Project Overview
This project implements a convolutional neural network (CNN) for image classification. The model was trained on a dataset of 10,000 images across 10 categories and achieves an accuracy of 95% on the test set.
Technical Details
The model architecture consists of:
- 3 convolutional layers with max pooling
- 2 fully connected layers
- Dropout for regularization
- Softmax activation for classification
Data Processing
The training data underwent several preprocessing steps:
- Normalization
- Data augmentation (rotation, zoom, flip)
- Train/validation/test split (70%/15%/15%)
Results
The final model achieved the following metrics:
- Training accuracy: 97%
- Validation accuracy: 95%
- Test accuracy: 95%
- F1-score: 0.94
Sample Predictions
Examples of correctly classified images
Future Work
- Implement transfer learning with pre-trained models
- Create a web interface for real-time predictions
- Extend the model to handle video classification