Journal
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
Volume 14, Issue -, Pages 1497-1508Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/JSTARS.2020.3041316
Keywords
Aerial imagery; fusion; multimodal; object detection; remote sensing (RS)
Categories
Funding
- National Geospatial-Intelligence Agency [HM0476-19-1-2014]
- National Science Foundation [OAC-1 808 582]
- Air Force Office of Scientific Research [FA9550-20-1-0039]
Ask authors/readers for more resources
Deep-learning object detection methods designed for computer vision applications perform poorly on remote sensing data due to difficulties in collecting training data, small target sizes, and arbitrary perspective transformations. Fusion of data from multiple remote sensing modalities can improve detection performance. YOLOrs is a new convolutional neural network specifically designed for real-time object detection in multimodal remote sensing imagery, capable of detecting objects at multiple scales and predicting target orientations.
Deep-learning object detection methods that are designed for computer vision applications tend to underperform when applied to remote sensing data. This is because contrary to computer vision, in remote sensing, training data are harder to collect and targets can be very small, occupying only a few pixels in the entire image, and exhibit arbitrary perspective transformations. Detection performance can improve by fusing data from multiple remote sensing modalities, including red, green, blue, infrared, hyperspectral, multispectral, synthetic aperture radar, and light detection and ranging, to name a few. In this article, we propose YOLOrs: a new convolutional neural network, specifically designed for real-time object detection in multimodal remote sensing imagery. YOLOrs can detect objects at multiple scales, with smaller receptive fields to account for small targets, as well as predict target orientations. In addition, YOLOrs introduces a novel mid-level fusion architecture that renders it applicable to multimodal aerial imagery. Our experimental studies compare YOLOrs with contemporary alternatives and corroborate its merits.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available