This project intends to implement a Content-based Image Retrieval (CBIR) system used for image classification and object detection. The aim of this task is to compare several pre-trained CNN architectures to be fine-tuned with several dataset and to explore its capabilities with differences in spatial resolution.

The CBIR is designed to be composed to include feature representation, feature indexing and feature similarity matching. The framework begins by fine-tuning remote sensing datasets on EfficientNet and ResNet, that are used to extract features. In addtion, these features are stored in an image database, known as KD-tree, that is used to implement search and retrieval when a query image comes in place.