Biology Presentation Group 5

You might also like

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 20

Plant Disease

Detection Using
Machine Learning
Team Members:

● Sami Nayeem (011 193 020)


● Farhana Hossain (011 193 004)
● Maruf Ur Rahman (011 193 019)
● Nujhat Samiha (011 212 156 )
Table of Contents
1. Introduction
2. How Computer Science is Related In This Field
3. Related Works
4. Methodology
5. Basics of Classification
6. Dataset
7. Result Analysis
8. Conclusion
Introduction

Plant disease is defined as the state of local or systemic


abnormal physiological functioning of a plant, resulting
from the continuous, prolonged 'irritation' caused by
phytopathogenic organisms (infectious or biotic disease
agents).
How Computer Science Is Related In This
Field
As a result, we are living in a technological age when
Artificial Intelligence is sweeping the globe. The ailments of
all living creatures can be detected using this AI. This AI can
be utilized to gather information from plant photos and build
a dataset. The plant disease can then be classified using the
classification approach.
Related Works

Reference:

Ramesh, S., Hebbar, R.,


Niveditha, M., Pooja, R.,
Shashank, N., & Vinod, P. V.
(2018, April). Plant disease
detection using machine
learning. In 2018
International conference on
design innovations for 3Cs
compute communicate
control (ICDI3C) (pp. 41-
45). IEEE.
Related Works: Continued

Reference:

Bhise, N., Kathet, S.,


Jaiswar, S., & Adgaonkar, A.
(2020). Plant disease
detection using machine
learning. International
Research Journal of
Engineering and Technology
(IRJET), 7(7), 2924-2929.
Proposed Methodology
A neural network chain called Convolutional Neural Networks (CNN) is used to
classify a picture from a dataset. By converting the photos from the dataset to
numerical values, they are trained. Their primary benefit is the absence of
human oversight as they automatically recognize significant traits.

The following dataset, which includes a variety of photos of damaged leaves


from many different crops, will be used in the proposed work.

(a) (b) (c)


Proposed Methodology: Continued

Smartphones are growing in popularity as a


result of their super-fast processing, crystal-clear
displays, and excellent cameras. In this project, a
plant's name and degree of confidence are
provided together with an image that is taken
using an android app.
Proposed Methodology: Continued

Detection of disease is performed in two steps i.e. detection of the type of crop and
detection of type of disease. Convolutional Neural Network is used to do this.
Classification also functions as fully connected classifiers that are created utilizing a
variety of model learnings. This step essentially entails importing the leaf image from
the mobile device.The software includes a camera module that allows users to take
pictures of distinct groups. When there is a disease, it classifies them and the results
are given as what disease they are and the confidence of the classification. When the
leaf is healthy and there is no classification, the results are shown as healthy.
Workflow
Basics of Classification

Classification is a supervised machine learning


method where the model tries to predict the correct
label of a given input data.
Basics of Classification: Continued
Suppose you are showing a child the multiplication of zero (0).

5*0=0
10 * 0 = 0
15 * 0 = 0 Then you are asking the same child the
60 * 0 = 0 following question:
75 * 0 = 0
85 * 0 = 0 9*0=?

What will be the answer?


Yes the answer will
be 0.
Basics of Classification: Continued
Now let’s forward to a bit tricky part. Suppose there is a following dataset:

Favorite Song
Age Gender
Type

25 Male Metal

45 Female Classic Question: Age = 35,


Gender = Female,
19 Male Metal
Favorite Song Type = ??
26 Female Classic

20 Male Jazz

15 Female K-pop

30 Male Classic
Dataset
Dataset
Result Analysis

A neural network was trained to detect plant diseases such as


Corn Leaf Blight and Tomato Bacterial Spots by focusing
on visual features like blemishes and leaf shape.
However, low-resolution images, background noise, and images
with multiple classes resulted in inaccurate results.
Faster R-CNN with Inception Resnet V2 performed the best,
while MobileNet's performance decreased when pre-trained on
COCO+Plant Village compared to only COCO.
Conclusion

In order to correctly identify and categorize leaf illnesses, this project will propose a
revolutionary deep convolutional neural network model. This model can automatically
identify leaf disease distinguishing characteristics and enable a high-accuracy end-to-end
learning process.
Reference
1. Dataset: Davinder Singh, Naman Jain, Pranjali Jain, Pratik Kayal, Sudhakar Kumawat, and Nipun
Batra. 2020. PlantDoc: A Dataset for Visual Plant Disease Detection. In Proceedings of the 7th ACM
IKDD CoDS and 25th COMAD (CoDS COMAD 2020). Association for Computing Machinery, New
York, NY, USA, 249–253. https://doi.org/10.1145/3371158.3371196

2. Dataset (Kaggle):
https://www.kaggle.com/code/sumanismcse/plant-disease-detection-using-keras/notebook

3. Ramesh, S., Hebbar, R., Niveditha, M., Pooja, R., Shashank, N., & Vinod, P. V. (2018, April). Plant
disease detection using machine learning. In 2018 International conference on design innovations for
3Cs compute communicate control (ICDI3C) (pp. 41-45). IEEE.

4. Bhise, N., Kathet, S., Jaiswar, S., & Adgaonkar, A. (2020). Plant disease detection using machine
learning. International Research Journal of Engineering and Technology (IRJET), 7(7), 2924-2929.
Thank You

You might also like