Deep learning technique for image satellite processing
Keywords
Convolutional neural network, Satellite Images, Protocol, Remote SensingAbstract
Recently, several technological solutions based on deep learning have been developed for the processing of multispectral satellite images. These solutions have broadened the scope of spatial remote sensing and further explored the earth and space by avoiding human efforts in heavy manual tasks. However, even though this progress gives even better performance it is difficult to design an efficient deep-learning pipeline for satellite image processing in the absence of a well-developed standard. Thus, we propose a new framework to design and implement a machine-learning model for satellite image processing through a set of machine-learning methods in the form of a protocol adapted to the context of space imagery. The choice of the dataset adapted metrics, exploitation of the product format, data enrichment, spatiotemporal indexing and analysis, contrast enhancement and noise reduction/suppression, network parameterization, model scalability, data normalization, spectral dependency, and processing complexity reduction are among the methods adapted to the processing of multispectral satellite images by deep learning. All these methods are analyzed in depth in order to understand their usefulness and their contribution to the performance of learning models before testing them on real use cases. This method constitutes the first standard framework for the use of deep learning in the processing of multispectral satellite images.
Downloads
References
Published: 2023-03-21
Issue: Vol. 2 No. 1 (2023) (view)
Section: Research Articles
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IMIENS open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.