TY - JOUR
T1 - A robust end-to-end deep learning framework for detecting Martian landforms with arbitrary orientations
AU - Jiang, Shancheng
AU - Wu, Fan
AU - Yung, K. L.
AU - Yang, Yingqiao
AU - Ip, W. H.
AU - Gao, Ming
AU - Foster, James Abbott
N1 - Funding Information:
This work was supported in part by the project grant ZG3K Chang’e phase 3 sample return instruments, in part by the National Nature Science and Foundation of China Grand No. 71801031 , in part by the Guangdong Basic and Applied Basic Research Foundation project, China , No. 2019A1515011962 and 2020A1515110431 , in part by the National Nature Science and Foundation of China Grand No. 71772033 , in part by the Natural Science Foundation of Liaoning Province, China (Joint Funds for Key Scientific Innovation Bases, 2020-KF-11-11 ), and in part by the Scientific Research Project of the Education Department of Liaoning Province, China ( LN2019Q14 ).
Publisher Copyright:
© 2021 Elsevier B.V.
PY - 2021/12/25
Y1 - 2021/12/25
N2 - With increasingly massive amounts of high-resolution images of Mars, automated detection of geological landforms on Mars has received widespread interest. It is significant for acquiring knowledge of distant planetary surfaces and processes, or manifold onboard applications such as spacecraft motion estimation and obstacle avoidance. This is a challenging task, not only because of the multiple sizes of targets and complex image backgrounds, but also the various orientations of some bar-shaped landforms in satellite images captured from the top view. The existing methods for directed landform detection require several pre or post-processing operations to extract possible regions of interest and final detection results with orientation, which are very time consuming. In this paper, a new end-to-end deep learning framework is developed for detecting arbitrarily-directed landforms. This framework, named Rotated-SSD (Single Shot MultiBox Detector, SSD), can locate and identify different landforms on Mars in one pass, by using rotatable anchor-box based mechanism. To enhance its robustness against angle variation of the targets and complex backgrounds, a new efficient match strategy is proposed for anchoring default boxes to ground truth boxes in the model training process and an autoencoder-based unsupervised pre-training operation is introduced to improve both the model training and inference performance. The proposed framework is tested for detection of bar-shaped buttes and dark slope streaks on satellite images. The detection results show that our framework can significantly contribute to onboard motion estimation systems. The comparative results demonstrate that the proposed match strategy outperforms other state-of-the-art match strategies with regard to model training efficiency and prediction accuracy. The pre-training strategy can facilitate the training of deep architectures in case of limited available training data.
AB - With increasingly massive amounts of high-resolution images of Mars, automated detection of geological landforms on Mars has received widespread interest. It is significant for acquiring knowledge of distant planetary surfaces and processes, or manifold onboard applications such as spacecraft motion estimation and obstacle avoidance. This is a challenging task, not only because of the multiple sizes of targets and complex image backgrounds, but also the various orientations of some bar-shaped landforms in satellite images captured from the top view. The existing methods for directed landform detection require several pre or post-processing operations to extract possible regions of interest and final detection results with orientation, which are very time consuming. In this paper, a new end-to-end deep learning framework is developed for detecting arbitrarily-directed landforms. This framework, named Rotated-SSD (Single Shot MultiBox Detector, SSD), can locate and identify different landforms on Mars in one pass, by using rotatable anchor-box based mechanism. To enhance its robustness against angle variation of the targets and complex backgrounds, a new efficient match strategy is proposed for anchoring default boxes to ground truth boxes in the model training process and an autoencoder-based unsupervised pre-training operation is introduced to improve both the model training and inference performance. The proposed framework is tested for detection of bar-shaped buttes and dark slope streaks on satellite images. The detection results show that our framework can significantly contribute to onboard motion estimation systems. The comparative results demonstrate that the proposed match strategy outperforms other state-of-the-art match strategies with regard to model training efficiency and prediction accuracy. The pre-training strategy can facilitate the training of deep architectures in case of limited available training data.
KW - Autoencoder
KW - Deep learning
KW - Match strategy
KW - Object detection
KW - Rotated single shot multibox detector
UR - http://www.scopus.com/inward/record.url?scp=85117269668&partnerID=8YFLogxK
U2 - 10.1016/j.knosys.2021.107562
DO - 10.1016/j.knosys.2021.107562
M3 - Journal article
AN - SCOPUS:85117269668
SN - 0950-7051
VL - 234
JO - Knowledge-Based Systems
JF - Knowledge-Based Systems
M1 - 107562
ER -