A web-based application to detect paddy crop deaease and recommend suitable crop protection measures using a deep learning model (CNN). The application achieves high accuracy in identifying common paddy pests and suggests optimized agricultural practices to improve yield. It leverages Flask for deployment, Google Cloud Storage for model hosting, and real-time camera feed integration for instant analysis.
This project aims to assist farmers in detecting paddy crop pests and receiving tailored crop management recommendations. By utilizing a Convolutional Neural Network (CNN), the system identifies pests from captured images and classifies them into different categories. Additionally, the system provides recommendations on pest control measures and alternative crop strategies. The trained model is deployed using Django . The web application offers an intuitive user interface, allowing users to capture images using a real-time camera feed for instant analysis and feedback.
- Deep Learning Model: A CNN trained on a diverse dataset of paddy crop pests.
- Crop Recommendations: Provides suitable recommendations for pest control and alternative crops.
- Cloud Integration: Google Cloud Storage for model hosting and easy access.
- Web Interface: User-friendly interface built with Flask for seamless interaction.
We provide a training dataset of 10,407 (75%) labeled paddy leaf images across ten classes (nine diseases and normal leaf). Each image is accompanied by metadata, including the paddy variety and age. Your task is to develop an accurate disease classification model using this training dataset and classify each sample in the test dataset of 3,469 (25%) paddy leaf images into one of the ten categories.
- image_id - Unique image identifier corresponding to image file names (
.jpg) in thetrain_imagesdirectory. - label - Type of paddy disease (target class). There are ten categories, including normal leaf.
- variety - Name of the paddy variety.
- age - Age of the paddy in days.
3️⃣ train_images/ - Contains 10,407 training images stored in sub-directories corresponding to the ten target classes. File names match the image_id column in train.csv.
The dataset includes images categorized into the following ten classes:
- paddy_bacterial_leaf_blight
- paddy_bacterial_leaf_streak
- paddy_bacterial_panicle_blight
- paddy_blast
- paddy_brown_spot
- paddy_dead_heart
- paddy_downy_mildew
- paddy_hispa
- paddy_normal (Healthy leaves)
- paddy_tungro
📦 Paddy-disease-classification
Directory structure:
└── ai-hackathon-2025-scarecrow/
├── README.md
├── manage.py
├── requirement.txt
├── disease_detector/
│ ├── _init_.py
│ ├── admin.py
│ ├── apps.py
│ ├── crop_analyser.py
│ ├── image_classifier_code.py
│ ├── models.py
│ ├── tempCodeRunnerFile.py
│ ├── tests.py
│ ├── train.py
│ ├── views.py
│ ├── dataset/
│ │ └── Crop_recommendation.csv
│ ├── migrations/
│ │ └── _init_.py
│ ├── model_training/
│ │ └── crop-dataset.ipynb
│ └── models/
│ ├── crop_recommendation_model.pkl
│ ├── label_encoder.pkl
│ └── paddy_image_classifier.joblib
├── model_traning/
│ └── Paddy_CNN_clasification.ipynb
├── scarecrowApp/
│ ├── _init_.py
│ ├── asgi.py
│ ├── settings.py
│ ├── urls.py
│ └── wsgi.py
├── static/
│ ├── css/
│ │ └── style.css
│ └── js/
│ └── script.js
└── templates/
└── index.html
We would like to extend our heartfelt gratitude to everyone who contributed to this project. Your hard work and dedication made this possible!
|
Srujan Rana 🏆 Project Lead, Backend Developer |
Rudra Prasad Jena 💻 Frontend Developer & 🌐 API Integration |
Debasish Mohanty 💻 Frontend Developer |
🌟 Want to contribute?
We welcome contributions from the community! If you'd like to improve the project or report issues, feel free to fork the repo and submit a pull request.