Abstract
AbstractDeep learning has proved to be useful for ship detection in SAR satellite images. SAR is a satellite that can be used to capture images from Earth’s surface even in unfavorable weather conditions. Although many object detection models have been applied to this problem previously, we wanted to use a model that was fast and accurate. In this paper, we use You Only Look Once version-3 (YOLOv3) and compare its results with You Only Look Once version-2 (YOLOv2). The motivation was to improve upon the results of YOLOv2. The results show that YOLOv3 achieved 90.25 average precision (AP) compared to 90.05 AP of YOLOv2. Furthermore, YOLOv3 gave an inference time of 22 ms against 25 ms of YOLOv2. The dataset used is A SAR Dataset of Ship Detection for Deep Learning under Complex Backgrounds; the dataset consists of 43,819 images of 256 × 256 pixels. The dataset contains Gaofen-3 and Sentinel-1 satellite images.KeywordsYOLOv3Remote sensingSynthetic-aperture radar (SAR) imagesObject detectionShip detection
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
