Search

Beijing Satellite 3 is a high-performance optical remote sensing satellite with a spatial resolution of 0.3–0.5 m. It can provide timely and independent ultra-high-resolution spatial big data and comprehensive spatial information application services. At present, there is no relevant research on the fusion method of BJ-3A satellite images. In many applications, high-resolution panchromatic images alone are insufficient. Therefore, it is necessary to fuse them with multispectral images that contain spectral color information. Currently, there is a lack of research on the fusion method of BJ-3A satellite images. This article explores six traditional pixel-level fusion methods (HPF, HCS, wavelet, modified-IHS, PC, and Brovey) for fusing the panchromatic image and multispectral image of the BJ-3A satellite. The fusion results were analyzed qualitatively from two aspects: spatial detail enhancement capability and spectral fidelity. Five indicators, namely mean, standard deviation, entropy, correlation coefficient, and average gradient, were used for quantitative analysis. Finally, the fusion results were comprehensively evaluated from three aspects: spectral curves of ground objects, absolute error figure, and object-oriented classification effects. The findings of the research suggest that the fusion method known as HPF is the optimum and appropriate technique for fusing panchromatic and multispectral images obtained from BJ-3A. These results can be utilized as a guide for the implementation of BJ-3A panchromatic and multispectral data fusion in real-world scenarios. [ABSTRACT FROM AUTHOR]

In the face of complex scenarios, the information insufficiency of classification tasks dominated by a single modality has led to a bottleneck in classification performance. The joint application of multimodal remote sensing data for surface observation tasks has garnered widespread attention. However, issues such as sample differences between modalities and the lack of correlation in physical features have limited the performance of classification tasks. Establishing effective interaction between multimodal data has become another significant challenge. To fully integrate heterogeneous information from multiple modalities and enhance classification performance, this paper proposes a dual-branch cross-Transformer feature fusion network aimed at joint land cover classification of hyperspectral imagery (HSI) and Light Detection and Ranging (LiDAR) data. The core idea is to leverage the potential of convolutional operators to represent spatial features, combined with the advantages of the Transformer architecture in learning remote dependencies. The framework employs an improved self-attention mechanism to aggregate features within each modality, highlighting the spectral information of HSI and the spatial (elevation) information of LiDAR. The feature fusion module based on cross-attention integrates deep features from two modalities, achieving complementary information through cross-modal attention. The classification task is performed using jointly obtained spectral and spatial features. Experiments were conducted on three multi-source remote sensing classification datasets, demonstrating the effectiveness of the proposed model compared to existing methods. [ABSTRACT FROM AUTHOR]

Chlorophyll-a concentration (Chl-a) is an important indicator of coastal eutrophication. Remote sensing technology provides a global view of it. However, different types of sensors are subject to design constraints and cannot meet the requirements of high temporal and spatial resolution on nearshore engineering simultaneously. To obtain high-spatiotemporal-resolution images, this study examines the performance of the enhanced spatial and temporal adaptive reflectance fusion model (ESTARFM) on GOCI and Landsat Chl-a data fusion. Considering the rapidly changing rate and consistency of oceanic Chl-a, the ESTARFM was modified via segmented fitting and numerical conversion. The results show that both fusion models can fuse multiple data advantages to obtain high-spatiotemporal-resolution Chl-a images. Compared with the ESTARFM, the modified solution has a better performance in terms of the root mean square error and correlation coefficient, and its results have better spatial consistency for coastal Chl-a. In addition, the new solution expands the data utilization range of data fusion by reducing the influence of the time interval of original data and realizes better monitoring of nearshore Chl-a changes. [ABSTRACT FROM AUTHOR]

Leaf area index (LAI) is an important biophysical parameter of vegetation and serves as a significant indicator for assessing forest ecosystems. Multi-source remote sensing data enables large-scale and dynamic surface observations, providing effective data for quantifying various indices in forest and evaluating ecosystem changes. However, employing single-source remote sensing spectral or LiDAR waveform data poses limitations for LAI inversion, making the integration of multi-source remote sensing data a trend. Currently, the fusion of active and passive remote sensing data for LAI inversion primarily relies on empirical models, which are mainly constructed based on field measurements and do not provide a good explanation of the fusion mechanism. In this study, we aimed to estimate LAI based on physical model using both spectral imagery and LiDAR waveform, exploring whether data fusion improved the accuracy of LAI inversion. Specifically, based on the physical model geometric-optical and radiative transfer (GORT), a fusion strategy was designed for LAI inversion. To ensure inversion accuracy, we enhanced the data processing by introducing a constraint-based EM waveform decomposition method. Considering the spatial heterogeneity of canopy/ground reflectivity ratio in regional forests, calculation strategy was proposed to improve this parameter in inversion model. The results showed that the constraint-based EM waveform decomposition method improved the decomposition accuracy with an average 12% reduction in RMSE, yielding more accurate waveform energy parameters. The proposed calculation strategy for the canopy/ground reflectivity ratio, considering dynamic variation of parameter, effectively enhanced previous research that relied on a fixed value, thereby improving the inversion accuracy that increasing on the correlation by 5% to 10% and on R² by 62.5% to 132.1%. Based on the inversion strategy we proposed, data fusion could effectively be used for LAI inversion. The inversion accuracy achieved using both spectral and LiDAR data (correlation=0.81, R² = 0.65, RMSE=1.01) surpassed that of using spectral data or LiDAR alone. This study provides a new inversion strategy for large-scale and high-precision LAI inversion, supporting the field of LAI research. [ABSTRACT FROM AUTHOR]

In order to expand the application of remote sensing prospecting technology in gold prospecting, this paper takes the Zhuanshanzi area in Inner Mongolia as an example, and selects Landsat 8 OLI remote sensing data. On the basis of image data preprocessing, the regional tectonic information is enhanced by the fusion of B7B5B3 band. The principal component analysis is used to extract the alteration anomaly information through threshold segmentation. Based on the geological characteristics and remote sensing interpretation information of the region and Zhuanshanzi Gold Deposit, the prospecting indicator of gold deposits in Zhuanshanzi area is established, and 3 prospect areas for prospecting prediction are delineated. The field verification shows that there is gold mineralization in the prospect area. It is believed that the linear and ring structure and remote sensing anomaly information extraction based on OLI remote sensing images can be used as auxiliary means for gold prospecting prediction. [ABSTRACT FROM AUTHOR]

In a world in continuous evolution and in which human needs grow exponentially according to the increasing world population, the advent of new technologies plays a fundamental role in all fields of industry, especially in agriculture. Optimizing times, automating machines, and guaranteeing product quality are key objectives in the field of Agriculture 4.0, which integrates various innovative technologies to meet the needs of producers and consumers while guaranteeing respect for the environment and the planet's resources. In this context, our research aims to propose an integrated system using data coming from an innovative experimental atmospheric and forecasting simulator (capable of predicting some characteristic climate variables subsequently validated with local sensors), combined with indices deriving from Remote Sensing and UAV images (treated with the data fusion method), that can give fundamental information related to Agriculture 4.0 with particular reference to the subsequent phases of system automation. These data, in fact, can be collected in an open-source GIS capable of displaying areas that need irrigation and fertilization and, moreover, establishing the path of an automated drone for the monitoring of the crops and the route of a self-driving tractor for the irrigation of the areas of interest. [ABSTRACT FROM AUTHOR]

The NASA Terra and Aqua satellites have been successfully operating for over two decades, exceeding their original design life. However, the era of NASA's Earth Observing System may be coming to a close as early as 2023. Similarities between the Moderate Resolution Imaging Spectroradiometer (MODIS), aboard Aqua and Terra, and the Visible Infrared Imaging Radiometer Suite (VIIRS) sensors aboard the Suomi NPP, NOAA‐20 and NOAA‐21 satellites enable potential continuity of long‐term earth observational records in the VIIRS era. We conducted a comprehensive calibration and validation of the MODIS MOD17 product, which provided the first global, continuous, weekly estimates of ecosystem gross primary productivity (GPP) and annual estimates of net primary productivity (NPP). Using Bayesian model‐data fusion, we combined 18 years of tower fluxes with prior data on plant traits and hundreds of field measurements of NPP to benchmark MOD17 and to develop the first terrestrial productivity estimates from VIIRS. The updated mean global GPP (NPP) flux from the future MOD17 Collection 7 product and new VNP17 product for 2012–2018 is 127 ± 2.8 Pg C year−1 (58 ± 1.1 Pg C year−1), which compares well with independent top‐down and bottom‐up estimates. MOD17 and VNP17 depict upward productivity trends over recent decades, with 2000–2018 MOD17 GPP (NPP) rising by 0.47 (0.25) Pg C year−2 but slowing to 0.35–0.44 (0.11–0.13) Pg C year−2 over 2012–2021, with a greater reduction in the NPP growth rate. The new VIIRS VNP17 product has the potential to extend these estimates of global, terrestrial primary productivity beyond 2030. Plain Language Summary: The NASA Terra and Aqua satellites have been successfully operating for over two decades, far longer than their original 5‐year design life. However, both satellites may be unable to maintain their original orbits as early as 2023. These satellites carry the Moderate Resolution Imaging Spectroradiometer (MODIS) sensors, which are very similar to the Visible Infrared Imaging Radiometer Suite (VIIRS) sensors aboard newer satellites. The long record of MODIS data collected so far may therefore be extended by the VIIRS sensors, particularly the global estimates of the amount of carbon in the atmosphere taken up and stored by plants. We used multiple independent data sets to figure out if and how the MODIS MOD17 computer model should be changed to improve its accuracy and to use data from VIIRS. The new VIIRS VNP17 data could extend our record of plant‐atmosphere carbon exchange beyond the year 2030. Key Points: Over two decades of global productivity estimates from Moderate Resolution Imaging Spectroradiometer (MODIS) cannot be continued without use of Visible Infrared Imaging Radiometer Suite (VIIRS) dataWe performed a comprehensive calibration and validation, and sensitivity and uncertainty analyses of MODIS MOD17 and new VIIRS VNP17Both MOD17 and new VNP17 depict upward productivity trends and mean and interannual variability consistent with independent data [ABSTRACT FROM AUTHOR]

The fusion of dissimilar data modalities in neural networks presents a significant challenge, particularly in the case of multimodal hyperspectral and lidar data. Hyperspectral data, typically represented as images with potentially hundreds of bands, provide a wealth of spectral information, while lidar data, commonly represented as point clouds with millions of unordered points in 3D space, offer structural information. The complementary nature of these data types presents a unique challenge due to their fundamentally different representations requiring distinct processing methods. In this work, we introduce an alternative hyperspectral data representation in the form of a hyperspectral point cloud (HSPC), which enables ingestion and exploitation with point cloud processing neural network methods. Additionally, we present a composite fusion-style, point convolution-based neural network architecture for the semantic segmentation of HSPC and lidar point cloud data. We investigate the effects of the proposed HSPC representation for both unimodal and multimodal networks ingesting a variety of hyperspectral and lidar data representations. Finally, we compare the performance of these networks against each other and previous approaches. This study paves the way for innovative approaches to multimodal remote sensing data fusion, unlocking new possibilities for enhanced data analysis and interpretation. [ABSTRACT FROM AUTHOR]

Space Archaeology (SA), also known as Satellite Archaeology, Satellite Remote Sensing for Archaeology, or Archaeology from Space, is part of the wider interdisciplinary field of Remote Sensing for Archaeology. The application of satellite data in archaeological investigations has proven useful for landscape observation and analysis, the detection of archaeological traces, the reconstruction and monitoring of natural and anthropic processes, and the management and promotion of archaeological heritage. During the last few decades, the increasing number of SA studies has demonstrated innovative growth in archaeological disciplines due to the significant enhancement of spatial technologies, the advancement of visual inspection and image processing techniques, the development of data fusion methodologies, and the improvement of multi-temporal analysis methods. Therefore, a broad overview of the current situation in the concepts and approaches of SA is necessary to gain greater awareness of the current potentialities and limitations of this science to better address future studies. The present work provides a review of the scientific literature by exploring the different aspects of SA and the goals achieved to date in three main application fields: Detection, monitoring, and promotion of archaeological heritage. The contributions reviewed were divided within each of these three categories by analyzing the type of SA data and methods applied. The results indicate that (1) most studies aim to detect archaeological sites compared to monitoring and promotion; (2) optical images are used more than SAR data; and (3) techniques of image enhancement and visual interpretation are applied more than other data processing methods. This brief overview allows us to consider SA as an evolving discipline, an engine for cross-skills training, as well as a promising future science that can play a key role in the frontier of sustainable development and the new space economy. [ABSTRACT FROM AUTHOR]

Objectives: The scale of buildings and their distribution is key indicators to measure the economic and social development of a region. Therefore, it is significant to study the extraction of buildings based on remote sensing images. Existing neural network methods still have shortcomings in the completeness of building extraction and the accuracy of building edges. To solve the above problems, this paper proposes a multi-level feature fusion network (MFFNet) based on high-resolution images. Methods: Firstly, we use edge detection operators to improve the ability of the network to recognize the boundaries of buildings. Secondly, we use a multi-path convolution fusion module to extract building features from multiple dimensions, and introduce a large receptive field convolution module to break through feature extraction. The process is limited by the size of the receptive field. After fusing the extracted features, the convolutional attention module is used to compress them, and the global features are further mined by pyramid pooling, so as to achieve high-precision extraction of buildings. Results: The current mainstream UNet, pyramid scene parsing network (PSPNet),multi attending path neural network (MAPNet) and multiscale-feature fusion deep neural networks with dilated convolution (MDNNet) are used as the comparison methods, and we use Wuhan University Aerial Image Dataset, Satellite Dataset II (East Asia) and Inria Aerial Image Dataset as experimental data for testing. Compared with the other four methods, MFFNet improves intersection over union, precision, recall, F1-score and mean average precision by 1.53%, 2.65%, 2.41%, 3.32% and 1.19% on average, achieves a better effect. Conclusions: MFFNet not only accurately captures the detail features of buildings, but also strengthens the extraction and utilization of global features. It has better ex‐ traction effect on large buildings and buildings in complex environment. [ABSTRACT FROM AUTHOR]

Land use changes are analyzed correctly, a series of improvements according to the changes are carried out appropriately, the relationship between land use development and economic and human survival is handled correctly, and the healthy and orderly development of the entire society is promoted. Aiming at the combination of multisource remote sensing data and monitoring changes in land planning, this study uses CBERS data and ASAR data as multisource remote sensing data sources to conduct in-depth research and discussion on the land use change in this area and uses the HPF pixel-level fusion method for data fusion to generate HPF. The data are integrated, and then, the CBERS data and HPF fusion data are used to extract the land use type information of Zhenning County, respectively, and a confusion matrix is built based on the field sample points to verify the accuracy, compare and analyze the relative error of the land use type information extraction before and after data fusion, and evaluate the CBERS data. Regarding the extraction effect of land use type information of fusion data with HPF, the results show that the two kinds of remote sensing data have good effects in extracting water body type information, and the accuracy has reached 100%. Using multisource remote sensing image processing can well summarize and analyze land use changes and make the changes in various indicators in the study area. Accurate statistics is obtained. [ABSTRACT FROM AUTHOR]

Adaptive management of large herbivores requires an understanding of how spatial‐temporal fluctuations in forage biomass and quality influence animal performance. Advances in remote sensing have yielded information about the spatial‐temporal dynamics of forage biomass, which in turn have informed rangeland management decisions such as stocking rate and paddock selection for free‐ranging cattle. However, less is known about the spatial‐temporal patterns of diet quality and their influence on large herbivore performance. This is due to infrequent concurrent ground observations of forage conditions with performance (e.g., mass gain), and previously limited satellite data at fine spatial and temporal scales. We combined multi‐temporal field observations of diet quality (weekly) and mass gain (monthly) with satellite‐derived phenological metrics (pseudo‐daily, using data fusion and interpolation) to model daily mass gains of free‐ranging yearling cattle in shortgrass steppe. We used this model to predict grazing season (mid‐May to October) mass gains, a key management indicator, across 40 different paddocks grazed over a 10‐year period (n = 138). We found strong relationships between diet quality and the satellite‐derived phenological metrics, especially metrics related to the timing and rate of green‐up and senescence. Satellite‐derived diet quality estimates were strong predictors of monthly mass gains (R2 = 0.68) across a wide range of aboveground net herbaceous production. Season‐long predictions of average daily gain and cattle off‐mass had mean absolute errors of 8.9% and 2.9%, respectively. The model performed better temporally (across repeated observations in the same paddock) than spatially (across all paddocks within a given year), highlighting the need for accurate vegetation maps and robust field data collection across both space and time. This study demonstrates that free‐ranging cattle performance in rangelands is strongly affected by diet quality, which is related to the timing of vegetation green‐up and senescence. Senescing vegetation suppressed mass gains, even if adequate forage was available. The satellite‐based pseudo‐daily approach presented here offers new opportunities for adaptive management of large herbivores, such as identifying within‐season triggers to move livestock among paddocks, predicting wildlife herd health, or timing the grazing season to better match earlier spring green‐up caused by climate change and plant species invasion. [ABSTRACT FROM AUTHOR]

Multi-source incomplete mixed data abound in real life, like medical data, biological data, remote sensing data, military data, etc. However, some of these sources are of less importance than others, and some are essentially worthless. Therefore, their information fusion and attribute reduction face many challenges. This paper studies information fusion for multi-source incomplete mixed data via conditional information entropy (CIE) and considers its application to attribute reduction based on D-S evidence theory. First of all, a new distance function is defined to measure the difference between nominal attribute values with missing information, and the neighborhood rough set model is used to establish the granularity structure of multi-source incomplete mixed data. Then, a source selection method is given via CIE, which is used to fuse multi-source incomplete mixed data into single-source incomplete mixed data. Based on the maximization of CIE, this method allows worthy and reliable information sources to be chosen. Next, the connection between neighborhood rough set model and D-S evidence theory is established. Moreover, two attribute reduction algorithms for the fused incomplete mixed data are proposed based on the belief and plausibility. Finally, experiments are done to verify the effectiveness of the proposed fusion and reduction algorithms. The results of experiment and statistical test on 12 datasets show that the proposed algorithms exceed other advanced algorithms in classification performance. • The neighborhood rough set model is used to establish the granularity structure. • A source selection method is given via conditional information entropy. • The connection between the neighborhood rough set model and D-S evidence theory is established. • Two attribute reduction algorithms for the fused incomplete mixed data are proposed. • Experiments are done to verify the effectiveness of the proposed fusion and reduction algorithms. [ABSTRACT FROM AUTHOR]

Image fusion is a challenging research area which is useful in various image processing applications. Image fusion integrates information from multiple source images into a single composite image for better visual quality and information content than any of its source images. In the present paper, we have proposed a new median based image fusion algorithm using nonsubsampled shearlet transform. Nonsubsampled shearlet transform is a powerful multiscale geometrical analysis (MGA) tool having rich mathematical structure, high directionality, anisotropy and shift-invariance features. Due to these features nonsubsampled shearlet transform can efficiently capture information of the source images in its coefficient sets. The coefficient sets of the source images are fused by using a new median based fusion rule. Median is an important statistical measurement, which is enriched with two outstanding properties that are edge preserving and robustness against noise. Hence, median based fusion rule increases the quality of fused image. The proposed fusion rule is simple and easy to understand. Strength of the proposed fusion method is verified visually as well as quantitatively by comparing it with different state of the art methods. We have performed experiments on three different types of images (medical, remote sensing and multifocus). Results of the experiments confirm that the proposed method outperform in comparison with other state-of-the-art fusion methods visually as well as quantitatively in terms of different quantitative performance measures such as entropy, standard deviation, edge strength, fusion factor, and running time. [ABSTRACT FROM AUTHOR]

With the development of remote sensing techniques, the fusion of multimodal data, particularly hyperspectral-Light Detection And Ranging (HS-LiDAR) and hyperspectral-SAR, has become an important research field in numerous application areas. Multispectral, HS, LiDAR, and Synthetic Aperture Radar (SAR) images contain detailed information about the monitored surface that are complementary to each other. Thus, data fusion methods have become a promising solution to obtain high spatial resolution remote-sensing images. The main point of this review paper is to classify hyperspectral-LiDAR and hyperspectral-SAR data fusion with approaches. Moreover, recent achievements in the fusion of hyperspectral-LiDAR and hyperspectral-SAR data are highlighted in terms of faced challenges and applications. Most frequently used data fusion datasets that include IEEE GRSS Data Fusion Contests are also described. [ABSTRACT FROM AUTHOR]

Image fusion combines the images of different spectral, spatial, multi-date, as well as radiometric data to achieve a better quality image for improved classification results. Recently, Convolution Neural Network (CNN)-based classification algorithms are extensively used for remote sensing applications. Keeping this in view, present work proposes to use CNN-based fusion and classification of Sentinel 1 (VV and VH polarization) and Sentinel 2 datasets acquired over an agricultural area near Hisar (India). For image fusion, three CNN-based approaches are used to fuse Sentinel 2 (10 m) data with VV and VH bands of Sentinel 1 data. After fusion, classification was performed using 2D-CNN classifier to judge the performance of fused images in terms of classification accuracy. Results suggest that out of the three fusion approaches, only infrared image fusion (IVF) approach performed well with the considered dataset in terms of fusion indicators and classification accuracy. Keeping in view of its better performance, this study proposes a modified IVF approach by using different image pyramid methods. Comparison of results suggests an improved performance by modified IVF approach for the fusion of Sentinel 2 and Sentinel 1 data in comparison with the original IVF approach. [ABSTRACT FROM AUTHOR]

Cropping intensity (CI) is essentially related to the utilization condition of arable land. A classification method based on high spatial resolution data (TM), and a peak-counting method based on high temporal resolution NDVI data (MODIS NDVI) are mainly used to extract CI in remotely sensed field. However, it is difficult to acquire the high spatial resolution data with high quality characteristics in the most classification method, as three requirements cannot be satisfied concurrently. The first condition is the time requirement, where the high spatial resolution data can be observed in the growing season. The second one is the frequency requirements, where at least one period of data can be obtained in the growing season. The third one is the quality requirement, where the data cannot be polluted by the cloud, mist and other environmental surroundings. Furthermore, the performance was relatively low, when the peak-counting method was applied in some complex surface regions, since the high temporal resolution data has the low spatial resolution (>250 m). A feasible strategy can be made to obtain a high extraction accuracy of CI, in order to improve the spatial resolution of high temporal resolution data in the use of spatiotemporal data fusion (Spatial and Temporal Adaptive Reflectance Fusion Model, STARFM). Taking the hilly region in the eastern Sichuan as a research area, the 2010-2011 time series Landsat-like NDVI dataset with high temporal and spatial resolution was collected to extract the spatial distribution of cropping area for the winter-season in2010 and the CI for summer- and winter-seasons in 2011 in Yanting County, using the STARFM algorithm, Landsat TM and MODIS data. Precision validation between different data in various methods was done at a series of spatial scales (30-4 000 m), referring to the visual interpretation of 1509 random samples, and the spatial distribution of cropping area for winter-season in2010 using Landsat data. The results showed that: 1) Compared to the peak-counting method using Landsat-like time-series NDVI dataset, the overall accuracy of verification was higher for the classification method using Landsat data (89.73% vs. 54.94%), at the scale of 30 m. 2) Compared to the peak-counting method using MODIS time-series NDVI dataset, the overall accuracy of verification was higher 3%-7% for the peak-counting method using Landsat-like time-series NDVI dataset, at the scales from 250 m to 4 000 m. 3) The overall accuracy of verification was higher 0.17, when the data from the new method was spatially aggregated from 30 m to 4 000 m. In the CI result of 2011 in Yanting County, the data in the Mianyang Statistical Yearbook of 2011, and 73 samples from the field survey data in August 2013 were used to validate the practicability of new method. There were very close CI values from the statistical data (1.69) and new method (1.67), indicating that the overall accuracy of verification was 73.97% in the field survey. A high accuracy of extraction and fine spatial details of CI can be acquired, when using the fusion of spatiotemporal data in some complex surface regions. The findings can contribute to the evaluation of food production and carbon sequestration potential in complex surface regions. [ABSTRACT FROM AUTHOR]

Spatiotemporal data fusion is a methodology to generate images with both high spatial and temporal resolution. Most spatiotemporal data fusion methods generate the fused image at a prediction date based on pairs of input images from other dates. The performance of spatiotemporal data fusion is greatly affected by the selection of the input image pair. There are two criteria for selecting the input image pair: the “similarity” criterion, in which the image at the base date should be as similar as possible to that at the prediction date, and the “consistency” criterion, in which the coarse and fine images at the base date should be consistent in terms of their radiometric characteristics and imaging geometry. Unfortunately, the “consistency” criterion has not been quantitatively considered by previous selection strategies. We thus develop a novel method (called “cross-fusion”) to address the issue of the determination of the base image pair. The new method first chooses several candidate input image pairs according to the “similarity” criterion and then takes the “consistency” criterion into account by employing all of the candidate input image pairs to implement spatiotemporal data fusion between them. We applied the new method to MODIS–Landsat Normalized Difference Vegetation Index (NDVI) data fusion. The results show that the cross-fusion method performs better than four other selection strategies, with lower average absolute difference (AAD) values and higher correlation coefficients in various vegetated regions including a deciduous forest in Northeast China, an evergreen forest in South China, cropland in North China Plain, and grassland in the Tibetan Plateau. We simulated scenarios for the inconsistency between MODIS and Landsat data and found that the simulated inconsistency is successfully quantified by the new method. In addition, the cross-fusion method is less affected by cloud omission errors. The fused NDVI time-series data generated by the new method tracked various vegetation growth trajectories better than previous selection strategies. We expect that the cross-fusion method can advance practical applications of spatiotemporal data fusion technology. [ABSTRACT FROM AUTHOR]

Combining a hyperspectral (HS) image and a multispectral (MS) image—an example of image fusion—can result in a spatially and spectrally high-resolution image. Despite the plethora of fusion algorithms in remote sensing, a necessary prerequisite, namely registration, is mostly ignored. This limits their application to well-registered images from the same source. In this article, we propose and validate an integrated registration and fusion approach (code available at https://github.com/zhouyuanzxcv/Hyperspectral). The registration algorithm minimizes a least-squares (LSQ) objective function with the point spread function (PSF) incorporated together with a nonrigid freeform transformation applied to the HS image and a rigid transformation applied to the MS image. It can handle images with significant scale differences and spatial distortion. The fusion algorithm takes the full high-resolution HS image as an unknown in the objective function. Assuming that the pixels lie on a low-dimensional manifold invariant to local linear transformations from spectral degradation, the fusion optimization problem leads to a closed-form solution. The method was validated on the Pavia University, Salton Sea, and the Mississippi Gulfport datasets. When the proposed registration algorithm is compared to its rigid variant and two mutual information-based methods, it has the best accuracy for both the nonrigid simulated dataset and the real dataset, with an average error less than 0.15 pixels for nonrigid distortion of maximum 1 HS pixel. When the fusion algorithm is compared with current state-of-the-art algorithms, it has the best performance on images with registration errors as well as on simulations that do not consider registration effects. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Image & Graphics is the property of Editorial Office of Journal of Image & Graphics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

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]

Multi-modal data fusion has recently been shown promise in classification tasks in remote sensing. Optical data and radar data, two important yet intrinsically different data sources, are attracting more and more attention for potential data fusion. It is already widely known that a machine learning-based methodology often yields excellent performance. However, the methodology relies on a large training set, which is very expensive to achieve in remote sensing. The semi-supervised manifold alignment (SSMA), a multi-modal data fusion algorithm, has been designed to amplify the impact of an existing training set by linking labeled data to unlabeled data via unsupervised techniques. In this paper, we explore the potential of SSMA in fusing optical data and polarimetric synthetic aperture radar (SAR) data, which are multi-sensory data sources. Furthermore, we propose a MAPPER-induced manifold alignment (MIMA) for the semi-supervised fusion of multi-sensory data sources. Our proposed method unites SSMA with MAPPER, which is developed from the emerging topological data analysis (TDA) field. To the best of our knowledge, this is the first time that SSMA has been applied on fusing optical data and SAR data, and also the first time that TDA has been applied in remote sensing. The conventional SSMA derives a topological structure using $k$ -nearest neighbor (kNN), while MIMA employs MAPPER, which considers the field knowledge and derives a novel topological structure through the spectral clustering in a data-driven fashion. The experimental results on data fusion with respect to land cover land use classification and local climate zone classification suggest superior performance of MIMA. [ABSTRACT FROM AUTHOR]

Road recognition and extraction based on remotely sensed data is efficient and applicable in much urban management studies. In this research, the capabilities of SPOT and SAR images are investigated for road recognition. Spectral and textural similarities between roads and other urban objects such as building's roofs many cause some difficulties in road recognition based on SPOT image. On the other hand, SAR images are good for small road recognition but, may have some difficulties for detecting roads among vegetation. The proposed method in this paper is a decision level fusion of SPOT and SAR classification results in order to modify extracted road regions. This method has three main steps; 1) texture feature extraction from each of the SPOT and SAR images, 2) classifying each of the SPOT and SAR images based on SVM classifier, 3) decision level fusion of classification results in order to reduce road recognition difficulties and having optimum road regions. Performing the capabilities of the proposed decision level fusion algorithm for road recognition can improve the quality of the classification for about 21%. [ABSTRACT FROM AUTHOR]

Land-cover classification of Remote Sensing (RS) data in urban area has always been a challenging task due to the complicated relations between different objects. Recently, fusion of aerial imagery and light detection and ranging (LiDAR) data has obtained a great attention in RS communities. Meanwhile, convolutional neural network (CNN) has proven its power in extracting high-level (deep) descriptors to improve RS data classification. In this paper, a CNN-based feature-level framework is proposed to integrate LiDAR data and aerial imagery for object classification in urban area. In our method, after generating low-level descriptors and fusing them in a feature-level fusion by layer-stacking, the proposed framework employs a novel CNN to extract the spectral-spatial features for classification process, which is performed using a fully connected multilayer perceptron network (MLP). The experimental results revealed that the proposed deep fusion model provides about 10% improvement in overall accuracy (OA) in comparison with other conventional feature-level fusion techniques. [ABSTRACT FROM AUTHOR]

Accurate and timely winter wheat yield estimation has significant effect to grain markets and policy. Most crop estimation methods can be divided into two categories, one is based on the crop model and the other is the statistical learning method. For statistical learning methods with recent advances in deep learning, convolutional neural network (CNN) have become state-of-the-art algorithms. can extract the depth-dependent features of crop growth. However, the pivotal challenge is to combine remote sensing images with CNN. In this paper, we employ the method of histogram dimensionality reduction and time series fusion to generate the input layer. The experiment firstly performed projection transformation, splicing, mask, fusion, and clipping for 6 different MODIS images in the research area from 2006 to 2016, and then generated 21 600 fusion images of 12 bands (surface reflectance data of 7 different wavelengths in MOD09A1, surface temperature of day and night in MYD11A2, NDVI and EVI in MOD13A1, and FPAR in MOD15A2H). Then, the sensitivity range of winter wheat growth in each band is divided into 36 sections, and the histogram statistics are used to reduce the dimension to generate a vector of length 36, so the remote sensing image generates a matrix of 36×36×12 in the 228-day growing season. The corresponding time and regional statistics are applied as the output layer to construct a complete sample. The yield estimation sample database of 12 indices in the winter wheat region of north China (60 prefecture-level cities) from 2006 to 2016 was constructed, and the training set and verification set were divided into 10:1 for the training and evaluation of yield estimation model. Finally, the neural network structure is designed according to the sample, which consists of the input layer, 7 convolution layers (c1-c7), 7 activation layers, 7 batch normalization layers, 3 dropout layers, 2 full connection layers, and output layer. The number of c1-c7 convolution kernels is 64, 64, 128, 128, 256, 256, 256, the convolution kernel size is 3×3 dpi, and the sliding step length is 2, 1, 2, 2, 2, 1 and 2 respectively, 1 zero paddings per convolutional layer. At the same time, batch normalization and Relu function activation are performed on each convolutional layer, and the Dropout layer is used in the fully connected layer. The results show that: 1) The root-mean-square error (RMSE) and coefficient of determination (R²) of the convolutional neural network model on the training set are 183.82 kg/hm² and 0.98 respectively. In the validation set, RMSE and R² are 689.72 kg/hm² and 0.71. 2) With the same neural network structure, the average RMSE of the estimated samples from 2006 to 2016 trained as validation sets for 11 models was 772.03 kg/hm². The error of the yield estimation model was the largest in 2007 and the smallest in 2012, and the RMSE was 920.45 kg/hm² and 632.08 kg/hm² respectively. Crop estimation algorithm based on CNN has high robustness and precision; 3) The accuracy analysis of prediction yield at the municipal level of different provinces in three temporal points of 2007, 2012 and 2016 indicates that the model has higher accuracy in most areas of the northern winter wheat region, especially, RMSE of Hebei and Shandong provinces is approximately 500 kg/hm². The result shows that CNN is well applied to the estimation of winter wheat production. This is a great thought of remote sensing combined with the deep learning algorithm. This method can be used to estimate yield by remote sensing in different scales and regions. Compared with the traditional method, this “start-to-end” learning method has the advantage of synergy and can obtain the optimal estimation model relative to the whole area. Meanwhile, As data accumulates, the estimation accuracy will be continuously improved, and it has a good application prospect in the national agricultural production forecast. [ABSTRACT FROM AUTHOR]

The multifractal fusion method has proved to be an effective algorithm to mitigate the noise of the sea surface salinity (SSS) of Soil Moisture Ocean Salinity (SMOS) mission. However, the traditional nonparametric weight function used in this method is unable to fully capture the dynamic evolution of the oceanic environment. Considering the multiscale, nonuniform, anisotropic, and flow-dependent nature of the ocean, a prototype with the so-called flexible circle (FLC) weight function or flexible ellipse (FLE) weight function with a set of predefined parameters is proposed in this paper. The improved weight functions could draw dynamic information from the sea surface temperature, Rossby radius of deformation, and surface geostrophic flow to improve the quality of the remotely sensed SSS. The validation against the in situ data indicates that the improved weight functions perform better than the traditional one with a reduced root-mean-square (RMS) and standard deviation (STD) of the differences with respect to EN 4.2.0 profiles (from 0.50 and 0.46 to 0.42 and 0.38 for FLC and 0.39 and 0.36 for FLE in the global ocean). In particular, the FLE scheme could highlight the variation of the strong currents without affecting the computational efficiency. Furthermore, this paper discusses the influences of the error distribution on the fusion results and underlines the importance of error-based adaptions for further improvements. [ABSTRACT FROM AUTHOR]

This paper aims to enhance the accuracy and reduce the cost of the fusion of multi-source remote sensing data. For this purpose, the existing multi-source remote sensing data fusion methods were reviewed in detail. Then, a new back propagation (BP) neural network (BPNN) fusion algorithm for the groundwater was put forward based on hybrid soft computing. Using the function approximation ability of BP neural network, it was combined with the Kalman filter to form an optimization method. The BP neural network was coupled with the particle swarm optimization (PSO) algorithm into the PSO-BPNN-EKF data fusion algorithm. On this basis, the least squares support vector machine (LSSVM) was introduced to create the LSSVM-PSO data fusion algorithm. Through simulation experiments, it is learned that the proposed algorithm can effectively fuse the multi-source remote sensing data on groundwater, especially in the case of big data. The research findings shed a new light on the fusion of remote sensing data collected by multiple sensors. [ABSTRACT FROM AUTHOR]

After an earthquake, efficiently and accurately acquiring information about damaged buildings can help reduce casualties. Earth observation data have been widely used to map affected areas after earthquakes. However, accurate post-earthquake assessment results are needed to manage recovery and reconstruction and estimate economic losses. In this paper, for quantification and precision purposes, information on earthquake-induced building damage is extracted using multi-source remote sensing images collected after an earthquake. The multi-source remote sensing data include optical data, synthetic aperture radar (SAR) data, and digital surface model (DSM) data generated by interpolating light detection and ranging (LiDAR) point cloud data. Features that describe texture, colour, and geometry are included in our analysis. The feature analysis is carried out according to the rough set theory to further determine the feature parameters. A logistic regression model (LRM) is employed to find the optimal fitting function to describe the relationship between the occurrence and absence of destroyed buildings within an individual object. In our experiment, old Beichuan County, China, the area most devastated by the Wenchuan earthquake on May 12, 2008, is used to test the proposed hypothesis. Through comparison with a ground survey, the experimental results show that the detection accuracy of the proposed method is 94.2 %; the area under the receiver operating characteristic (ROC) curve is 0.827. The efficiency of the proposed method is demonstrated using 6 modes of data combination acquired from the same area in old Beichuan County. The approach is one of the first attempts to extract damaged buildings through the fusion of three types of data with different features. The approach addresses multivariate regression methodologies and compares the potentials of different features for application in the field of damage detection. [ABSTRACT FROM AUTHOR]

The objectives of this study were to accurately model relationships between spectral reflectance and water-quality parameters, including blue-green algae phycocyanin, chlorophyll a, total suspended solids, turbidity, and total dissolved solids; evaluate feature-level fusion to spectral data for water-quality modeling; and evaluate the effectiveness of machine learning regression techniques and decision-level fusion for water-quality variable prediction. We introduce the application of canonical correlation analysis fusion as a method for water-based spectral analysis to overcome the low signal-to-noise ratio of the data. Water-quality variables and spectral reflectance were used to create predictive models via machine learning regression models, including multiple linear regression, partial least-squares regression, Gaussian process regression, support vector machine regression, and extreme learning machine regression. The models were then combined using decision-level fusion. Results indicate that canonical correlation analysis feature-level fusion and machine learning techniques are superior to traditional methods. [ABSTRACT FROM AUTHOR]

Tree trunks detection and their location information are needed to perform effective production and management in forestry and fruit farming. A novel algorithm based on data fusion with a vision camera and a 2D laser scanner was developed to detect tree trunks accurately. The transformation was built from a laser coordinate system to an image coordinate system, and the model of a rectangle calibration plate with two inward concave regions was established to implement data alignment between two sensors data. Then, data fusion and decision with Dempster-Shafer theory were achieved through integration of decision level after designing and determining basic probability assignments of regions of interesting (RoIs) for laser and vision data respectively. Tree trunk width was calculated by using laser data to determine basic probability assignments of RoIs of laser data. And a stripping segmentation algorithm was presented to determine basic probability assignments of RoIs of vision data, by calculating the matching level of RoIs like tree trunks. A robot platform was used to acquire data from sensors and to perform the developed tree trunk detection algorithm. Combined calibration tests were conducted to calculate a conversion matrix transforming from the laser coordinate system to the image coordinate system, and then field experiments were carried out in a real pear orchard under sunny and cloudy conditions, with trunk width measurement of 120 trees and 40 images processed by the presented stripping segmentation algorithm. Results showed the algorithm was successful to detect tree trunks and data fusion improved the ability for tree trunk detection. This algorithm could provide a new method for tree trunk detection and accurate production and management in orchards. [ABSTRACT FROM AUTHOR]

Dense time-series high-resolution satellite images are extremely valuable for long-term monitoring of land dynamics. Spatiotemporal fusion (STF) techniques have been developed to integrate multi-resolution satellite images to produce data with high spatial resolution and temporal frequency. Due to the large volume and diversity of higher resolution global Earth Observation (EO) data, large-scale data processing methods need to be computationally efficient, free of parameter fine-tuning, and adaptive to various data structures without process customization. These requirements are especially critical for the production of global Seamless Data Cube (SDC). Considering the limitations of existing STF methods, we propose a ROBust OpTimization-based (ROBOT) fusion model that exploits time-series information to obtain more accurate predictions. ROBOT maintains stable under varied data conditions by adopting a temporal-coherence regularization term. And being free of parameter tuning, ROBOT can be applied to arbitrary spatiotemporally distributed data without repetitive process customization. Its performance was compared with eight representative STF methods. Results show that ROBOT outperforms existing STF methods in most cases and is computationally efficient, about 4000-fold faster than ESTARFM and 600-fold faster than FSDAF. The proposed method has demonstrated its potential for global-scale SDC generation to support subsequent remote sensing applications. • ROBOT is proposed to employ time-series data for better fusion results. • It performs stably under varied data conditions and is free of parameter tuning. • It does not require input images to be completely clean and cloud-free. • It outperforms existing algorithms in both accuracy and efficiency. • It makes the on-the-run spatiotemporal data fusion possible and thus saves storage. [ABSTRACT FROM AUTHOR]

The fraction of absorbed photosynthetically active radiation (FAPAR) is a critical input in many climate and ecological models. The accuracy of satellite FAPAR products directly influences estimates of ecosystem productivity and carbon stocks. The targeted accuracy of FAPAR products is 10% or 0.05 for many applications. However, most current FAPAR products do not meet such requirements, and further improvements are still needed. In this paper, a data fusion scheme based on the multiple resolution tree (MRT) approach is developed to integrate multiple satellite FAPAR estimates at site and regional scales. MRT was chosen because of the superior computational efficiency compared with other fusion methods. The fusion scheme removed the bias in FAPAR estimates and resulted in a 15% increase in the $R^{2}$ and 3% reduction in the root-mean-square error compared with the average of individual FAPAR estimates. The regional-scale fusion filled in the missing values, and provided spatially consistent FAPAR distributions at different resolutions. Overall, MRT can be used to efficiently and accurately generate spatially and temporally continuous FAPAR data across both site and regional scales. [ABSTRACT FROM AUTHOR]

Summary: Monitoring networks contain monitoring nodes that observe an area of interest to detect any possible existing object and estimate its states. Each node has characteristics such as probability of detection and clutter density that may have different values for distinct nodes in nonhomogeneous monitoring networks. This paper proposes a modified covariance intersection method for data fusion in such networks. It is derived by formulating a mixed game model between neighbor monitoring nodes as players and considering the inverse of the trace of fused covariance matrix as players' utility function. Monitoring nodes estimate the states of any possible existing object by applying joint target detection and tracking filter on their own observations. Processing nodes fuse the estimated states received from neighbor monitoring nodes by the proposed modified covariance intersection. It is validated by simulating target detection and tracking problem in 2 situations: 1 target and unknown number of targets. [ABSTRACT FROM AUTHOR]

This paper presents a sensor-driven hierarchical domain adaptation method that aims at transferring the knowledge from a source domain (RS image where reference data are available) to a different but related target domain (RS image where no labeled reference data are available) for solving a classification problem. Due to the different acquisition conditions, a difference in the source and target distributions of the features representing the same class is generally expected. To solve this problem, the proposed method takes advantage from the availability of multisensor data to hierarchically detect features subspaces where for some classes data manifolds are partially (or completely) aligned. These feature subspaces are associated with invariant physical properties of classes measured by the sensors in the scene, i.e., measures having almost the same behavior in both domains. The detection of these invariant feature subspaces allows us to infer labels of the target samples that result more aligned to the source data for the considered subset of classes. Then, the labeled target samples are analyzed in the full feature space to classify the remaining target samples of the same classes. Finally, for those classes for which none of the sensors can measure invariant features, we perform the adaptation via a standard active learning technique. Experimental results obtained on two real multisensor data sets confirm the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Correlation analysis has been widely used to validate the accuracy of synthetic medium-resolution data generated by spatial and temporal fusion methods. However, as the temporal resolution of Landsat data is 16 days, this method only can be used for the validation of multi-temporal Landsat data. This means that the fusion accuracy of most images in a synthetic daily time series have not been validated. Furthermore, the fusion accuracy of each image in a synthetic Landsat time series is different, because there is a negative correlation with the time interval length between the fusion date and the base image date. Therefore, there is a need for temporal validation of synthetic daily Landsat time series data. We propose a suitable validation method in this paper. The improved spatial and temporal data fusion approach (ISTDFA) was applied to generate synthetic daily Landsat Normalized Difference Vegetation Index (NDVI) time series, which were then validated for both spatial and temporal dimensions using actual MODIS NDVI time series. For temporal validation, the correlation coefficients ( R ) between the actual and synthetic 500 m NDVI time series were calculated by pixel-by-pixel to generate imagery with an R value for each pixel. For spatial validation, R between MODIS NDVI imagery and synthetic Landsat NDVI imagery was calculated day by day to generate an R time series. This method was tested and validated in two locations (Bole and Luntai) in Xinjiang Province, China. The results show that, for temporal validation, the R values of 86.08% pixels in Bole and 94.71% pixels in Luntai are higher than 0.9, and in spatial validation, R values are higher than 0.8 on most days. Synthetic daily Landsat NDVI data was used to monitor the phenology of vegetation at a spatial resolution of 30 m successfully, while the MODIS product is limited to 500 m. [ABSTRACT FROM AUTHOR]

As a list of remotely sensed data sources is available, how to efficiently exploit useful information from multisource data for better Earth observation becomes an interesting but challenging problem. In this paper, the classification fusion of hyperspectral imagery (HSI) and data from other multiple sensors, such as light detection and ranging (LiDAR) data, is investigated with the state-of-the-art deep learning, named the two-branch convolution neural network (CNN). More specific, a two-tunnel CNN framework is first developed to extract spectral-spatial features from HSI; besides, the CNN with cascade block is designed for feature extraction from LiDAR or highresolution visual image. In the feature fusion stage, the spatial and spectral features of HSI are first integrated in a dual-tunnel branch, and then combined with other data features extracted from a cascade network. Experimental results based on several multisource data demonstrate the proposed two-branch CNN that can achieve more excellent classification performance than some existing methods. [ABSTRACT FROM AUTHOR]

Pansharpening usually refers to the fusion of a high spatial resolution panchromatic (PAN) image with a higher spectral resolution but coarser spatial resolution multispectral (MS) image. Owing to the wide applicability of related products, the literature has been populated by many papers proposing several approaches and studies about this issue. Many solutions require a preliminary spectral matching phase wherein the PAN image is matched with the MS bands. In this paper, we propose and properly justify a new approach for performing this step, demonstrating that it yields state-of-the-art performance. The comparison with existing spectral matching procedures is performed by employing four data sets, concerning different kinds of landscapes, acquired by the Pléiades, WorldView-2, and GeoEye-1 sensors. [ABSTRACT FROM AUTHOR]

Snow depth is a general input variable in many models of agriculture, hydrology, climate, and ecology. However, there are some uncertainties in the retrieval of snow depth by remote sensing. Errors occurred in snow depth evaluation under the D-InSAR methods will affect the accuracy of snow depth inversion to a certain extent. This study proposes a scheme to estimate spatial snow depth that combines remote sensing with site observation. On the one hand, this scheme adopts the Sentinel-1 C-band of the European Space Agency (ESA), making use of the two-pass method of differential interferometry for inversion of spatial snow depth. On the other hand, the 3DVAR (three dimensional variational) fusion algorithm is used to integrate actual snow depth data of virtual stations and real-world observation stations into the snow depth inversion results. Thus, the accuracy of snow inversion will be improved. This scheme is applied in the study area of Bayanbulak Basin, which is located in the central hinterland of Tianshan Mountains in Xinjiang, China. Observation data from stations in different altitudes are selected to test the fusion method. According to the results, most of the obtained snow depth values using interferometry are lower than the observed ones. However, after the fusion using the 3DVAR algorithm, the snow depth accuracy is slightly higher than it was in the inversion results (R² = 0.31 vs. R² = 0.50, RMSE = 2.51 cm vs. RMSE = 1.96 cm; R² = 0.27 vs. R² = 0.46, RMSE = 4.04 cm vs. RMSE = 3.65 cm). When compared with the inversion results, the relative error (RE) improved by 6.97% and 3.59%, respectively. This study shows that the scheme can effectively improve the accuracy of regional snow depth estimation. Therefore, its future application is of great potential. [ABSTRACT FROM AUTHOR]

Time series vegetation indices with high spatial resolution and high temporal frequency are important for crop growth monitoring and management. However, due to technical constraints and cloud contamination, it is difficult to obtain such datasets. In this study, a spatio-temporal vegetation index image fusion model (STVIFM) was developed to generate high spatial resolution Normalized Difference Vegetation Index (NDVI) time-series images with higher accuracy, since most of the existing methods have some limitations in accurately predicting NDVI in heterogeneous regions, or rely on very computationally intensive steps and land cover maps for heterogeneous regions. The STVIFM aims to predict the fine-resolution NDVI through understanding the contribution of each fine-resolution pixel to the total NDVI change, which was calculated from the coarse-resolution images acquired on two dates. On the one hand, it considers the difference in relationships between the fine- and coarse-resolution images on different dates and the difference in NDVI change rates at different growing stages. On the other hand, it neither needs to search similar pixels nor needs to use land cover maps. The Landsat-8 and MODIS data acquired over three test sites with different landscapes were used to test the spatial and temporal performance of the proposed model. Compared with the spatial and temporal adaptive reflectance fusion model (STARFM), enhanced spatial and temporal adaptive reflectance fusion model (ESTARFM) and the flexible spatiotemporal data fusion (FSDAF) method, the proposed STVIFM outperforms the STARFM and ESTARFM at three study sites and different stages when the land cover or NDVI changes were captured by the two pairs of fine- and coarse-resolution images, and it is more robust and less computationally intensive than the FSDAF. [ABSTRACT FROM AUTHOR]

G reen tides, a worldwide problem, are harmful to aquaculture, tourism, marine ecosystems, and maritime traffic. Currently, green tide detection relies on remote sensing (RS) images, which are often missing or unusable. Thus, the observation and detection of green tides cannot be performed daily, which makes it difficult to improve environmental quality and ecological health. To address this problem, this study proposed a novel green tide estimation framework (GTEF) through convolutional long short-term memory, which learned the historical spatial-temporal seasonal and trend patterns of green tides from 2008 to 2021 and fused the previously observed or estimated data and biological (optional) and physical (optional) data over the preceding seven days when RS images were absent or unusable for daily observation and detection tasks. The results showed that the overall accuracy (OA), false-alarm rating (FAR), and missing-alarm rating (MAR) of the GTEF were 0.9592 ± 0.0375, 0.0885 ± 0.1877 and 0.4315 ± 0.2848, respectively. The estimated results described the green tides in terms of attributes, geometry and position features. Especially in the latitudinal features, the Pearson correlation coefficient of the predicted data and observed data were over 0.8 (P < 0.05), which showed a strong correlation. In addition, this study also discussed the role of biological and physical factors in the GTEF. Sea surface salinity may be the dominant factor in the early stages of green tides; in the late stage, solar irradiance may be the dominant factor. Sea surface winds and sea surface currents also played a significant role in green tide estimation. Results showed the OA, FAR and MAR of the GTEF which, with physical factors but without biological factors, were 0.9556 ± 0.0389, 0.1311 ± 0.3338 and 0.4297 ± 0.3180, respectively. In short, the proposed approach could generate a daily map of green tides, even if RS images were missing or unusable. [Display omitted] • Proposing a novel framework for green tide estimation in the Yellow Sea. • Fusing the historical green tide data, marine biological factor, marine physical factors. • Analyzing the import feature of the process of green tides in the Yellow Sea. • Aiming to help continuous green tide monitoring in the absence of remote sensing images. [ABSTRACT FROM AUTHOR]

Data fusion is a powerful tool for the merging of multiple sources of information to produce a better output as compared to individual source. This study describes the data fusion of five land use/cover types, that is, bare land, fertile cultivated land, desert rangeland, green pasture, and Sutlej basin river land derived from remote sensing. A novel framework for multispectral and texture feature based data fusion is designed to identify the land use/land cover data types correctly. Multispectral data is obtained using a multispectral radiometer, while digital camera is used for image dataset. It has been observed that each image contained 229 texture features, while 30 optimized texture features data for each image has been obtained by joining together three features selection techniques, that is, Fisher, Probability of Error plus Average Correlation, and Mutual Information. This 30-optimized-texture-feature dataset is merged with five-spectral-feature dataset to build the fused dataset. A comparison is performed among texture, multispectral, and fused dataset using machine vision classifiers. It has been observed that fused dataset outperformed individually both datasets. The overall accuracy acquired using multilayer perceptron for texture data, multispectral data, and fused data was 96.67%, 97.60%, and 99.60%, respectively. [ABSTRACT FROM AUTHOR]

Remote sensing imagery to monitor global biophysical dynamics requires the availability of thermal infrared data at high temporal and spatial resolution because of the rapid development of crops during the growing season and the fragmentation of most agricultural landscapes. Conversely, no single sensor meets these combined requirements. Data fusion approaches offer an alternative to exploit observations from multiple sensors, providing data sets with better properties. A novel spatio-temporal data fusion model based on constrained algorithms denoted as multisensor multiresolution technique (MMT) was developed and applied to generate TIR synthetic image data at both temporal and spatial high resolution. Firstly, an adaptive radiance model is applied based on spectral unmixing analysis of . TIR radiance data at TOA (top of atmosphere) collected by MODIS daily 1-km and Landsat - TIRS 16-day sampled at 30-m resolution are used to generate synthetic daily radiance images at TOA at 30-m spatial resolution. The next step consists of unmixing the 30 m (now lower resolution) images using the information about their pixel land-cover composition from co-registered images at higher spatial resolution. In our case study, TIR synthesized data were unmixed to the Sentinel 2 MSI with 10 m resolution. The constrained unmixing preserves all the available radiometric information of the 30 m images and involves the optimization of the number of landcover classes and the size of the moving window for spatial unmixing. Results are still being evaluated, with particular attention for the quality of the data streams required to apply our approach. [ABSTRACT FROM AUTHOR]

The inconsistency between the freely available remote sensing datasets and crowd-sourced data from the resolution perspective forms a big challenge in the context of data fusion. In classical classification problems, crowd-sourced data are represented as points that may or not be located within the same pixel. This discrepancy can result in having mixed pixels that could be unjustly classified. Moreover, it leads to failure in retaining sufficient level of details from data inferences. In this paper we propose a method that can preserve detailed inferences from remote sensing datasets accompanied with crowd-sourced data. We show that advanced machine learning techniques can be utilized towards this objective. The proposed method relies on two steps, firstly we enhance the spatial resolution of the satellite image using Convolutional Neural Networks and secondly we fuse the crowd-sourced data with the upscaled version of the satellite image. However, the covered scope in this paper is concerning the first step. Results show that CNN can enhance Landsat 8 scenes resolution visually and quantitatively. [ABSTRACT FROM AUTHOR]

Dempster-Shafer evidence theory has been extensively used in many information fusion systems since it was proposed by Dempster and extended by Shafer. Many scholars have been conducted on conflict management of Dempster-Shafer evidence theory in past decades. However, how to determine a potent parameter to measure evidence conflict, when the given environment is in an open world, namely the frame of discernment is incomplete, is still an open issue. In this paper, a new method which combines generalized conflict coefficient, generalized evidence distance, and generalized interval correlation coefficient based on ordered weighted averaging (OWA) operator, to measure the conflict of evidence is presented. Through ordered weighted average of these three parameters, the combinatorial coefficient can still measure the conflict effectively when one or two parameters are not valid. Several numerical examples demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Flood events cause substantial damage to urban and rural areas. Monitoring water extent during large-scale flooding is crucial in order to identify the area affected and to evaluate damage. During such events, spatial assessments of floodwater may be derived from satellite or airborne sensing platforms. Meanwhile, an increasing availability of smartphones is leading to documentation of flood events directly by individuals, with information shared in real-time using social media. Topographic data, which can be used to determine where floodwater can accumulate, are now often available from national mapping or governmental repositories. In this work, we present and evaluate a method for rapidly estimating flood inundation extent based on a model that fuses remote sensing, social media and topographic data sources. Using geotagged photographs sourced from social media, optical remote sensing and high-resolution terrain mapping, we develop a Bayesian statistical model to estimate the probability of flood inundation through weights-of-evidence analysis. Our experiments were conducted using data collected during the 2014 UK flood event and focus on the Oxford city and surrounding areas. Using the proposed technique, predictions of inundation were evaluated against ground-truth flood extent. The results report on the quantitative accuracy of the multisource mapping process, which obtained area under receiver operating curve values of 0.95 and 0.93 for model fitting and testing, respectively. [ABSTRACT FROM AUTHOR]

Hypersharpening aims at combining an observable low-spatial resolution hyperspectral image with a high-spatial resolution remote sensing image, in particular a multispectral one, to generate an unobservable image with the high spectral resolution of the former and the high spatial resolution of the latter. In this paper, two such new fusion methods are proposed. These methods, related to linear spectral unmixing techniques, and based on nonnegative matrix factorization (NMF), optimize a new joint criterion and extend the recently proposed joint NMF (JNMF) method. The first approach, called gradient-based joint-criterion NMF (Grd-JCNMF), is a gradient-based method. The second one, called multiplicative JCNMF (Mult-JCNMF), uses new designed multiplicative update rules. These two JCNMF approaches are applied to synthetic and semireal data, and their effectiveness, in spatial and spectral domains, is evaluated with commonly used performance criteria. Experimental results show that the proposed JCNMF methods yield sharpened hyperspectral data with good spectral and spatial fidelities. The obtained results are compared with the performance of two NMF-based methods and one approach based on a sparse representation. These results show that the proposed methods significantly outperform the well-known coupled NMF sharpening method for most performance figures. Also, the proposed Mult-JCNMF method provides the results that are similar to those obtained by JNMF, with a lower computational cost. Compared with the tested sparse-representation-based approach, the proposed methods give better results. Moreover, the proposed Grd-JCNMF method considerably surpasses all other tested methods. [ABSTRACT FROM AUTHOR]

Although many advances have been made in past decades, land cover classification of fine-resolution remotely sensed (RS) data integrating multiple temporal, angular, and spectral features remains limited, and the contribution of different RS features to land cover classification accuracy remains uncertain. We proposed to improve land cover classification accuracy by integrating multi-source RS features through data fusion. We further investigated the effect of different RS features on classification performance. The results of fusing Landsat-8 Operational Land Imager (OLI) data with Moderate Resolution Imaging Spectroradiometer (MODIS), China Environment 1A series (HJ-1A), and Advanced Spaceborne Thermal Emission and Reflection (ASTER) digital elevation model (DEM) data, showed that the fused data integrating temporal, spectral, angular, and topographic features achieved better land cover classification accuracy than the original RS data. Compared with the topographic feature, the temporal and angular features extracted from the fused data played more important roles in classification performance, especially those temporal features containing abundant vegetation growth information, which markedly increased the overall classification accuracy. In addition, the multispectral and hyperspectral fusion successfully discriminated detailed forest types. Our study provides a straightforward strategy for hierarchical land cover classification by making full use of available RS data. All of these methods and findings could be useful for land cover classification at both regional and global scales. [ABSTRACT FROM AUTHOR]

Human activity recognition has received a lot of attention recently, mainly thanks to the advancements in sensing technologies and systems' increasing computational power. However, complexity in human movements, sensing devices' noise and person-specific characteristics impose challenges that still remain to be overcome. In the proposed work, a novel, multi-modal human action recognition method is presented for handling the aforementioned issues. Each action is represented by a basis vector and spectral analysis is performed on an affinity matrix of new action feature vectors. Using modality-dependent kernel regressors for computing the affinity matrix, complexity is reduced and robust low-dimensional representations are achieved. The proposed scheme supports online adaptivity of modalities, in a dynamic fashion, according to their automatically inferred reliability. Evaluation on three publicly available datasets demonstrates the potential of the approach. [ABSTRACT FROM AUTHOR]