Filters








57,501 Hits in 4.9 sec

Training general representations for remote sensing using in-domain knowledge [article]

Maxim Neumann, André Susano Pinto, Xiaohua Zhai, Neil Houlsby
2020 arXiv   pre-print
For this analysis, five diverse remote sensing datasets are selected and used for both, disjoint upstream representation learning and downstream model training and evaluation.  ...  This paper investigates development of generic remote sensing representations, and explores which characteristics are important for a dataset to be a good source for representation learning.  ...  In this paper we explore representation learning for remote sensing, and in particular in how much in-domain knowledge from related datasets could help in representation learning.  ... 
arXiv:2010.00332v1 fatcat:woi2j23t3jeurokgff2h3ydiv4

In-domain representation learning for remote sensing [article]

Maxim Neumann, Andre Susano Pinto, Xiaohua Zhai, Neil Houlsby
2019 arXiv   pre-print
Specifically, we investigate in-domain representation learning to develop generic remote sensing representations and explore which characteristics are important for a dataset to be a good source for remote  ...  To address it and to establish baselines and a common evaluation protocol in this domain, we provide simplified access to 5 diverse remote sensing datasets in a standardized form.  ...  REPRESENTATION/TRANSFER LEARNING IN REMOTE SENSING Since labeling remote sensing data is expensive, for a long time there was no equivalent to ImageNet and most benchmark datasets were small.  ... 
arXiv:1911.06721v1 fatcat:4izoowzbcfcrdeuppqbjyw7llu

Research Progress on Few-Shot Learning for Remote Sensing Image Interpretation

Xian Sun, Bing Wang, Zhirui Wang, Hao Li, Hengchao Li, Kun Fu
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
., the data-augmentation-based and the prior-knowledge-based, are introduced for the interpretation of remote sensing images.  ...  This article gives a reference for scholars working on few-shot learning research in the remote sensing field.  ...  One of the most commonly used deep generative models in remote sensing field are GANs [27] .  ... 
doi:10.1109/jstars.2021.3052869 fatcat:ldos3sx6mvaapjkgsua73l7tve

Self-Supervision, Remote Sensing and Abstraction: Representation Learning Across 3 Million Locations [article]

Sachith Seneviratne, Kerry A. Nice, Jasper S. Wijnands, Mark Stevenson, Jason Thompson
2022 arXiv   pre-print
However, the performance of such methods on remote sensing imagery domains remains under-explored.  ...  We show that self-supervised methods can build a generalizable representation from as few as 200 cities, with representations achieving over 95\% accuracy in unseen cities with minimal additional training  ...  This research was undertaken using University of Melbourne Research Computing facilities established by the Petascale Campus Initiative.  ... 
arXiv:2203.04445v1 fatcat:ae2yu7dasvbjxfrteay4vnloju

Recent advances and opportunities in scene classification of aerial images with deep models [article]

Fan Hu, Gui-Song Xia, Wen Yang, Liangpei Zhang
2018 arXiv   pre-print
Scene classification is a fundamental task in interpretation of remote sensing images, and has become an active research topic in remote sensing community due to its important role in a wide range of applications  ...  Over the past years, tremendous efforts have been made for developing powerful approaches for scene classification of remote sensing images, evolving from the traditional bag-of-visual-words model to the  ...  Considering that it is challenging to well annotate a highcoverage and high-diversity dataset (time-consuming and call for expert domain knowledge), labeling scene images using geographic crowdsource data  ... 
arXiv:1806.00899v1 fatcat:zqvihhgao5cztc3b6dip7ushbu

EXPLAIN IT TO ME – FACING REMOTE SENSING CHALLENGES IN THE BIO- AND GEOSCIENCES WITH EXPLAINABLE MACHINE LEARNING

R. Roscher, B. Bohn, M. F. Duarte, J. Garcke
2020 ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
In order to reach this goal and obtain explanation, knowledge from the respective domain is necessary, which can be integrated into the model or applied post-hoc.  ...  For some time now, machine learning methods have been indispensable in many application areas.  ...  We are convinced that incorporating domain knowledge to gain explainability is a crucial next step in enabling explainability for ML in remote sensing applications.  ... 
doi:10.5194/isprs-annals-v-3-2020-817-2020 fatcat:2gyr2hmsxzhe7grexelm2dhehi

Learning a Multi-Branch Neural Network from Multiple Sources for Knowledge Adaptation in Remote Sensing Imagery

Mohamad Al Rahhal, Yakoub Bazi, Taghreed Abdullah, Mohamed Mekhalfi, Haikel AlHichri, Mansour Zuair
2018 Remote Sensing  
In this paper we propose a multi-branch neural network, called MB-Net, for solving the problem of knowledge adaptation from multiple remote sensing scene datasets acquired with different sensors over diverse  ...  In the experiments, we demonstrate the effectiveness of the proposed method on a new multiple domain dataset created from four heterogonous scene datasets well known to the remote sensing community, namely  ...  Although multisource domain adaptation has been shown to be very useful in general computer vision literature, it is yet to find its way into remote sensing applications.  ... 
doi:10.3390/rs10121890 fatcat:zqynzfic6jhl3kghbcbhl5fuye

Background Subtraction Based on GAN and Domain Adaptation for VHR Optical Remote Sensing Videos

Wentao Yu, Jing Bai, Licheng Jiao
2020 IEEE Access  
The application of deep learning techniques in background subtraction for VHR optical remote sensing videos holds the potential to facilitate multiple intelligent remote sensing processing tasks.  ...  videos and VHR optical remote sensing videos so that the background subtraction performance for VHR optical remote sensing videos is improved significantly.  ...  The proposed method utilizes the generated background and video frame pairs and obtains the background subtraction result for VHR optical remote sensing videos by using domain adaptation.  ... 
doi:10.1109/access.2020.3004495 fatcat:hban6wuoxvdi5dkypk6yfxnuve

Remote Sensing Images Semantic Segmentation with General Remote Sensing Vision Model via a Self-Supervised Contrastive Learning Method [article]

Haifeng Li, Yi Li, Guo Zhang, Ruoyun Liu, Haozhe Huang, Qing Zhu, Chao Tao
2021 arXiv   pre-print
Therefore, we propose Global style and Local matching Contrastive Learning Network (GLCNet) for remote sensing semantic segmentation.  ...  A new learning paradigm, self-supervised learning (SSL), can be used to solve such problems by pre-training a general model with large unlabeled images and then fine-tuning on a downstream task with very  ...  In this work, we use contrastive learning to learn general spatiotemporal invariance features for remote sensing.  ... 
arXiv:2106.10605v1 fatcat:eohfunmdrzdflmhkupdda4wyry

State-of-the-art and gaps for deep learning on limited training data in remote sensing [article]

John E. Ball, Derek T. Anderson, Pan Wei
2018 arXiv   pre-print
However, most remote sensing applications only have limited training data, of which a small subset is labeled.  ...  The first topic is transfer learning, in which some aspects of one domain, e.g., features, are transferred to another domain.  ...  , and v) how does transfer learning work in the context of multisensor fusion for remote sensing?  ... 
arXiv:1807.11573v1 fatcat:q6vtrod6nvgtrihjafo25iz3wi

Remote Sensing Image Scene Classification with Self-Supervised Paradigm under Limited Labeled Samples [article]

Chao Tao, Ji Qi, Weipeng Lu, Hao Wang, Haifeng Li
2020 arXiv   pre-print
The insights distilled from our studies can help to foster the development of SSL in the remote sensing community.  ...  With the development of deep learning, supervised learning methods perform well in remote sensing images (RSIs) scene classification.  ...  Generally, the objective function for such a task uses L2 loss as shown in (1).  ... 
arXiv:2010.00882v1 fatcat:xqxmrld7p5avpng67hvbsrraly

Seasonal Contrast: Unsupervised Pre-Training from Uncurated Remote Sensing Data [article]

Oscar Mañas, Alexandre Lacoste, Xavier Giro-i-Nieto, David Vazquez, Pau Rodriguez
2021 arXiv   pre-print
In this work, we propose Seasonal Contrast (SeCo), an effective pipeline to leverage unlabeled data for in-domain pre-training of remote sensing representations.  ...  However, most of these methods are pre-trained on ImageNet and their generalization to remote sensing imagery is not guaranteed due to the domain gap.  ...  Method We propose a methodology for pre-training rich, transferable representations for remote sensing imagery, consisting of a general procedure for collecting an unsupervised pre-training dataset (Section  ... 
arXiv:2103.16607v2 fatcat:r7dukovfavckfbl4blkwqsdsiy

Learning Physical Scattering Patterns from PolSAR Images by Using Complex-Valued CNN

Juanping Zhao, Mihai Datcu, Zenghui Zhang, Huilin Xiong, Wenxian Yu
2019 IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium  
Deep learning algorithms are widely used in remote sensing image scene understanding. Generally, a large-scale annotated dataset is essential to train a deep neural network for classification.  ...  In this paper, an unsupervised domain adaptation framework based on ResNet-18 is presented to transfer the knowledge of an existing annotated land cover dataset to other remote sensing data, decreasing  ...  Transfer learning is also applied to remote sensing data to overcome the difficulty of limited training samples [4] . However, those methods usually use the same dataset for training and testing.  ... 
doi:10.1109/igarss.2019.8900150 dblp:conf/igarss/ZhaoDZXY19 fatcat:uyqx3efedbbjrmib5o2w2wjavi

Attention-based Domain Adaptation Using Residual Network for Hyperspectral Image Classification

Robiul Hossain Mdrafi, Qian Du, Ali Cafer Cafer Gurbuz, Bo Tang, Li Ma, Nick H Younan
2020 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
In remote sensing images, domain adaptation (DA) deals with the regions where labeling information is unknown.  ...  in DA domain.  ...  She was the General Chair for the fourth IEEE GRSS Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing held at Shanghai, in 2012.  ... 
doi:10.1109/jstars.2020.3035382 fatcat:qgqxthmlzba5bnh2algsdbl7za

RSMT: A Remote Sensing Image-to-Map Translation Model via Adversarial Deep Transfer Learning

Jieqiong Song, Jun Li, Hao Chen, Jiangjiang Wu
2022 Remote Sensing  
In this work, we intend to seek a remote sensing image-to-map translation model that approaches the challenge of generating maps for the remote sensing images of unseen areas.  ...  Maps can help governments in infrastructure development and emergency rescue operations around the world. Using adversarial learning to generate maps from remote sensing images is an emerging field.  ...  [35] use a new CNN architecture for transferring knowledge cross-domain and cross-task. For adversarial adaptation, Tzeng et al. [36] then first outline a new generalized framework called ADDA.  ... 
doi:10.3390/rs14040919 fatcat:ik7oiue2zbcl3p2wjsapbpuubm
« Previous Showing results 1 — 15 out of 57,501 results