Detection Computer Vision Project

This dataset is part of a graduation project focused on enhancing the tourism experience in Egypt by creating an AI-powered tool for identifying famous Egyptian landmarks. Our goal is to provide tourists with an easy and efficient way to recognize and learn about landmarks through image recognition.

The dataset includes high-quality, labeled images of various landmarks across Egypt, curated to ensure diverse environmental conditions, angles, and lighting scenarios. These images are annotated with bounding boxes to train and evaluate landmark detection models. The final model will serve as the backbone of a landmark detection tool accessible via a user-friendly website, making it easy for tourists to get information about landmarks in real time, thus overcoming language and informational barriers.

This project utilizes the YOLO v7 model for initial deployment, with planned future improvements using YOLO NAS. The dataset is meticulously labeled with Roboflow to ensure accurate annotations, and it will continue to evolve as we collect more diverse images to improve model accuracy.

Applications:
The resulting landmark detection model can be used in travel apps, tourism websites, and other educational resources, offering a modernized approach to tourism guidance in Egypt.