Plant Disease Detection

System

Submitted To: Mr. Pradeep Kandulna

B.Tech 7th Semester Final Year Project

Team Members:
1. Ganupati Thirumala Reddy - 21052760
2. Nomula Srimlika -21051911
3. Nancy Bakhla - 2105472
SCHOOL OF COMPUTER ENGINEERING KALINGA INSTITUTE OF
INDUSTRIAL TECHNOLOGY 4. Harshita Olive Arohan - 2105718
BHUBANESWAR, ODISHA - 751024
5. Reddy Swamy - 2105727
 Introduction
● P l a n t d i s e a s e s s i g n i f i c a n t l y t h r e a t e n g l o b a l a g r i c u l t u r a l p r o d u c t i v i t y, l e a d i n g t o
reduced crop yields and economic losses.

● Using a dataset of annotated plant leaf images, the system employs

Convolutional Neural Networks (CNNs) to detect and classify diseases
a c c u r a t e l y. P r e p r o c e s s i n g s t e p s s u c h a s r e s i z i n g , n o r m a l i z a t i o n , a n d
augmentation enhance the model's training process.

● The final model achieves high accuracy and precision in distinguishing

b e t w e e n h e a l t h y a n d d i s e a s e d l e a v e s , o ff e r i n g a s c a l a b l e , e ff i c i e n t , a n d
automated solution for farmers and agricultural experts.

● By integrating this system into farming practices, stakeholders can achieve

early detection and prevention of plant diseases, ensuring sustainable
a g r i c u l t u r e a n d f o o d s e c u r i t y.

● It provides a technological solution to enhance agricultural productivity by

d e t e c t i n g d i s e a s e s e a r l y.
 Problem Statement :
“Manual detection of plant diseases is time-intensive, prone to errors, and often leads to
delayed interventions, affecting agricultural productivity. ”

● Farmers and agricultural professionals often face significant

challenges in identifying plant diseases at an early stage. Manual
inspection methods require expertise, are susceptible to human error,
and can lead to delays in diagnosis, further exacerbating crop damage.

● Additionally, many regions lack access to skilled personnel or the

resources required for regular monitoring, making it difficult to mitigate
the impact of plant diseases effectively.
 Objective:
“Develop an AI-based system using image processing and machine learning techniques
to identify and classify plant diseases accurately. “

● This project aims to bridge this gap by developing an automated, AI-

driven solution that can detect and classify plant diseases with high
accuracy.

● The proposed system addresses the need for a scalable, cost-

effective, and accessible tool to support sustainable farming practices
and reduce crop losses.
 Dataset and Preprocessing

● The dataset comprises images of healthy and diseased plants.

● Preprocessing steps include:

- Image resizing and normalization

- Data augmentation to enhance model robustness

Requirements : [with versions]

• tensorflow==2.10.0
• scikit-learn==1.3.0
• numpy==1.24.3
• matplotlib==3.7.2
• seaborn==0.13.0
• pandas==2.1.0
• streamlit
• librosa==0.10.1
 Model Selection
● We e m p l o y e d a C o n v o l u t i o n a l N e u r a l N e t w o r k ( C N N ) f o r i t s e ff i c i e n c y i n i m a g e
classification tasks. CNNs are a proven architecture for plant disease
detection tasks in research and real-world applications.

Q: Why Convolutional Neural Networks (CNNs)?

- S u i t a b i l i t y f o r I m a g e D a t a : C N N s a r e h i g h l y e ff e c t i v e i n h a n d l i n g i m a g e d a t a d u e
to their ability to extract spatial features and patterns.

Key Features:
• Convolutional Layers:
Detect patterns such as
edges, textures, and
shapes in leaf images.
• Pooling Layers: Reduce
dimensionality while
preserving important
features, improving
computational
efficiency.
[sourced from Amara et al. 2017]
 System Workflow
● The workflow consists of:

1. Input: Plant leaf image

2. Preprocessing: Image cleaning and enhancement

3. Prediction: Model identifies the disease

Training and Validation

Training involves:
- optimizing the model on the dataset with hyperparameter tuning.

Validation ensures:
- the model's accuracy and generalization.
 Results
Test results:
Model Prediction:
Demo Video:
 Challenges Faced
1. Dat as et imbalance w ith few er examples of rare diseases.

● - Da tas et imbalance occurs when there are significantly f ewer examples of

r ar e p la nt diseases com pared to comm on ones.

● - Th is u neven distribut ion causes the m odel t o favor well-represented

c la ss es , m aking it less effective at identifying rare diseases.

● - A s a r esult, the model st ruggles to generalize and often m isclassif ies or

i gno re s these minority classes, reducing it s overall accuracy and
r el ia bil it y.

2. L imit ed computational resources for training deep models.

● - De ep l earning m odels like CN Ns require substant ial com putational power

fo r t ra in ing, especially w ith large datasets.

● - In re source-limited settings, t he lack of GP Us or TP Us leads to slow er

tr ai ni ng tim es and restricts experim ent ation w ith advanced architectures.

● - Th is m akes it challenging to optimize the model effectively, often

r es ul tin g in com prom ised performance.
 Applications
1. Smart agriculture for real-time disease monitoring :
● -Using IoT-enabled devices such as drones or smartphones, farmers
can capture leaf images directly from the field and receive instant
disease diagnoses.
● - This reduces the spread of diseases and minimizing crop losses .
● - By automating disease detection, the system enhances efficiency,
reduces the dependency on manual inspections, and supports
precision agriculture for improved yield and sustainability.
2. Research for understanding plant pathology :
● -Provides a reliable tool and method to identify and classify
diseases.
● - Helps in studying disease patterns, identifying factors influencing
disease spread, and evaluating the effectiveness of treatments.
● - The system's datasets and machine learning insights can
contribute to developing new strategies for disease prevention and
management.
 Future Scope
Integrating IoT sensors Expanding the dataset to Integration with Mobile
for real-time data include more plant Applications
species and diseases

• Improved Monitoring: • Wider Application: A • User-Friendly Access: A

IoT sensors provide diverse dataset makes mobile app allows
real-time the system useful for farmers to capture leaf
environmental data, various crops and images and receive
enhancing disease regions, supporting instant disease
predictions and global agriculture. diagnoses, making the
enabling timely • Higher Accuracy: More system accessible and
interventions. data improves the easy to use.
• Scalability: Facilitates model’s ability to detect • Data-Driven Insights:
deployment over large rare diseases, ensuring The app can store
areas using drones or reliable and precise historical data and
automated devices, results. provide trends, helping
ensuring efficient, farmers track crop
continuous monitoring. health over time and
make informed
decisions.
 Conclusion
● Effective Disease Detection

The plant disease detection system demonstrates high accuracy in

identifying and classifying diseases, making it a valuable tool for modern
agriculture.

● Automation and Efficiency

By replacing manual inspection with AI-driven analysis, the system

reduces time and labor costs while improving reliability.

● Scalability and Versatility

Its adaptability for various crops and integration with smart farming
technologies ensures broad usability in diverse agricultural practices.

"Empowering agriculture through technology is the key to a greener, healthier

future."
 Refrences & Bibliography
● Sladojevic, S., et al. (2016). Deep Neural Networks Based Recognition of Plant Diseases
by Leaf Image Classification.

● M o h a n t y, S . P. , e t a l . ( 2 0 1 6 ) . U s i n g D e e p L e a r n i n g f o r I m a g e - B a s e d P l a n t D i s e a s e
Detection.

● Amara, J., et al. (2017). A Deep Learning-Based Approach for Banana Leaf Diseases
Classification.

● K a g g l e . P u b l i c D a t a s e t s f o r M a c h i n e L e a r n i n g . R e t r i e v e d f r o m h t t p s : / / w w w. k a g g l e . c o m .

● Te n s o r F l o w. D e e p L e a r n i n g F r a m e w o r k D o c u m e n t a t i o n . R e t r i e v e d f r o m
h t t p s : / / w w w. t e n s o r f l o w. o r g .

We extend our gratitude to our guide, team members, and the resources
that helped us complete this project .

Group Members: 1. Ganupati Thirumala Reddy - 21052760

2. Nomula Srimlika -21051911
3. Nancy Bakhla - 2105472
4. Harshita Olive Arohan - 2105718
5. Reddy Swamy - 2105727