Deep learning has certainly become the dominant trend in hyper-spectral (HS) remote sensing image classification owing to its excellent capabilities to extract highly discriminating spectral-spatial features. In this context, transformer networks have recently shown prominent results in distinguishing even the most subtle spectral differences because of their potential to characterize sequential spectral data. Nonetheless, many complexities affecting HS remote sensing data (e.g. atmospheric effects, thermal noise, quantization noise, etc.) may severely undermine such potential since no mode of relieving noisy feature patterns has still been developed within transformer networks. To address the problem, this paper presents a novel masked autoencoding spectral-spatial transformer (MAEST), which gathers two different collaborative branches: (i) a reconstruction path, which dynamically uncovers the most robust encoding features based on a masking auto-encoding strategy; and (ii) a classification path, which embeds these features onto a transformer network to classify the data focusing on the features that better reconstruct the input. Unlike other existing models, this novel design pursues to learn refined transformer features considering the aforementioned complexities of the HS remote sensing image domain. The experimental comparison, including several stateof-the-art methods and benchmark datasets, shows the superior results obtained by MAEST. The codes of this paper will be available at https://github.com/ibanezfd/MAEST.

The increasing availability of remote sensing data raises important challenges in terms of operational data provision and spatial coverage for conducting global studies and analyses. In this regard, existing multi-temporal mosaicing techniques are generally limited to produce spectral image composites without considering the particular features of higher-level biophysical and other derived products, such as those provided by the Sentinel-3 (S3) and Fluorescence Explorer (FLEX) tandem missions. To relieve these limitations, this paper proposes a novel multitemporal mosaicing algorithm specially designed for operational S3 derived products and also studies its applicability within the FLEX mission context. Specifically, we design a new operational methodology to automatically produce multi-temporal mosaics from derived S3/FLEX products with the objective of facilitating the automatic processing of high-level data products where weakly, monthly, seasonal or annual biophysical mosaics can be generated by means of four processes proposed in this work: (1) operational data acquisition, (2) spatial mosaicing and rearrangement, (3) temporal compositing and (4) confidence measures. The experimental part of the work tests the consistency of the proposed framework over different S3 product collections while showing its advantages with respect to other standard mosaicing alternatives. The source codes of this work will be made available for reproducible research.

No abstract

No abstract

Sentinel missions provide widespread opportunities of exploiting inter-sensor synergies to improve the operational monitoring of terrestrial photosynthetic activity and canopy structural variations using vegetation indices (VI). In this context, continuous and consistent temporal data are logically required to rapidly detect vegetation changes across sensors. Nonetheless, the existing temporal limitations inherent to satellite orbits, cloud occlusions, data degradation and many other factors may severely constrain the availability of data involving multiple satellites. In response, this paper proposes a novel deep 3D convolutional regression network (3CRN) for temporally enhancing Sentinel-3 VI by taking advantage of inter-sensor Sentinel-2 observations. Unlike existing regression and deep learning-based methods, the proposed approach allows convolutional kernels to slide across the temporal dimension in order to exploit not only the higher spatial resolution of the Sentinel-2 instrument but also its own temporal evolution to better estimate time-resolved VI in Sentinel-3. To validate the proposed approach, we built a database made of multiple day-synchronized Sentinel-2 and Sentinel-3 operational products from a study area in Extremadura (Spain). The conducted experimental comparison, including multiple state-of-theart regression and deep learning models, shows the statistically significant advantages of the presented framework. The codes of this work will be made available at https://github.com/rufernan/3CRN.