🍚 Rice Image Classification with CNNs using PyTorch

📝 Project Overview

This repository contains a deep learning project aimed at classifying different rice grain varieties using a Convolutional Neural Network (CNN) built with PyTorch. The dataset includes five rice categories: Arborio, Basmati, Jasmine, Ipsala, and Karacadag.

The project demonstrates:

• Building and training a custom CNN model using PyTorch.
• Analyzing class distribution and data preprocessing techniques.
• Evaluating the model's performance with accuracy, loss metrics, and a confusion matrix.

📂 Dataset

The Rice Image Dataset used in this project contains 75,000 images, evenly distributed across five classes:

• Arborio: 15,000 images
• Basmati: 15,000 images
• Jasmine: 15,000 images
• Ipsala: 15,000 images
• Karacadag: 15,000 images

The dataset was split into 75% training and 25% validation using a 75/25 ratio, resulting in:

• Training set: 56,250 images
• Validation set: 18,750 images

⚙️ Features

• Custom CNN Architecture: A simple yet effective model with convolutional, pooling, and fully connected layers.
• Data Augmentation: Using techniques like random horizontal flipping for better generalization.
• Performance Metrics: Tracking training accuracy, validation accuracy, and loss over epochs.
• Visualization: Plotting accuracy, loss curves, and confusion matrix for detailed analysis.

📊 Results

• The custom CNN model achieved 99.73% training accuracy and 99.56% validation accuracy after 15 epochs.
• Below are visualizations for the model's performance:

🛠️ Model Architecture

The CNN model consists of:

1. Two convolutional layers with ReLU activations and batch normalization.
2. Max pooling for downsampling.
3. Fully connected layers for classification.

class CNNModel(nn.Module):
    def __init__(self, num_classes=5):
        super(CNNModel, self).__init__()
        self.conv1 = nn.Conv2d(3, 128, kernel_size=3, padding=1)
        self.conv2 = nn.Conv2d(128, 64, kernel_size=3, padding=1)
        self.pool = nn.MaxPool2d(2, 2)
        self.fc1 = nn.Linear(64 * 62 * 62, 128)
        self.fc2 = nn.Linear(128, 32)
        self.fc3 = nn.Linear(32, num_classes)

    def forward(self, x):
        x = self.pool(F.relu(self.conv1(x)))
        x = self.pool(F.relu(self.conv2(x)))
        x = x.view(x.size(0), -1)
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x

📈 Performance Metrics

Metric	Value
Training Accuracy	99.73%
Validation Accuracy	99.56%
Training Loss	0.86

🤝 Contributing

Contributions are welcome! Feel free to submit a Pull Request if you want to enhance the model or add new features.

📧 Contact

Developed by Hamed Mahmoudi.

If you have any questions or suggestions, please feel free to reach out!