Search

Accurate cloud and shadow detection is a crucial prerequisite for optical remote sensing image analysis and application. Multi-temporal-based cloud and shadow detection methods are a preferable choice to detect clouds in complex scenes (e.g., thin clouds, broken clouds and clouds with interference from artificial surfaces with high reflectivity). However, such methods commonly require cloud-free reference images, and this may be difficult to achieve in time-series data since clouds are often prevalent and of varying spatial distribution in optical remote sensing images. Furthermore, current multi-temporal-based methods have limited feature extraction capability and rely heavily on prior assumptions. To address these issues, this paper proposes a Siamese network (Siamnet) for cloud and shadow detection based on Time-Series cloudy Images, namely TSI-Siamnet, which consists of two steps: 1) low-rank and sparse component decomposition of time-series cloudy images is conducted to construct a composite reference image to cope with dynamic changes in the cloud distribution in time-series images; 2) an extended Siamnet with optimal difference calculation module (DM) and multi-scale difference features fusion module (MDFM) is constructed to extract reliable disparity features and alleviate semantic information feature dilution during the decoder part. TSI-Siamnet was tested extensively on seven land cover types in the well-known Landsat 8 Biome dataset. Compared to six state-of-the-art methods (including four deep learning-based methods and two classical non-deep learning-based methods), TSI-Siamnet produced the best performance with an overall accuracy of 95.05% and MIoU of 84.37%. In three more challenging experiments, TSI-Siamnet showed enhanced detection of thin and broken clouds and greater anti-interference to highly reflective surfaces. TSI-Siamnet provides a novel strategy to explore comprehensively the valid information in time-series cloudy images and integrate the extracted spectral-spatial–temporal features for reliable cloud and shadow detection. [ABSTRACT FROM AUTHOR]

Young forest age mapping at a fine spatial resolution is important for increasing the accuracy of estimating land–atmosphere carbon fluxes and guiding forest management practices. In recent decades, China has actively conducted afforestation and forest protection projects, thereby laying the foundation for the realization of carbon neutrality. However, very few studies have been conducted which map the ages of young forests for the whole of China at a fine spatial resolution. In this research, a continuous change detection and classification (CCDC)-based method suitable for large-scale forest age mapping is proposed and used to estimate young forest ages across China in 2020 at a spatial resolution of 30 m. First, a 10 m spatial-resolution land cover dataset (WorldCover2020) from the European Space Agency (ESA) was used to determine the forest cover areas in 2020. Then, the CCDC algorithm was used to identify stand-replacing disturbances to determine the stand age based on 436 967 Landsat tiles across China from 1990 to 2020. A validation sample set composed of multiple land use and land cover (LULC) products was used to calculate the overall accuracy (OA) of the 2020 young forest age (1–31-year) map of China, and the OA was 90.28 %. The reliability and applicability of the proposed CCDC-based forest age mapping method were validated by comparing the forest age map with Hansen's forest change dataset, Max Planck Institute for Biogeochemistry (MPI-BGC) 1 km global forest age datasets, and field measurements. The CCDC-based method has strong application potential in real-time mapping of the age of young forests at the global scale. The produced forest age map provides a basic dataset for research on the forest carbon cycle and forest ecosystem services as well as important guidance for government departments, such as the National Forestry and Grassland Administration and the National Development and Reform Commission in China. Data presented in this study is available at 10.6084/m9.figshare.21627023.v7 (Xiao, 2022). [ABSTRACT FROM AUTHOR]

Land surface temperature (LST) is an important factor in studies of surface energy fluxes between the Earth's surface and atmosphere. The Landsat LST product has been applied widely due to its fine spatial resolution and high data quality. Frequent cloud coverage, however, results in different degrees of gaps in the Landsat LST images, restricting greatly their application in practical cases requiring continuous LST time-series. Gap filling is an important technique used to reconstruct the missing data in the Landsat LST images. Using only temporally neighboring data from the same sensor, however, current gap filling methods struggle to deal with large spatial gaps and large temporal changes in LST. To solve these challenging issues, in this paper, multi-scale data (i.e., the 2 km, all-sky, hourly GOES LST product) were proposed for gap filling of Landsat LST images, and a spatial-temporal random forest (STRF) method was developed. The GOES LST product can provide effective information on LST in the gaps of the target Landsat LST data, with great potential to decreasing the uncertainty in gap filling when large LST change exists or the gap size is large. Through experiments in six regions of the USA, the proposed STRF method was shown to be more accurate than three typical gap filling and spatio-temporal fusion methods, and particularly advantageous for cases involving large temporal changes or large spatial gaps in LST. Specifically, STRF produces an average root mean square error (RMSE) of 1.6444 K and average correlation coefficient (CC) of 0.8653 for the six regions, which are 0.1133 K smaller and 0.0176 larger than the most competitive benchmark method (i.e., the spatial random forest (SRF) method). Conceptually, by spatial-temporal fusion of multi-scale data in gap filling, STRF provides a new paradigm for both spatial-temporal fusion and gap filling. Furthermore, the proposed STRF has the potential to be applied to various gap filling tasks in other situations. • 2 km, all-sky, hourly GOES LST data were used to assist gap filling of Landsat LST. • A spatial-temporal random forest (STRF) method was proposed. • STRF provides a new paradigm for both spatial-temporal fusion and gap filling. • The method is advantageous for large temporal changes and spatial gaps in LST. [ABSTRACT FROM AUTHOR]

Haze contamination is a common issue in optical remote sensing images, including hyperspectral images (HSIs), which can distort the spectral features of land cover objects. Over the last decades, although many haze removal solutions have been developed, very few studies have focused on haze removal of HSIs. Moreover, most of these methods cannot fully explore the abundant spectral information of HSIs in haze removal. To cope with the issues, a data driven method is proposed for haze removal of HSIs in this paper. Specifically, we design a spectral grouping network (SG-Net) to fully utilize the useful information in each spectral band during the reconstruction. To facilitate the relationship construction between hazy image and the corresponding haze-clear image, the proposed SG-Net first groups each HSI into several spectral subsets based on the intra-spectral correlations. Then, these subsets are convoluted in parallel with multiple branches for feature extraction. Furthermore, a novel attention block is designed to connect the adjacent branches for feature transmission, which can distill the useful information (e.g., uncontaminated information in longer wavelength bands) of each subset and assist the reconstruction of HSIs. Comprehensive experiments on both simulated and real haze HSIs showed that SG-Net is more accurate than seven state-of-the-art haze removal methods and is also more robust to different haze levels and shapes. [ABSTRACT FROM AUTHOR]

Subpixel mapping (SPM) is a technique to tackle the mixed-pixel problem and produces land cover and land use (LCLU) maps at a finer spatial resolution than the original coarse data. However, uncertainty exists unavoidably in SPM, which is an ill-posed downscaling problem. Spatio-temporal SPM methods have been proposed to deal with this uncertainty, but current methods fail to explore fully the information in the time-series images, especially more rapid changes over a short-time interval. In this article, a fast and slow changes constrained spatio-temporal subpixel mapping (FSSTSPM) method is proposed to account for fast LCLU changes over a short time interval and slow changes over a long time interval. Both fast and slow changes-based temporal constraints are proposed and incorporated simultaneously into the FSSTSPM to increase the accuracy of SPM. The proposed FSSTSPM method was validated using two synthetic datasets with various proportion errors. It was also applied to oil-spill mapping using a real PlanetScope-Sentinel-2 dataset and Amazon deforestation mapping using a real Landsat-Moderate Resolution Imaging Spectroradiometer (MODIS) dataset. The results demonstrate the superiority of FSSTSPM. Moreover, the advantage of FSSTSPM is more obvious with an increase in proportion errors. The concepts of the fast and slow changes, together with the derived temporal constraints, provide a new insight to enhance SPM by taking fuller advantage of the temporal information in the available time-series images. [ABSTRACT FROM AUTHOR]

Sentinel-3 is a newly launched satellite implemented by the European Space Agency (ESA) for global observation. The Ocean and Land Colour Imager (OLCI) sensor onboard Sentinel-3 provides 21 band images with a fine spectral resolution and is of great value for ocean, land and atmospheric monitoring. The two platforms (Sentinel-3A and -3B) can provide OLCI images at an almost daily temporal resolution. The coarse spatial resolution of the 21 band OLCI images (i.e., 300 m), however, limits greatly their utility for local, precise monitoring. Sentinel-2, another satellite provided by ESA, carries the Multispectral Imager (MSI) sensor which can supply much finer spatial resolution (e.g., 10 m and 20 m) images. This paper introduces a new fusion framework integrating spatio-temporal-spectral information for downscaling Sentinel-3 OLCI images, which has two parts. Based on bands with similar wavelengths (i.e., bands 2, 3, 4 and 8a for Sentinel-2 and bands Oa4, Oa6, Oa8 and Oa17 for Sentinel-3), the four Sentinel-3 bands are first downscaled to the spatial resolution of Sentinel-2 images by applying spatio-temporal fusion to Sentinel-2 MSI and Sentinel-3 OLCI images. Then, to take full advantage of all 21 available OLCI bands of the Sentinel-3 images, the extended image pair-based spatio-spectral fusion (EIPSSF) method is proposed in this paper to downscale the other 17 bands. EIPSSF is performed based on the new concept of the extended image pair (EIP) and by exploiting existing spatio-temporal fusion approaches. The framework consisting of spatio-temporal and spatio-spectral fusion is entirely general, which provides a practical solution for comprehensive downscaling of Sentinel-3 OLCI images for fine spatial, temporal and spectral resolution monitoring. [ABSTRACT FROM AUTHOR]

The point spread function (PSF) is ubiquitous in remote sensing. This paper investigated the effect of the PSF on the downscaling of continua. Geostatistical approaches were adopted to incorporate explicitly, and reduce the influence of, the PSF effect in downscaling. Two general cases were considered: univariate and multivariate. In the univariate case, the input coarse spatial resolution image is the only image available for downscaling. Area-to-point kriging was demonstrated to be a suitable solution in this case. For the multivariate case, a finer spatial resolution image (or images) observed under different conditions (e.g., at a different wavelength) is available as auxiliary data for downscaling. Area-to-point regression kriging was shown to be a suitable solution for this case. Moreover, a new solution was developed for estimating the PSF in image scale transformation. The experiments show that the PSF effect influences downscaling greatly and that downscaling can be enhanced obviously by considering the PSF effect through the geostatistical approaches and the PSF estimation solution proposed. [ABSTRACT FROM AUTHOR]

In this paper, a new method was investigated to enhance remote sensing images by alleviating the point spread function (PSF) effect. The PSF effect exists ubiquitously in remotely sensed imagery. As a result, image quality is greatly affected, and this imposes a fundamental limit on the amount of information captured in remotely sensed images. A geostatistical filter was proposed to enhance image quality based on a downscaling-then-upscaling scheme. The difference between this method and previous methods is that the PSF is represented by breaking the pixel down into a series of sub-pixels, facilitating downscaling using the PSF and then upscaling using a square-wave response. Thus, the sub-pixels allow disaggregation as an attempt to remove the PSF effect. Experimental results on simulated and real data sets both suggest that the proposed filter can enhance the original images by reducing the PSF effect and quantify the extent to which this is possible. The predictions using the new method outperform the original coarse PSF-contaminated imagery as well as a benchmark method. The proposed method represents a new solution to compensate for the limitations introduced by remote sensors (i.e., hardware) using computer techniques (i.e., software). The method has widespread application value, particularly for applications based on remote sensing image analysis. [ABSTRACT FROM AUTHOR]

Spatial downscaling is an ill-posed, inverse problem, and information loss (IL) inevitably exists in the predictions produced by any downscaling technique. The recently popularized area-to-point kriging (ATPK)-based downscaling approach can account for the size of support and the point spread function (PSF) of the sensor, and moreover, it has the appealing advantage of the perfect coherence property. In this article, based on the advantages of ATPK and the conceptualization of IL, an IL-guided image fusion (ILGIF) approach is proposed. ILGIF uses the fine spatial resolution images acquired in other wavelengths to predict the IL in ATPK predictions based on the geographically weighted regression (GWR) model, which accounts for the spatial variation in land cover. ILGIF inherits all the advantages of ATPK, and its prediction has perfect coherence with the original coarse spatial resolution data which can be demonstrated mathematically. ILGIF was validated using two data sets and was shown in each case to predict downscaled images more accurately than the compared benchmark methods. [ABSTRACT FROM AUTHOR]

• A spatial seamless, daily SMAP SSM dataset was generated from 2015 to 2021. • A highly comparative time-series analysis (HCTSA)-based method was employed. • Spatially seamless characteristics were extracted by HCTSA. • The unbiased root mean square error of the SSD_SMAP dataset can reach ∼ 0.040 m3/m3. Surface soil moisture (SSM) is a vital component in terrestrial hydrological processes. As a type of important microwave remote sensing-based SSM dataset, the Soil Moisture Active Passive (SMAP) SSM data were available since 2015 and have been applied to many studies. However, the spatial gaps in the SMAP SSM data affect greatly its applicability. To address this issue, this study developed a method to fill the spatial gaps in the SMAP SSM data, generating a spatially seamless, daily SMAP (SSD_SMAP) SSM dataset from 2015 to 2021. The method makes full use of the long SMAP SSM time-series data with inherent spatial gaps. To deal with the continuously changed spatial gaps in the daily SMAP data, we used a highly comparative time-series analysis (HCTSA) to dig the temporal profiles of the SMAP SSM time-series, producing 17 seamless HCTSA-based time-series characteristics (HTCs). Both spatially seamless topography characteristics (TCs) and location characteristics (LCs) were also considered. Moreover, a random forest (RF) model was trained to construct the relation between the valid daily SMAP data at the prediction time and the seamless auxiliary data (i.e., 17 HTCs coupled with TCs and LCs). Subsequently, the constructed RF model was migrated to fill the spatial gaps of the SMAP data at the corresponding time. The proposed method was validated in both simulated gaps (real data as reference) and real gaps (in-situ SSM data as reference). The produced SSD_SMAP dataset was also compared with the model-based Global Land Evaporation Amsterdam Model (GLEAM) dataset. The results indicated that the SSD_SMAP data are closer to the in-situ data, with an unbiased root mean square error (ubRMSE) of ∼ 0.040 m3/m3. The SSD_SMAP dataset can provide data support for global and regional research in related fields. [ABSTRACT FROM AUTHOR]

Sentinel-2 and Sentinel-3 are two newly launched satellites for global monitoring. The Sentinel-2 Multispectral Imager (MSI) and Sentinel-3 Ocean and Land Colour Instrument (OLCI) sensors have very different spatial and temporal resolutions (Sentinel-2 MSI sensor 10 m, 20 m and 60 m, 10 days, albeit 5 days with 2 sensors, conditional upon clear skies; Sentinel-3 OLCI sensor 300 m, < 1.4 days with 2 sensors). For local monitoring (e.g., the growing cycle of plants) one either has the desired spatial or temporal resolution, but not both. In this paper, spatio-temporal fusion is considered to fuse Sentinel-2 with Sentinel-3 images to create nearly daily Sentinel-2 images. A challenging issue in spatio-temporal fusion is that there can be very few cloud-free fine spatial resolution images temporally close to the prediction time, or even available, strong temporal (i.e., seasonal) changes may exist. To this end, a three-step method consisting of regression model fitting (RM fitting), spatial filtering (SF) and residual compensation (RC) is proposed, which is abbreviated as Fit-FC. The Fit-FC method can be performed using only one Sentinel-3–Sentinel-2 pair and is advantageous for cases involving strong temporal changes (i.e., mathematically, the correlation between the two Sentinel-3 images is small). The effectiveness of the method was validated using two datasets. The created nearly daily Sentinel-2 time-series images have great potential for timely monitoring of highly dynamic environmental, agricultural or ecological phenomena. [ABSTRACT FROM AUTHOR]

Sentinel-2 is a wide-swath and fine spatial resolution satellite imaging mission designed for data continuity and enhancement of the Landsat and other missions. The Sentinel-2 data are freely available at the global scale, and have similar wavelengths and the same geographic coordinate system as the Landsat data, which provides an excellent opportunity to fuse these two types of satellite sensor data together. In this paper, a new approach is presented for the fusion of Landsat 8 Operational Land Imager and Sentinel-2 Multispectral Imager data to coordinate their spatial resolutions for continuous global monitoring. The 30 m spatial resolution Landsat 8 bands are downscaled to 10 m using available 10 m Sentinel-2 bands. To account for the land-cover/land-use (LCLU) changes that may have occurred between the Landsat 8 and Sentinel-2 images, the Landsat 8 panchromatic (PAN) band was also incorporated in the fusion process. The experimental results showed that the proposed approach is effective for fusing Landsat 8 with Sentinel-2 data, and the use of the PAN band can decrease the errors introduced by LCLU changes. By fusion of Landsat 8 and Sentinel-2 data, more frequent observations can be produced for continuous monitoring (this is particularly valuable for areas that can be covered easily by clouds, thereby, contaminating some Landsat or Sentinel-2 observations), and the observations are at a consistent fine spatial resolution of 10 m. The products have great potential for timely monitoring of rapid changes. [ABSTRACT FROM PUBLISHER]

Sub-pixel mapping (SPM) is a process for predicting spatially the land cover classes within mixed pixels. In existing SPM methods, the effect of point spread function (PSF) has seldom been considered. In this paper, a generic SPM method is developed to consider the PSF effect in SPM and, thereby, to increase prediction accuracy. We first demonstrate that the spectral unmixing predictions (i.e., coarse land cover proportions used as input for SPM) are a convolution of not only sub-pixels within the coarse pixel, but also sub-pixels from neighboring coarse pixels. Based on this finding, a new SPM method based on optimization is developed which recognizes the optimal solution as the one that when convolved with the PSF, is the same as the input coarse land cover proportion. Experimental results on three separate datasets show that the SPM accuracy can be increased by considering the PSF effect. [ABSTRACT FROM AUTHOR]

Sentinel-2 is a very new programme of the European Space Agency (ESA) that is designed for fine spatial resolution global monitoring. Sentinel-2 images provide four 10 m bands and six 20 m bands. To provide more explicit spatial information, this paper aims to downscale the six 20 m bands to 10 m spatial resolution using the four directly observed 10 m bands. The outcome of this fusion task is the production of 10 Sentinel-2 bands with 10 m spatial resolution. This new fusion problem involves four fine spatial resolution bands, which is different to, and more complex than, the common pan-sharpening fusion problem which involves only one fine band. To address this, we extend the existing two main families of image fusion approaches (i.e., component substitution, CS, and multiresolution analysis, MRA) with two different schemes, a band synthesis scheme and a band selection scheme. Moreover, the recently developed area-to-point regression kriging (ATPRK) approach was also developed and applied for the Sentinel-2 fusion task. Using two Sentinel-2 datasets released online, the three types of approaches (eight CS and MRA-based approaches, and ATPRK) were compared comprehensively in terms of their accuracies to provide recommendations for the task of fusion of Sentinel-2 images. The downscaled ten-band 10 m Sentinel-2 datasets represent important and promising products for a wide range of applications in remote sensing. They also have potential for blending with the upcoming Sentinel-3 data for fine spatio-temporal resolution monitoring at the global scale. [ABSTRACT FROM AUTHOR]

Land cover/land use (LCLU) information extraction from multitemporal sequences of remote sensing imagery is becoming increasingly important. Mixed pixels are a common problem in Landsat and MODIS images that are used widely for LCLU monitoring. Recently developed subpixel mapping (SPM) techniques can extract LCLU information at the subpixel level by dividing mixed pixels into subpixels to which hard classes are then allocated. However, SPM has rarely been studied for time-series images (TSIs). In this paper, a spatiotemporal SPM approach was proposed for SPM of TSIs. In contrast to conventional spatial dependence-based SPM methods, the proposed approach considers simultaneously spatial and temporal dependences, with the former considering the correlation of subpixel classes within each image and the latter considering the correlation of subpixel classes between images in a temporal sequence. The proposed approach was developed assuming the availability of one fine spatial resolution map which exists among the TSIs. The SPM of TSIs is formulated as a constrained optimization problem. Under the coherence constraint imposed by the coarse LCLU proportions, the objective is to maximize the spatiotemporal dependence, which is defined by blending both spatial and temporal dependences. Experiments on three data sets showed that the proposed approach can provide more accurate subpixel resolution TSIs than conventional SPM methods. The SPM results obtained from the TSIs provide an excellent opportunity for LCLU dynamic monitoring and change detection at a finer spatial resolution than the available coarse spatial resolution TSIs. [ABSTRACT FROM AUTHOR]

Pan-sharpening is a technique to combine the fine spatial resolution panchromatic (PAN) band with the coarse spatial resolution multispectral bands of the same satellite to create a fine spatial resolution multispectral image. In this paper, area-to-point regression kriging (ATPRK) is proposed for pan-sharpening. ATPRK considers the PAN band as the covariate. Moreover, ATPRK is extended with a local approach, called adaptive ATPRK (AATPRK), which fits a regression model using a local, non-stationary scheme such that the regression coefficients change across the image. The two geostatistical approaches, ATPRK and AATPRK, were compared to the 13 state-of-the-art pan-sharpening approaches summarized in Vivone et al. (2015) in experiments on three separate datasets. ATPRK and AATPRK produced more accurate pan-sharpened images than the 13 benchmark algorithms in all three experiments. Unlike the benchmark algorithms, the two geostatistical solutions precisely preserved the spectral properties of the original coarse data. Furthermore, ATPRK can be enhanced by a local scheme in AATRPK, in cases where the residuals from a global regression model are such that their spatial character varies locally. [ABSTRACT FROM AUTHOR]

The availability of the panchromatic (PAN) band in remote sensing images gives birth to so-called image fusion techniques for increasing the spatial resolution of images to that of the PAN band. The spatial resolution of such spatially sharpened images, such as for the MODIS and Landsat sensors, however, may not be sufficient to provide the required detailed land-cover/land-use information. This paper proposes an area-to-point regression kriging (ATPRK)-based geostatistical solution to increase the spatial resolution of remote sensing images beyond that of any input images, including the PAN band. The new approach is a two-stage approach, including covariate downscaling and ATPRK-based image fusion. The new approach treats the PAN band as the covariate and takes advantages of its textural information. It explicitly accounts for the size of support, spatial correlation, and the point spread function of the sensor and has the characteristic of perfect coherence with the original coarse data. Moreover, the new downscaling approach can be extended readily by incorporating other ancillary information. The proposed approach was examined using both Landsat and MODIS images. The results show that it can produce more accurate sharpened images than four benchmark approaches. [ABSTRACT FROM AUTHOR]

The first seven bands of the Moderate Resolution Imaging Spectroradiometer (MODIS) data have been used widely for global land-cover/land-use (LCLU) monitoring (e.g., deforestation over the Amazon basin). However, the spatial resolution of MODIS bands 3–7 (i.e., 500 m) is coarser than that of bands 1 and 2 (i.e., 250 m), and may be too coarse for a large number of applications. In this paper, a new geostatistical approach based on area-to-point regression kriging (ATPRK) is proposed for downscaling coarse spatial resolution bands 3–7 such as to produce a complete set of MODIS images at 250 m. ATPRK takes advantages of the fine spatial resolution information in bands 1 and 2 by regression modeling, and uses area-to-point kriging to downscale the coarse residuals from the regression. ATPRK was compared to four existing methods, including the principal component analysis, wavelets, high-pass filter and kriging with external drift (KED) methods for downscaling in two experiments on MODIS data from the Brazilian Amazon. Both visual and quantitative evaluations (in terms of the root mean square error, correlation coefficient, relative global-dimensional synthesis error, universal image quality index, spectral angle mapper and spectral information divergence) showed that ATPRK produced sharpened images with the greatest quality. In addition, ATPRK perfectly preserved the spectral properties of the original coarse data and was faster than KED. The results reveal the great potential of ATPRK applied to MODIS data for a wide variety of applications, including global monitoring of deforestation. The ATPRK proposed in this paper is an entirely new image fusion approach based on a new conceptualization. [ABSTRACT FROM AUTHOR]

Haze contamination is a very common issue in remote sensing images, which inevitably limits data usability and further applications. Several methods have been developed for haze removal, which is an ill-posed problem. However, most of these methods involve various strong assumptions coupled with manually-determined parameters, which limit their generalization to different scenarios. Moreover, temporal information amongst time-series images has rarely been considered in haze removal. In this paper, the temporal information is proposed to be incorporated for more reliable haze removal, and guided by this general idea, a temporal information injection network (TIIN) is developed. The proposed TIIN solution for haze removal extracts the useful information in the temporally neighboring images provided by the regular revisit of satellite sensors. The TIIN method is suitable for images with various haze levels. Moreover, TIIN is also applicable for temporal neighbors with inherent haze or land cover changes due to a long-time interval between images. The proposed method was validated through experiments on both simulated and real haze images as well as comparison with five state-of-the-art benchmark methods. This research provides a new paradigm for enhancing haze removal by incorporating temporally neighboring images. • Temporal information is incorporated to enhance haze removal. • A deep learning-based haze removal method is proposed to use temporal information. • More accurate results can be produced by incorporating temporal information. • The method is applicable to various haze contamination. • The method is applicable to temporal neighbors with haze and land cover changes. [ABSTRACT FROM AUTHOR]

The soft-then-hard sub-pixel mapping (STHSPM) algorithm is a type of sub-pixel mapping (SPM) algorithm that first estimates the soft class values for sub-pixels at the target fine spatial resolution and then predicts the hard class labels for sub-pixels. In this article, four fast STHSPM algorithms (i.e. bilinear, bicubic, kriging, and radial basis function interpolation) were enhanced by using multiple shifted images (MSIs). The proportion images of the MSIs were first downscaled to the desired fine spatial resolution and then the multiple downscaled images were integrated for each class, followed by the class allocation process. Three remote-sensing images were used to test the proposed methods, and the results showed that MSIs can help to increase the SPM accuracy of the four STHSPM algorithms. The approach to incorporating MSIs into the STHSPM algorithms is non-iterative and fast. [ABSTRACT FROM PUBLISHER]

Abstract: In this paper, a new sub-pixel mapping (SPM) method based on radial basis function (RBF) interpolation is proposed for land cover mapping at the sub-pixel scale. The proposed method consists of sub-pixel soft class value estimation and subsequent class allocation for each sub-pixel. The sub-pixel soft class values are calculated by RBF interpolation. Taking the coarse proportion images as input, an interpolation model is built for each visited coarse pixel. First, the spatial relations between any sub-pixel within a visited coarse resolution pixel and its surrounding coarse resolution pixels are quantified by the basis function. Second, the coefficients indicating the contributions from neighboring coarse pixels are calculated. Finally, the basis function values are weighted by the coefficients to predict the sub-pixel soft class values. In the class allocation process, according to the class proportions and estimated soft class values, sub-pixels are allocated one of each available class in turn. Three remote sensing images were tested and the new method was compared to bilinear-, bicubic-, sub-pixel/pixel spatial attraction model- and Kriging-based SPM methods. Results show that the proposed RBF interpolation-based SPM is more accurate. Hence the proposed method provides an effective new option for SPM. [Copyright &y& Elsevier]

Fuzzyc-means (FCM) is a widely used unsupervised classifier for remote sensing images. This letter presents an uncertainty analysis-based FCM (UAFCM) classification method. The uncertainty in this letter refers to the discriminative ability of class attributes in fuzzy classification on a per-pixel basis. UAFCM is performed by analysing the uncertainty in FCM classification result and reclassifying the pixels with large uncertainty. Specifically, the uncertainty in FCM classification is measured by entropy and a proposed square error-based criterion. A threshold value is then determined to recognize the pixels with large uncertainty, which are reclassified with spatial connectivity subsequently. Experiments on three remote sensing images show that the proposed UAFCM consistently obtains more accurate classification results than does FCM, and hence provides an effective new unsupervised classification method for remote sensing images. [ABSTRACT FROM AUTHOR]

Mixed pixels are widely existent in remote-sensing imagery. Although the proportion occupied by each class in mixed pixels can be determined by spectral unmixing, the spatial distribution of classes remains unknown. Sub-pixel mapping (SPM) addresses this problem and a sub-pixel/pixel spatial attraction model (SPSAM) has been introduced to realize SPM. However, this algorithm fails to adequately consider the correlation between sub-pixels. Consequently, the SPM results created by SPSAM are noisy and the accuracy is limited. In this article, a method based on particle swarm optimization is proposed as post-processing on the SPM results obtained with SPSAM. It searches the most likely spatial distribution of classes in each coarse pixel to improve the SPSAM. Experimental results show that the proposed method can provide higher accuracy and reduce the noise in the results created by SPSAM. When compared with the available modified pixel-swapping algorithm, the proposed method often yields higher accuracy results. [ABSTRACT FROM AUTHOR]

Mixed pixels exist widely in remotely sensed images. To obtain more reliable land cover information than traditional hard classification, spectral unmixing methods have been developed to estimate the composition of the mixed pixels, in terms of the proportions of land cover classes. The existing spectral unmixing methods usually require pure spectra (i.e., endmembers) of each land cover class. However, in areas dominated by mixed pixels (e.g., highly heterogeneous areas), it can be a great challenge to extract a large number of pure endmembers, especially for long time-series data. Meanwhile, intra-class spectral variation remains a long-standing issue in spectral unmixing. In this paper, we propose a spatio-temporal spectral unmixing (STSU) approach to address these issues. The proposed method extends spectral unmixing from the traditional spatial domain to the spatio-temporal domain. It exploits fully the multi-scale spatio-temporal information, by using temporally neighboring fine spatial resolution images to detect land cover changes and, further, extracts the proportion information of unchanged mixed pixels required for training. The STSU method is free of the need for endmember extraction, using directly the extracted mixed training samples to construct a learning model, and it accounts for intra-class spectral variation. Therefore, it is a fully automatic method suitable for dynamic monitoring of land cover changes. The effectiveness of the STSU method was validated through experiments on Moderate Resolution Imaging Spectroradiometer (MODIS) data in five different areas. The proposed STSU method provides a new solution for spectral unmixing of time-series data based on the goal of continuous monitoring at the global scale. • A spatio-temporal spectral unmixing (STSU) approach is proposed. • Proportion information of detected unchanged pixels is extracted for training. • The proposed STSU method is free of the need for endmember extraction. • The proposed learning-based method can account for intra-class spectral variation. • The STSU method is suitable for dynamic monitoring of land cover changes. [ABSTRACT FROM AUTHOR]

Hyperspectral images (HSIs) always contain abundant spectral and spatial information. Most of the existing deep learning-based hyperspectral anomaly detection methods consider spectral differences between the background and anomalies, and the local spatial information is usually ignored. To make complete use of the spatial–spectral information, this article proposed a local spatial–spectral information-integrated semisupervised two-stream network (LS3T-Net) for hyperspectral anomaly detection. The two-stream network comprises an adaptive convolution and fully connected network and a variational autoencoder (VAE). The adaptive convolution and fully connected network is used to extract the local spatial features of patches, while VAE is trained to learn spectral information close to the background pixels. Furthermore, the detection maps from the two-stream network are incorporated through a process combining the benefits of spatial learning and spectral learning. This enhances the ability to separate the background and anomalies and suppress the false alarm. The experimental results for six real HSI datasets reveal that LS3T-Net can produce more accurate detection results than seven popular benchmark methods. [ABSTRACT FROM AUTHOR]

The Terra/Aqua MODerate resolution Imaging Spectroradiometer (MODIS) data have been used widely for global monitoring of the Earth's surface due to their daily fine temporal resolution. The spatial resolution of MODIS time-series (i.e., 500 m), however, is too coarse for local monitoring. A feasible solution to this problem is to downscale the coarse MODIS images, thus creating time-series images with both fine spatial and temporal resolutions. Generally, the downscaling of MODIS images can be achieved by fusing them with fine spatial resolution images (e.g., Landsat images) using spatio-temporal fusion methods. Among the families of spatio-temporal fusion methods, spatial unmixing-based methods have been applied widely owing to their lighter dependence on the available fine spatial resolution images. However, all techniques within this class of method suffer from the same serious problem, that is, the block effect, which reduces the prediction accuracy of spatio-temporal fusion. To our knowledge, almost no solution has been developed to tackle this issue directly. To address this need, this paper proposes a blocks-removed spatial unmixing (SU-BR) method, which removes the blocky artifacts by including a new constraint constructed based on spatial continuity. SU-BR provides a flexible framework suitable for any existing spatial unmixing-based spatio-temporal fusion method. Experimental results on a heterogeneous region, a homogeneous region and a region experiencing land cover changes show that SU-BR removes the blocks effectively and increases the prediction accuracy obviously in all three regions. SU-BR also outperforms two popular spatio-temporal fusion methods. SU-BR, thus, provides a crucial solution to overcome one of the longest standing challenges in spatio-temporal fusion. • All spatial unmixing-based fusion methods suffer from the block effect problem. • A blocks-removed spatial unmixing (SU-BR) method is proposed. • SU-BR improves existing spatial unmixing-based methods. • SU-BR is suitable for the existing spatial unmixing-based fusion method. [ABSTRACT FROM AUTHOR]

Deep learning-based methods (e.g., convolutional neural network (CNN)-based methods) have shown increasing potential in hyperspectral image (HSI) change detection (CD). However, the recent advances in CNN-based methods in HSI CD tasks are mostly devoted to designing more complex architectures or adding additional hand-designed blocks. This increases the number of parameters making model training difficult. In this article, we propose an end-to-end residual self-calibrated network (RSCNet) to increase the accuracy of HSI CD. To fully exploit the spatial information, the proposed RSCNet method adaptively builds interspatial and interspectral dependencies around each spatial location with fewer extra parameters and reduced complexity. Moreover, the introduced self-calibrated convolution (SCConv) helps to generate more discriminative representations by heterogeneously exploiting convolutional filters nested in the convolutional layer. The designed RSC module can explicitly incorporate richer information by introducing response calibration operation. The experiments on four bitemporal HSI datasets demonstrated that the proposed RSCNet method is more accurate than ten widely used benchmark methods. [ABSTRACT FROM AUTHOR]

The point spread function (PSF) effect is ubiquitous in remote sensing images and imposes a fundamental uncertainty on subpixel mapping (SPM). The crucial PSF effect has been neglected in existing SPM methods. This paper proposes a general model to reduce the PSF effect in SPM. The model is applicable to any SPM methods treating spectral unmixing as pre-processing. To demonstrate the advantages of the new technique it was necessary to develop a new approach for accuracy assessment of SPM. To-date, accuracy assessment for SPM has been limited to subpixel classification accuracy, ignoring the performance of reproducing spatial structure in downscaling. In this paper, a new accuracy index is proposed which considers SPM performances in classification and restoration of spatial structure simultaneously. Experimental results show that by considering the PSF effect, more accurate SPM results were produced and small-sized patches and elongated features were restored more satisfactorily. Moreover, using the novel accuracy index, the quantitative evaluation was found to be more consistent with visual evaluation. This paper, thus, addresses directly two of the longest standing challenges in SPM (i.e., the limitations of the PSF effect and accuracy assessment undertaken only on a subpixel-by-subpixel basis). • The PSF effect influences sub-pixel mapping (SPM) greatly and cannot be ignored. • A general solution is proposed to deal with the PSF effect in SPM. • A new accuracy index is proposed for comprehensive accuracy assessment of SPM. • SPM can be enhanced by considering the PSF effect. • Quantitative evaluation using the index is more consistent with visual evaluation. [ABSTRACT FROM AUTHOR]

Spatio-temporal fusion is a technique used to produce images with both fine spatial and temporal resolution. Generally, the principle of existing spatio-temporal fusion methods can be characterized by a unified framework of prediction based on two parts: (i) the known fine spatial resolution images (e.g., Landsat images), and (ii) the fine spatial resolution increment predicted from the available coarse spatial resolution increment (i.e., a downscaling process), that is, the difference between the coarse spatial resolution images (e.g., MODIS images) acquired at the known and prediction times. Owing to seasonal changes and land cover changes, there always exist large differences between images acquired at different times, resulting in a large increment and, further, great uncertainty in downscaling. In this paper, a virtual image pair-based spatio-temporal fusion (VIPSTF) approach was proposed to deal with this problem. VIPSTF is based on the concept of a virtual image pair (VIP), which is produced based on the available, known MODIS-Landsat image pairs. We demonstrate theoretically that compared to the known image pairs, the VIP is closer to the data at the prediction time. The VIP can capture more fine spatial resolution information directly from known images and reduce the challenge in downscaling. VIPSTF is a flexible framework suitable for existing spatial weighting- and spatial unmixing-based methods, and two versions VIPSTF-SW and VIPSTF-SU are, thus, developed. Experimental results on a heterogeneous site and a site experiencing land cover type changes show that both spatial weighting- and spatial unmixing-based methods can be enhanced by VIPSTF, and the advantage is particularly noticeable when the observed image pairs are temporally far from the prediction time. Moreover, VIPSTF is free of the need for image pair selection and robust to the use of multiple image pairs. VIPSTF is also computationally faster than the original methods when using multiple image pairs. The concept of VIP provides a new insight to enhance spatio-temporal fusion by making fuller use of the observed image pairs and reducing the uncertainty of estimating the fine spatial resolution increment. • VIPSTF is proposed for spatio-temporal fusion based on the new concept of VIP. • The VIP is closer to the data at the prediction time in the feature space. • VIPSTF is suitable for spatial weighting- and spatial unmixing-based methods. • VIPSTF is especially advantageous when the observed image pairs are temporally far. • VIPSTF is robust to the use of multiple image pairs and computationally faster. [ABSTRACT FROM AUTHOR]

Remote sensing images contain abundant land cover information. Due to the complex nature of land cover, however, mixed pixels exist widely in remote sensing images. Sub-pixel mapping (SPM) is a technique for predicting the spatial distribution of land cover classes within mixed pixels. As an ill-posed inverse problem, the uncertainty of prediction cannot be eliminated and hinders the production of accurate sub-pixel maps. In contrast to conventional methods that use continuous geospatial information (e.g., images) to enhance SPM, in this paper, a SPM method with point constraints into SPM is proposed. The method of fusing point constraints is implemented based on the pixel swapping algorithm (PSA) and utilizes the auxiliary point information to reduce the uncertainty in the SPM process and increase map accuracy. The point data are incorporated into both the initialization and optimization processes of PSA. Experiments were performed on three images to validate the proposed method. The influences of the performances were also investigated under different numbers of point data, different spatial characters of land cover and different zoom factors. The results show that by using the point data, the proposed SPM method can separate more small-sized targets from aggregated artifacts and the accuracies are increased obviously. The proposed method is also more accurate than the advanced radial basis function interpolation-based method. The advantage of using point data is more evident when the point data size and scale factor are large and the spatial autocorrelation of the land cover is small. As the amount of point data increases, however, the increase in accuracy becomes less noticeable. Furthermore, the SPM accuracy can still be increased even if the point data and coarse proportions contain errors. • Sub-pixel mapping is an ill-posed problem with inevitable uncertainty. • The point data are proposed to enhance sub-pixel mapping. • A new sub-pixel mapping method with point constraint is proposed. • More accurate sub-pixel maps can be produced by exerting point constraints. [ABSTRACT FROM AUTHOR]

Deep learning methods, especially convolutional neural network (CNN)-based methods, have shown promising performance for hyperspectral image (HSI) change detection (CD). It is acknowledged widely that different spectral channels and spatial locations in input image patches may contribute differently to CD. However, they are treated equally in existing CNN-based approaches. To increase the accuracy of HSI CD, we propose an end-to-end Siamese CNN (SiamNet) with a spectral–spatial-wise attention (SSA-SiamNet) mechanism. The proposed SSA-SiamNet method can emphasize informative channels and locations and suppress less informative ones to refine the spectral–spatial features adaptively. Moreover, in the network training phase, the weighted contrastive loss function is used for more reliable separation of changed and unchanged pixels and to accelerate the convergence of the network. SSA-SiamNet was validated using four groups of bitemporal HSIs. The accuracy of CD using the SSA-SiamNet was found to be consistently greater than for ten benchmark methods. [ABSTRACT FROM AUTHOR]

An accurate and timely inventory for major landslides triggered by seismic events is essential for hazard assessment and risk governance. Sentinel-2 MultiSpectral Instrument (MSI) data have been increasingly used for co-seismic landslide mapping in recent years. However, there have been few studies exploring the full use of Sentinel-2 near infrared (NIR) narrow, red-edge, and short-wave infrared (SWIR) bands for landslide mapping. In this study, Sentinel-2 10-m fused NIR narrow, red-edge-1, SWIR-1, and SWIR-2 bands were incorporated into the change detection-based Markov random field (CDMRF) that considers the spectral, spatial, and contextual information of landslides. The effectiveness of the developed method was proved in mapping co-seismic landslides triggered by two major earthquakes in the Yarlung Tsangpo Grand Canyon, China, in 2017 and Hokkaido, Japan, in 2018. The results demonstrated that the mapping performances with the fused Sentinel-2 bands were significantly improved for both seismic events. Such improvement is mainly attributed to the more abundant spectral information provided by the fused Sentinel-2 red-edge, NIR narrow, and SWIR bands. The novelties of this study are (1) the preliminary attempt to use the Sentinel-2 10-m fused bands for landslide mapping, and (2) the synergy of Sentinel-2 red-edge, NIR narrow, and SWIR bands for more reliable landslide mapping. [ABSTRACT FROM AUTHOR]

Many remote sensing soil moisture (SM) products have been developed with global coverage. However, most of them are derived from passive microwave observations with very coarse resolution, greatly constraining the applications at regional scales. To increase the spatial resolution, a downscaling method is developed to downscale the 36-km Soil Moisture Active Passive L3 SM (SMAP SM) product to 1 km using the Moderate Resolution Imaging Spectroradiometer (MODIS) products (8-d land surface temperature, LST, and 16-d normalized difference vegetation index, NDVI). In this method, a linking model is first established between SM and LST and NDVI, and a self-adaptive window method is applied with the use of the geographically weighted regression (GWR) method to obtain an optimal local regression. Then, the uncertainty of the linking model, expressed as the regression residual, is redistributed to fine-resolution pixels to analyze the consistency before and after downscaling. The method was applied to the Iberian Peninsula to produce the 8-d downscaled SM product in 2016. The downscaled SM was validated with the in-situ SM network (REMEDHUS). A good agreement was found between the two data sets, with a correlation coefficient ($R$) of 0.87 and an unbiased root-mean-squared error (ubRMSE) of 0.043 m3/m3 at a network level. At station level, the $R$ is larger than 0.6 for all the REMEDHUS stations, with an ubRMSE smaller than 0.06 m3/m3. The evaluation indicates the good potential of the proposed method in the SM downscaling, which achieves a robust consistency and provides rich spatial information while maintaining good accuracy. [ABSTRACT FROM AUTHOR]

Sentinel-1A synthetic aperture radar (SAR) data present an opportunity for acquiring crop information without restrictions caused by weather and illumination conditions, at a spatial resolution appropriate for individual rice fields and a temporal resolution sufficient to capture the growth profiles of different crop species. This study investigated the use of multi-temporal Sentinel-1A SAR data and Landsat-derived normalized difference vegetation index (NDVI) data to map the spatial distribution of paddy rice fields across parts of the Sanjiang plain, in northeast China. The satellite sensor data were acquired throughout the rice crop-growing season (May–October). A co-registered set of 10 dual polarization (VH/VV) SAR and NDVI images depicting crop phenological development were used as inputs to Support Vector Machine (SVM) and Random Forest (RF) machine learning classification algorithms in order to map paddy rice fields. The results showed a significant increase in overall classification when the NDVI time-series data were integrated with the various combinations of multi-temporal polarization channels (i.e. VH, VV, and VH/VV). The highest classification accuracies overall (95.2%) and for paddy rice (96.7%) were generated using the RF algorithm applied to combined multi-temporal VH polarization and NDVI data. The SVM classifier was most effective when applied to the dual polarization (i.e. VH and VV) SAR data alone and this generated overall and paddy rice classification accuracies of 91.6% and 82.5%, respectively. The results demonstrate the practicality of implementing RF or SVM machine learning algorithms to produce 10 m spatial resolution maps of paddy rice fields with limited ground data using a combination of multi-temporal SAR and NDVI data, where available, or SAR data alone. The methodological framework developed in this study is apposite for large-scale implementation across China and other major rice-growing regions of the world. [ABSTRACT FROM PUBLISHER]

This paper presents a novel adaptive fuzzy local information c-means (ADFLICM) clustering approach for remotely sensed imagery classification by incorporating the local spatial and gray level information constraints. The ADFLICM approach can enhance the conventional fuzzy c-means algorithm by producing homogeneous segmentation and reducing the edge blurring artifact simultaneously. The major contribution of ADFLICM is use of the new fuzzy local similarity measure based on pixel spatial attraction model, which adaptively determines the weighting factors for neighboring pixel effects without any experimentally set parameters. The weighting factor for each neighborhood is fully adaptive to the image content, and the balance between insensitiveness to noise and reduction of edge blurring artifact to preserve image details is automatically achieved by using the new fuzzy local similarity measure. Four different types of images were used in the experiments to examine the performance of ADFLICM. The experimental results indicate that ADFLICM produces greater accuracy than the other four methods and hence provides an effective clustering algorithm for classification of remotely sensed imagery. [ABSTRACT FROM PUBLISHER]

Graph-based classification algorithms have gained increasing attention in semi-supervised classification. Nevertheless, the graph cannot fully represent the inherent spatial distribution of the data. In this paper, a new classification methodology based on the spatial-spectral Label Propagation is proposed for semi-supervised classification of hyperspectral imagery. The spatial information was used in two aspects: on the one hand, the spatial features extracted by a 2-D Gabor filter were stacked with spectral features; on the other hand, the width of the Gaussian function, which was used to construct graph, was determined with an adaptive method. Subsequently, the unlabeled samples from the spatial neighbors of the labeled samples were selected and the spatial graph was constructed based on spatial smoothness. Finally, labels were propagated from labeled samples to unlabeled samples with spatial-spectral graph to update the training set for a basic classifier (e.g., Support Vector Machine, SVM). Experiments on four hyperspectral datasets show that the proposed Spatial-Spectral Label Propagation based on the SVM (SS-LPSVM) can effectively represent the spatial information in the framework of semi-supervised learning and consistently produces greater classification accuracy than the standard SVM, the Laplacian Support Vector Machine (LapSVM), Transductive Support Vector Machine (TSVM) and the Spatial-Contextual Semi-Supervised Support Vector Machine (SCS 3 VM). [ABSTRACT FROM AUTHOR]

A new subpixel mapping (SPM) algorithm combining pixel-level and subpixel-level spatial dependences is proposed in this letter. The pixel-level dependence is measured by the spatial attraction model (SAM) with either surrounding or quadrant neighbourhood, while the subpixel-level dependence is characterized by either the mean filter or the exponential weighting function. Both pixel-level and subpixel-level dependences are then fused as the weighted dependence in the constructed objective function. The branch-and-bound algorithm is employed to solve the optimization problem, and thus, obtain the optimal spatial distribution of subpixel classes. An artificial image and a set of real remote sensing images were tested for validation of the proposed method. The results demonstrated that the proposed method can achieve results with greater accuracy than two traditional SPM methods and the mixed SAM method. Meanwhile, the proposed method needs less computation time than the mixed SAM, and hence it provides a new solution to subpixel land cover mapping. [ABSTRACT FROM AUTHOR]

Due to the limitations of hardware technology and budget constraints, there always exists a tradeoff between spatial and spectral resolutions in a hyperspectral image (HSI). Because of the limited spatial resolution, mixed pixels are a common issue in HSIs, and consequently, some targets appear as subpixels. The effectiveness of hyperspectral target detection is affected greatly by the subpixel targets, especially when the size of the targets is small. In this article, we proposed a double dictionary-based nonlinear representation model for hyperspectral subpixel target detection (DDNRTD). DDNRTD represents HSIs with a nonlinear model based on background and target dictionaries, which fully considers the spatial property of background and targets and can separate background and targets reliably, especially for small-sized subpixel targets. In addition, we designed an over-completed background dictionary construction strategy to represent the background part more effectively, which integrates spectral angle distance (SAD) with sparse representation. Experiments on two simulated and five real HSI datasets showed that the proposed DDNRTD method produced more accurate detection results than six state-of-the-art methods. [ABSTRACT FROM AUTHOR]

• A seamless downscaling method for ESA CCI SSM product is proposed in this study. • The method well overcomes the impact from the gaps in ESA CCI SSM and MODIS product. • The downscaled SSM data show good temporal consistency with field measurements. • The downscaled result provides a good spatio-temporal continuity in SSM variations. Spatial downscaling has recently become a crucial process in the regional application of coarse-resolution passive microwave surface soil moisture (SSM) products. Extensive gaps in auxiliary optical/thermal infrared observation data (mainly caused by cloud cover) and gaps in coarse-resolution passive microwave SSM data lead to spatiotemporal discontinuity in downscaled SSM maps, thereby limiting their applications. An improved downscaling method for the 25-km European Space Agency (ESA) Climate Change Initiative (CCI) SSM product was proposed to obtain daily seamless downscaled SSM series at a 1-km scale. The Moderate Resolution Imaging Spectroradiometer (MODIS) Terra daily land surface temperature (LST) and normalized difference vegetation index (NDVI) products were used as auxiliary data for the downscaling process. Prior to the spatial downscaling, an annual temperature cycle model was applied to the 1-km daily daytime LST data to fill the data gaps caused by cloud cover and to derive the spatial-seamless LST (gap-filled). Subsequently, the gap-filled ESA CCI SSM was generated at the original resolution based on the relationships among the SSM, LST, and NDVI. Finally, these were utilized to obtain a seamless downscaled series SSM at 1-km spatial resolution with a value-consistent downscaling method. The proposed method was applied to data obtained for the Iberian Peninsula from January 1, 2016 to December 31, 2018. Based on the comparison with the precipitation dataset, the downscaled SSM exhibited strong temporal correlation with rainfall events. Evaluation using the in situ SSM from the REMEDHUS network highlighted the good performance of the downscaled SSM at network level with a correlation coefficient (R) of 0.820. The root-mean-square-error, unbiased root-mean-square error (ub RMSE), and bias were 0.091, 0.033, and 0.085 m3/m3, respectively. A comparison with an alternative downscaled SSM product produced by the Barcelona Expert Center, one of the soil moisture and ocean salinity mission (SMOS)–downscaled SSM datasets, also indicated that the downscaled SSM has better spatiotemporal coverage and performance in terms of R and ub RMSE with reference to the REMEDHUS network. These results confirmed that the proposed method is an efficient and convenient downscaling process that can be used to generate high-resolution SSM data without spatiotemporal gaps. The downscaled SSM data improves the accuracy of the original passive microwave SSM product in describing regional SSM variations and shows good potential for related applications at regional scale. [ABSTRACT FROM AUTHOR]

Spectral unmixing is one of the important techniques for hyperspectral data processing. The analysis of spectral mixing is often based on a linear, fully constrained (FC) (i.e., nonnegative and sum-to-one mixture proportions), and least squares criterion. However, the traditional iterative processing of FC least squares (FCLS) linear spectral mixture analysis (LSMA) (FCLS-LSMA) is of heavy computational burden. Recently developed geometric LSMA methods decreased the complexity to some degree, but how to further reduce the computational burden and completely meet the FCLS criterion of minimizing the unmixing residual needs to be explored. In this paper, a simple distance measure is proposed, and then, a new geometric FCLS-LSMA method is constructed based on the distance measure. The method is in line with the FCLS criterion, free of iteration and dimension reduction, and with very low complexity. Experimental results show that the proposed method can obtain the same optimal FCLS solution as the traditional iteration-based FCLS-LSMA, and it is much faster than the existing spectral unmixing methods, particularly the traditional iteration-based method. [ABSTRACT FROM AUTHOR]

Optical earth observation satellite sensors often provide a coarse spatial resolution (CR) multispectral (MS) image together with a fine spatial resolution (FR) panchromatic (PAN) image. Pansharpening is a technique applied to such satellite sensor images to generate an FR MS image by injecting spatial detail taken from the FR PAN image while simultaneously preserving the spectral information of MS image. Pansharpening methods are mostly applied on a per-pixel basis and use the PAN image to extract spatial detail. However, many land cover objects in FR satellite sensor images are not illustrated as independent pixels, but as many spatially aggregated pixels that contain important semantic information. In this article, an object-based pansharpening approach, termed object-based area-to-point regression kriging (OATPRK), is proposed. OATPRK aims to fuse the MS and PAN images at the object-based scale and, thus, takes advantage of both the unified spectral information within the CR MS images and the spatial detail of the FR PAN image. OATPRK is composed of three stages: image segmentation, object-based regression, and residual downscaling. Three data sets acquired from IKONOS and Worldview-2 and 11 benchmark pansharpening algorithms were used to provide a comprehensive assessment of the proposed OATPRK approach. In both the synthetic and real experiments, OATPRK produced the most superior pan-sharpened results in terms of visual and quantitative assessment. OATPRK is a new conceptual method that advances the pixel-level geostatistical pansharpening approach to the object level and provides more accurate pan-sharpened MS images. [ABSTRACT FROM AUTHOR]

• Thermokarst lake dynamics had significant correlations with peripheral ground deformations. • Permafrost degradation extensively accelerated surrounding evolved thermokarst lakes. • Hydrological connectivity predominates spatial-temporal dynamics of thermokarst lakes. • Constructing the Qinghai-Tibet railway limitedly promoted permafrost degradation. As a common landscape feature of permafrost degradation, the formation and evolution/expansion of thermokarst lakes have strong impacts on hydrology, ecosystem, and biogeochemical cycle. Characterizing the dynamics of thermokarst lakes and quantifying its links to permafrost degradation on the Tibetan Plateau (TP) are essential but lacking. Here, we investigated three representative areas of permafrost degradation on the central TP, where extensive small thermokarst lakes (<0.1 km2) were found to develop rapidly. We reconstructed spatial-temporal changes of thermokarst lakes using Google Earth Engine (GEE) and ground deformation derived from Interferometric Synthetic Aperture Radar (InSAR) during 2016–2020. We found that permafrost degradation extensively accelerated surrounding the evolved thermokarst lakes. The thermokarst lake dynamics had significant correlations with peripheral ground deformation. We infer that hydrological connectivity is the main driving factor for promoting permafrost degradation and facilitating the development of thermokarst lakes on the central TP. Moreover, the impact of the recently-constructed Qinghai-Tibet railway on accelerating permafrost degradation may be overestimated. Our results suggest that thermokarst lake change is an important indicator of climate change on the central TP and thus deserves long-term monitoring. [ABSTRACT FROM AUTHOR]

Spatio-temporal image fusion aims to increase the frequency and resolution of multispectral satellite sensor images in a cost-effective manner. However, practical constraints on input data requirements and computational cost prevent a wider adoption of these methods in real case-studies. We propose an ensemble of strategies to eliminate the need for cloud-free matching pairs of satellite sensor images. The new methodology called Unpaired Spatio-Temporal Fusion of Image Patches (USTFIP) is tested in situations where classical requirements are progressively difficult to meet. Overall, the study shows that USTFIP reduces the root mean square error by 2-to-13% relative to the state-of-the-art Fit-FC fusion method, due to an efficient use of the available information. Implementation of USTFIP through parallel computing saves up to 40% of the computational time required for Fit-FC. [Display omitted] • New spatio-temporal fusion method to relax strict input requirements. • Efficient use of available information increases method accuracy. • Robust method that relies on temporally close data. • Advanced programming techniques to save running time. [ABSTRACT FROM AUTHOR]

Forest mapping from satellite sensor imagery provides important information for the timely monitoring of forest growth and deforestation, bioenergy potential assessment, and modeling of carbon flux, among others. Due to the daily global revisit rate and wide swath width, MODerate-resolution Imaging Spectroradiometer (MODIS) images are used commonly for satellite-derived forest mapping at both regional and global scales. However, the spatial resolution of MODIS images is too coarse to observe fine spatial variation in forest cover. The last few decades have seen the production of several fine-spatial-resolution satellite-derived global forest cover maps, such as Hansen’s global tree canopy cover map of 2000, which includes abundant spectral, temporal, and spatial prior information about forest cover at a fine spatial resolution. In this paper, a novel learning-based spatial–temporal superresolution mapping approach is proposed to integrate both current MODIS images and prior maps of Hansen’s tree canopy cover, to map present forest cover with a fine spatial resolution. The novel approach is composed of three main stages: 1) automatic generation of 240-m forest proportion images from both 240- and 480-m MODIS images using a nonlinear learning-based spectral unmixing method; 2) downscaling the 240-m forest proportion images to 30 m to predict the class possibilities at the subpixel scale using a temporal-example learning-based downscaling method; and 3) final production of the fine-spatial-resolution forest map by solving a regularization-based optimization problem. The novel approach produced more accurate fine-spatial-resolution forest cover maps in terms of both visual and quantitative evaluation than traditional pixel-based classification and the latest subpixel based superresolution mapping methods. The results show the great efficiency and potential of the novel approach for producing fine-spatial-resolution forest maps from MODIS images. [ABSTRACT FROM PUBLISHER]

Landslide mapping (LM) is essential for hazard prevention, mitigation, and vulnerability assessment. Despite the great efforts over the past few years, there is room for improvement in its accuracy and efficiency. Existing LM is primarily achieved using field surveys or visual interpretation of remote sensing images. However, such methods are highly labor-intensive and time-consuming, particularly over large areas. Thus, in this paper a change detection-based Markov random field (CDMRF) method is proposed for near-automatic LM from aerial orthophotos. The proposed CDMRF is applied to a landslide-prone site with an area of approximately 40 km 2 on Lantau Island, Hong Kong. Compared with the existing region-based level set evolution (RLSE), it has three main advantages: 1) it employs a more robust threshold method to generate the training samples; 2) it can identify landslides more accurately as it takes advantages of both the spectral and spatial contextual information of landslides; and 3) it needs little parameter tuning. Quantitative evaluation shows that it outperforms RLSE in the whole study area by almost 5.5% in Correctness and by 4% in Quality . To our knowledge, it is the first time CDMRF is used to LM from bitemporal aerial photographs. It is highly generic and has great potential for operational LM applications in large areas and also can be adapted for other sources of imagery data. [ABSTRACT FROM AUTHOR]

Landslide inventory mapping (LIM) is an increasingly important research topic in remote sensing and natural hazards. Past studies achieve LIM mainly using on-screen interpretation of aerial photos, and little attention has been paid to developing more automated methods. In recent years, the use of multitemporal remote sensing images makes it possible to map landslides semi-automatically. Although numerous methods have been proposed, only a few methods are competent for some specific situations and there is large room for improvement in their degree of automation. For these reasons, a semi-automated approach is proposed for reliable and accurate LIM from bitemporal aerial orthophotos. Specifically, it consists of two principal steps: 1) change detection-based thresholding (CDT) and 2) level set evolution (LSE). CDT is mainly used to generate the initial zero-level curve (ZLC) for LSE, thus automating the proposed method considerably. It includes three substeps: 1) generating difference image (DI) using change vector analysis (CVA), 2) detecting landslide candidates using a thresholding method, and 3) removing errors using morphology operations. Then, landslide boundaries are detected using two types of LSE, i.e., edge-based LSE (ELSE) and region-based LSE (RLSE). Finally, the effectiveness and advantages of the proposed methods are corroborated by a series of experiments. Given its efficiency and accuracy, it can be applied to rapid responses of natural hazards. This study is the first attempt to apply LSE to LIM from bitemporal remote sensing images. [ABSTRACT FROM AUTHOR]