This project focuses on building a Convolutional Neural Network (CNN) to classify plant diseases from leaf images. Early and accurate detection is essential for boosting agricultural productivity and sustainability. The trained model leverages image data to recognize various plant diseases, enabling timely and informed decision-making for farmers and agronomists.
Plant diseases severely affect agricultural yield. Traditional detection involves manual inspection, which can be slow and error-prone. This project aims to automate disease detection using a CNN that classifies images of plant leaves into healthy or diseased categories.
The dataset consists of healthy and diseased plant leaf images, with multiple classes corresponding to specific diseases.
🔗 Download the dataset - PlantVillage
-
Prepare the Dataset
Structure your dataset in the required format. -
Train the Model
Run the training script to train the CNN. -
Evaluate the Model
Use the test dataset to evaluate model accuracy and performance. -
Classify New Images
Load the trained model and classify new leaf images.
The CNN consists of:
- Multiple convolutional layers
- Pooling layers to reduce dimensionality
- Fully connected layers for classification
🧩 The model processes leaf images to identify and categorize diseases with high precision.
📦 Download the trained model
The CNN is trained using supervised learning techniques. Adjustable parameters:
- Learning rate
- Batch size
- Number of epochs
Training incorporates data augmentation to enhance model generalization and performance.
The model is evaluated using metrics such as:
- Accuracy
- Precision
- Recall
- F1-Score
Additional visualizations like confusion matrix and ROC curves are used to analyze results.
- Python
- TensorFlow / Keras
- Flask
- HTML, CSS
- Matplotlib / Seaborn
A deep learning-based solution for early plant disease detection to support precision agriculture.
📍 Feel free to fork, star, or contribute!