VisionVocabClassifier

15-Scene Image Classification with SIFT and SVM

中文文档请戳

This repository contains the Python code for a machine learning project that classifies images from the 15-Scene dataset using Scale-Invariant Feature Transform (SIFT) features and Support Vector Machine (SVM) classifier. The project utilizes OpenCV for image processing, scikit-learn for clustering and classification, and joblib for caching.

Dataset

The dataset used is the 15-Scene Image Dataset, which can be downloaded from: 15-Scene Image Dataset

Installation

To run this project, you need Python 3.x and the following packages:

• OpenCV
• NumPy
• scikit-learn
• joblib
• tqdm

You can install the required packages using pip:

pip install numpy opencv-python scikit-learn joblib tqdm

Usage

1. First, clone the repository to your local machine:

   git clone [email protected]:hammershock/VisionVocabClassifier.git
   cd VisionVocabClassifier

2. Download the 15-Scene Image Dataset and extract it into the project directory under ./15-Scene Image Dataset.

3. Run the main script:

   python main.py

This will execute the data loading, feature extraction, training, and evaluation sequence. Results including the model's accuracy and confusion matrix will be displayed in the console.

Features

• Data Loading: Automatically loads and splits the dataset.
• Feature Extraction: Extracts SIFT descriptors from images.
• Clustering: Uses MiniBatchKMeans for clustering descriptors into visual words.
• Histogram Generation: Builds histogram features from clustered descriptors.
• Classification: Trains an SVM classifier with RBF kernel.
• Evaluation: Computes accuracy and displays a confusion matrix.

Contributing

Contributions to this project are welcome. Please fork the repository and submit a pull request with your changes.

License

This project is open-sourced under the MIT License. See the LICENSE file for more details.