S-track directions, respectively. In addition, the Compound 48/80 Cancer reference orbit derived by the unique
S-track directions, respectively. Moreover, the reference orbit derived by the various ECOM models was assessed by way of the orbit overlap at the day boundary. The orbit accuracy improvements of ECOMC more than ECOM2 have been 13.2 , 14.eight , and 42.6 for the IIF satellites and 7.4 , 7.7 , and 35.0 for the IIR satellites within the radial, along-track, and cross-track directions, respectively. This result shows that ECOMC considerably reduces the errors in the cross-track path, exactly where the SLR may not proficiently validate the outcome. We also assessed the impact from the reference orbit derived by ECOM1, ECOM2, and ECOMC on PPP. The outcome showed that the improvement in the ECOMC resolution over ECOM2 and ECOM1 was approximately 20 and 13 , respectively.Funding: This analysis is funded by the Ministry of ML-SA1 Neuronal Signaling Science and Technologies of Taiwan, grant number [MOST 110-2121-M-992-002]. Institutional Overview Board Statement: Not applicable. Informed Consent Statement: Not applicable. Information Availability Statement: We thank each IGS and International Earth Rotation Service (IERS) for delivering the GNSS orbits and Earth orientation parameters. Acknowledgments: We thank Geoscience Australia for supplying Ginan software program for processing GNSS data. We are also grateful to Simon McClusky from Geoscience Australia for offering the concept behind conducting the SRP work. Conflicts of Interest: The author declares no conflict of interest.
remote sensingArticleClassifying Crop Kinds Applying Two Generations of Hyperspectral Sensors (Hyperion and DESIS) with Machine Learning on the CloudItiya Aneece and Prasad S. ThenkabailU.S. Geological Survey, Western Geographic Science Center, Flagstaff, AZ 86001, USA; [email protected] Correspondence: [email protected]; Tel.: 1-928-556-Citation: Aneece, I.; Thenkabail, P.S. Classifying Crop Kinds Making use of Two Generations of Hyperspectral Sensors (Hyperion and DESIS) with Machine Understanding on the Cloud. Remote Sens. 2021, 13, 4704. https://doi.org/ ten.3390/rsAbstract: Advances in spaceborne hyperspectral (HS) remote sensing, cloud-computing, and machine learning can assist measure, model, map and monitor agricultural crops to address worldwide food and water security difficulties, like by offering precise estimates of crop region and yield to model agricultural productivity. Leveraging these advances, we utilized the Earth Observing-1 (EO-1) Hyperion historical archive as well as the new generation DLR Earth Sensing Imaging Spectrometer (DESIS) data to evaluate the performance of hyperspectral narrowbands in classifying key agricultural crops with the U.S. with machine mastering (ML) on Google Earth Engine (GEE). EO-1 Hyperion pictures in the 2010013 expanding seasons and DESIS photos from the 2019 increasing season were utilised to classify three planet crops (corn, soybean, and winter wheat) in addition to other crops and non-crops close to Ponca City, Oklahoma, USA. The supervised classification algorithms: Random Forest (RF), Assistance Vector Machine (SVM), and Naive Bayes (NB), along with the unsupervised clustering algorithm WekaXMeans (WXM) have been run working with selected optimal Hyperion and DESIS HS narrowbands (HNBs). RF and SVM returned the highest general producer’s, and user’s accuracies, with all the performances of NB and WXM getting substantially decrease. The most beneficial accuracies had been achieved with two or 3 photos all through the developing season, especially a combination of an earlier month (June or July) and a later month (August or September). The narrow two.55 nm bandwidth of DESIS supplied numero.