Accenture AIR Computer Vision Project

AIR: Aerial Inspection RetinaNet for supporting Land Search and Rescue Missions

From the Original Author:

AIR is a deep learning based object detection solution to automate the aerial drone footage inspection task frequently carried out during search and rescue (SAR) operations with drone units. It provides a fast, convenient and reliable way to augment aerial, high-resolution image inspection for clues about human presence by highlighting relevant image regions with bounding boxes, as done in the image below. With the assistance of AIR, SAR missions with aerial drone searches can likely be carried out much faster before, and with considerably higher success rate.

This code repository is based on the master's thesis work by Pasi Pyrrö from Aalto University, School of Science which was funded by Accenture.

Dataset Source:

This dataset is obtained from the listing in Robin Cole's satellite-image-deep-learning GitHub repository

• https://www.satellite-image-deep-learning.com/
• https://twitter.com/robmarkcole

AIR -> A deep learning object detector framework written in Python for supporting Land Search and Rescue Missions

• GitHub Repository
• https://issueantenna.com/repo/Accenture/AIR

The HERIDAL dataset (after conversion to keras-retinanet PASCAL VOC XML format) is used to train the AIR model.

• Link to dataset: https://zenodo.org/record/5662351#.Y6aA9-zMIeY

Original Citation:

@MastersThesis{pyrro2021air, title={{AIR:} {Aerial} Inspection RetinaNet for Land Search and Rescue Missions}, author={Pyrr{"o}, Pasi and Naseri, Hassan and Jung, Alexander}, school={Aalto University, School of Science}, year={2021} }