 million metric tons of potatoes were produced in 2021
which is down 2% from 2020 crop [2]. Farmers heavily rely
on human inspection to identify potato leaves diseases which
are time consuming and have a high chance of error. In the
present technological era, the use of new age technologies
such as artificial intelligence (AI), deep learning, and
computer vision (CV) etc. are very advantageous to speed up
the potato disease prediction process. AI and deep learning
have witnessed immense surge in the agriculture domain due
to its capabilities of image identification, processing, image
classification and image prediction [3].
A kind of machine learning [4-6] called deep learning
(DL) has been demonstrated to be particularly good at
EAI Endorsed Transactions on
Pervasive Health and Technology
2023 | Volume 9