Open AccessSatellite Segmentation with Pre-trained CNN Models
Author(s) -
Yefei Huang,
Tao Xu,
Zexu Zhang,
Hutao Cui,
Yu Su
Publication year - 2022
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2171/1/012003
Subject(s) - computer science , artificial intelligence , segmentation , satellite , intersection (aeronautics) , matching (statistics) , computer vision , pipeline (software) , task (project management) , object detection , feature extraction , pose , pattern recognition (psychology) , geography , cartography , mathematics , statistics , management , engineering , economics , programming language , aerospace engineering
In a generic satellite relative pose estimation pipeline, finding sufficient features in objects is quite essential to build the correct matching relationship and then solve the relative movement. However, for low-earth-orbit (LEO) satellites, since the earth background contains much more texture than objects, an object segmentation process is necessary to provide a prior range for feature extraction. In this work, we address this task with the pre-trained Deeplabv3 and fully convolutional network (FCN). Unlike the fine-tuning or transfer learning processes in other researches, we obtain probabilistic maps from the high-dimensional output of the above-mentioned CNN models and achieve a rough satellite extraction. Our method makes Deeplabv3 and FCN models work in a totally unfamiliar LEO scene and still achieves 0.2927 and 0.2122 in average intersection over union (IoU) respectively.