This project focuses on efficient indexing of aerial images. The primary objective is to develop and implement algorithms that can index a large collection of aerial images, enabling quick and accurate searches based on image content. This project leverages ResNet, a powerful convolutional neural network, for feature extraction from images, and FAISS (Facebook AI Similarity Search) for efficient indexing and retrieval.
To install and run this project, follow these steps:
- Clone the repository:
git clone https://github.com/bianca-ghx/aerial-image-indexing.git
- Navigate to the project directory:
cd aerial-image-indexing
- Ensure you have Python and the required libraries installed.
To use the aerial image indexing system, follow these instructions:
- Prepare your dataset by placing the aerial images in the appropriate directories as specified in the project.
- Open and run the
uc-merced-indexing.ipynb
notebook to preprocess the images, build the index, and perform retrieval tasks.
This project was developed for the Big Data Mining course at POLITEHNICA Bucharest National University of Science and Technology explore aerial image processing and indexing.