Project Overview
The AI-Based Tomato Plant Disease Detection System is a smart agriculture project that combines Artificial Intelligence (AI), Computer Vision, Internet of Things (IoT), and Edge Computing to automatically detect diseases in tomato plant leaves.
 
The system captures leaf images using a Raspberry Pi Camera Module and performs real-time disease classification using a TensorFlow Lite deep learning model running directly on Raspberry Pi. Environmental parameters such as temperature, humidity, and soil moisture are also monitored and displayed on a web dashboard.
 
The project works completely offline and does not require cloud connectivity for disease prediction.
 
Key Features
Real-time Tomato Disease Detection
Raspberry Pi Camera Integration
TensorFlow Lite AI Model
Edge AI Processing (Offline Inference)
Live Camera Streaming
Image Capture & Prediction
Temperature Monitoring (DHT11)
Humidity Monitoring (DHT11)
Soil Moisture Monitoring
Flask-Based Web Dashboard
Mobile & Laptop Accessible Dashboard
Local Network Access
Confidence Score Display
Real-Time Sensor Monitoring
Lightweight Raspberry Pi Deployment
 
Technologies Used
 
Hardware -
Raspberry Pi 4 Model B
Raspberry Pi Camera Module
DHT11 Temperature & Humidity Sensor
Soil Moisture Sensor
Jumper Wires
Power Supply
 
Software -
Python
TensorFlow Lite
MobileNetV2 Transfer Learning
Flask
OpenCV
Picamera2
NumPy
HTML
CSS
 
AI Model Information
 
Dataset Used - PlantVillage Dataset
Training Method - Transfer Learning using MobileNetV2
Deployment - TensorFlow Lite Model on Raspberry Pi
Validation Accuracy - Approximately 93%
Additional Testing - Model tested on:
PlantVillage Validation Images
Internet Images
Real-World Tomato Leaf Images
 
System Workflow
Capture Tomato Leaf Image
Preprocess Image
Run TensorFlow Lite Inference
Predict Disease Class
Calculate Confidence Score
Read Sensor Values
Display Results on Dashboard
 
Dashboard Features
 
Live Camera Feed - View real-time camera stream directly from Raspberry Pi.
Disease Prediction - Displays:
Disease Name
Confidence Percentage
Environmental Monitoring - Displays:
Temperature
Humidity
Soil Moisture Status
Captured Image Preview - Shows captured image alongside prediction result.
 
Available Options - 
 
Option 1: DIY Kit: Perfect for students who want to build and understand the project themselves.
 
Includes -
✅ Raspberry Pi 4
✅ Raspberry Pi Camera Module
✅ DHT11 Sensor
✅ Soil Moisture Sensor
✅ Connecting Wires
✅ Complete Source Code
✅ Trained AI Model (.tflite)
✅ Circuit Diagram
✅ Project Report
✅ Block Diagram
✅ Flow Chart
✅ Installation Guide
✅ Setup Documentation
✅ Video Demonstration Guide
✅ Technical Support
 
User Needs To - 
Assemble Hardware
Connect Sensors
Upload Code
Run Project
 
Option 2: Ready-Made Project: Ideal for final year submissions, exhibitions, and project demonstrations.
 
Includes -
✅ Fully Assembled Hardware
✅ Preloaded Raspberry Pi OS
✅ Configured AI Model
✅ Working Dashboard
✅ Camera Setup
✅ Sensor Integration
✅ Complete Testing
✅ Project Report
✅ Circuit Diagram
✅ Block Diagram
✅ Flow Chart
✅ Source Code
✅ Demonstration Video
✅ Technical Support
 
User Only Needs To
Power ON Device
Connect to Local WiFi
Access Dashboard
Start Testing
 
Applications
Smart Agriculture
Precision Farming
Disease Monitoring
Crop Health Analysis
Agricultural Research
Educational Projects
Final Year Engineering Projects
AI and IoT Demonstrations
 
For any queries, reach out to us at -
Email - learnelectronicsindia.com@gmail.com
WhatsApp - 7600948607