Abstract
In order to enhance the efficiency of the image transmission system and the robustness of the optical imaging system of the Association of Sino-Russian Technical Universities satellite, a new framework of on-board cloud detection by utilizing a lightweight U-Net and JPEG compression strategy is described. In this method, a careful compression strategy is introduced and evaluated to acquire a balanced result between the efficiency and power consuming. A deep-learning network combined with lightweight U-Net and Mobilenet is trained and verified with a public Landsat-8 data set Spatial Procedures for Automated Removal of Cloud and Shadow. Experiment results indicate that by utilizing image-compression strategy and depthwise separable convolutions, the maximum memory cost and inference speed are dramatically reduced into 0.7133 Mb and 0.0378 s per million pixels while the overall accuracy achieves around 93.1%. A good possibility of the on-board cloud detection based on deep learning is explored by the proposed method.
Highlights
As far as the CubeSat satellites are concerned, the capability of the data transmission system would be largely limited by power consumption and covering time due to its low cost and lack of ground stations.For example, The Association of Sino-Russian Technical Universities (ASRTU) satellite is designed by the Research Center of Satellite Technology (RCST) in Harbin Institute of Technology (HIT)
The increase of the spatial resolution of RS missions with small size and lightweight proposes more requirements of downlink capability, the on-board detection and classification of the RS images are of great value to improve the downlink efficiency and enhance the performance of the optical imaging system, especially in small satellites with low cost
A framework of the on-board cloud-detection system is investigated, and the experiment results have demonstrated that MobU-Net network combined with Haar wavelet compression algorithm shows the best performance in general on the SPARCS data set for the ASRTU mission
Summary
Take the LilacSat-2 satellite as a reference which was designed by HIT and launched on 20 September 2015. With downlink frequency of 437.2 MHz and orbit height of 524 km, the downlink data rate is less than 9600 bps, and the time window is around 10 min in every track, which means less than 3 images with size of 500 3 500 are able to be downloaded from satellite during one track. It highlights the importance of image processing and classification before the data transmission to improve its downlink efficiency
Sign-in/Register to view highlights & summary 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.
