An AI-powered desktop application that detects plant leaf diseases from either uploaded images or real-time webcam feed using a custom-trained VGG16 Convolutional Neural Network (CNN) model.
It provides a detailed diagnosis, including the disease name, confidence level, and a suggested cure/solution, all displayed in an intuitive Tkinter-based GUI.
The application also sends instant Telegram alerts containing the predicted disease along with the image.
This model supports two modes of disease detection:
- Real-time Detection: Using a webcam or external camera (e.g., phone camera via Wi-Fi)
- Image Upload Detection: Upload any leaf image for prediction
In image upload mode, the system provides the disease prediction, confidence score, and also reads a text file containing the cause and solution from the
solutions/folder.
For real-time detection, it can use either your laptop webcam or a phone camera over Wi-Fi (IP Webcam supported) for remote analysis.
- Detect plant diseases from images or webcam feed
- Built on a pre-trained VGG16 deep learning model
- Sends automatic Telegram alerts (with message + photo)
- Displays predicted disease, confidence, and recommended cure
- Simple and elegant Tkinter GUI
- Works with Python 3.6 – 3.10
| 📤 Upload Image | 🎥 Real-time Detection |
|---|---|
 |
 |
git clone https://github.com/your-username/plant-disease-detector.git
cd plant-disease-detectorUse the VS Code terminal or command line
pip install opencv-python==4.8.0.76 numpy==1.24.4 pillow requests tensorflow telepot
python tr.pyTelegram Notification Setup Get instant alerts with predictions via Telegram (message + photo).
Steps: Go to @BotFather on Telegram and create a new bot. Copy the BOT_TOKEN provided.
Get your CHAT_ID by sending a message to your bot, then visit: https://api.telegram.org/bot/getUpdates
Replace the values in tr.py: BOT_TOKEN = "your-bot-token" CHAT_ID = "your-chat-id"
If you need the complete project package including:
- Final Report / Synopsis
- Experiment Observations
- Accuracy Graphs / Confusion Matrix
- PPT / Documentation
Feel free to contact me directly. I’ll be happy to help!