The Normalized Difference Vegetation Index (NDVI) is widely used for vegetation monitoring, phenological analysis, and ecosystem assessment. However, obtaining continuous daily high-resolution NDVI time series remains challenging because high-resolution satellite observations are often affected by long revisit intervals and cloud contamination. Existing reconstruction methods generally rely on dense observations or auxiliary datasets, limiting their applicability in data-sparse regions. In this study, we propose a method by extracting phenological archetypes from tower observations to reconstruct GF NDVI time series. The proposed method successfully reconstructed daily NDVI for 2024 over a forested area. The results show that the reconstructed NDVI achieves an RMSE of 0.100 at 16 m resolution (R2=0.942) and 0.158 at 1.5 m resolution (R2=0.789), compared with tower observations. The method enhances spatial detail by more than five times relative to the original GF NDVI while maintaining temporal consistency, requiring only tower observations and limited satellite data. These results demonstrate the potential of the proposed approach for high-resolution daily NDVI reconstruction in data-sparse regions.

Bridging the point-to-pixel scale mismatch remains a central challenge in soil moisture (SM) product validation when only a limited number of in situ sites are available. Conventional arithmetic averaging gives equal importance to site observations without explicitly accounting for site-specific error magnitudes, intersite error dependence, or uncertainty in the resulting pixel-scale reference. Rather than pursuing higher point-estimate accuracy, this study emphasizes the explicit statistical characterization of observation errors. We propose a Bayesian upscaling framework that explicitly characterizes these error structures through a regularized observation error covariance matrix incorporated into the likelihood function. The contributions of calibrated observations to the posterior estimate are thereby determined jointly by their error variances and intersite covariances, while the framework simultaneously yields a pixel-scale posterior estimate and a quantified measure of its uncertainty. The method was evaluated using 56,950 valid three-site combinations derived from 72 sites in the Qinghai Lake Basin. Results show that the method reduces reference bias for site combinations with low representativeness by approximately 66.3%. Across 24 SM time series from eight products, the deviations in correlation coefficients and error metrics relative to the benchmark reference are reduced by approximately 58%, while propagation of posterior reference uncertainty enabled the uncertainty in the validation metrics to be explicitly quantified. Without adding sites, this framework improves the statistical properties of the upscaling process by reusing the error information inherent in existing observations, thereby providing a practical Bayesian bridge from sparse point measurements to defensible pixel-scale reference with quantifiable confidence under routine, site-limited validation conditions.

Thermal radiation directionality (TRD) characterizes the anisotropic signature of most surface targets in the thermal infrared (TIR) domain. It introduces uncertainties in land surface temperature (LST) retrieval from TIR remote sensing data. Multi-angle quasi-synchronous observation is the most direct and effective approach to addressing the TRD effect. Currently, only the ATSR series sensors offer dual-angle observations, which are insufficient to capture the TRD effect. Designing a wide-swath multi-angle TIR satellite sensor is expected to globally capture the TRD effect and further support the angular normalization of the LST product. For that, this study aimed to determine the optimal angle combination for multi-angular linear array imaging, based on 10,200 groups of 4SAIL model-simulated hemispherical distributed directional brightness temperature (DBT, containing 71 view zenith angle × 360 view azimuth angle for each group) and 220 groups of limited multi-angle combinations (considering the field of view effect and image displacement due to Earth’s curvature). The LSF-Chen kernel-driven model was selected as the fitting tool to simulate hemispherical distributed DBT with the inputs of limited multi-angular DBTs extracted from 4SAIL reference data. Then, the optimal angle combination could be determined according to the Root Mean Square Error (RMSE) and the Average Proportion of Absolute Error within 0.5 K (APAE≤0.5 K). Results indicate that: 1) The 5-angle combination (0°, ±22.1°, ±50.3°) with a typical satellite altitude of 800 km and a 100° field of view (i.e., with a swath width of 2127.6 km) is suggested. 2) The RMSE of the optimal 5-angle combination increases from the array center (0.49 K) to the array edge (1.30 K) when a random error of [-0.5 K, 0.5 K] is considered. In addition, this optimal 5-angle combination was evaluated using MODIS data with 0.5 K random noise, which shows its overall RMSE is 0.58 K. This study provides a new insight for designing a new wide-swath multi-angle TIR satellite sensor for upcoming Earth observation.

Cross-evaluation results of multi-source satellite solar-induced chlorophyll fluorescence (SIF) products depend critically on the analytical choices made during evaluation, including product combinations, statistical models, and spatiotemporal preprocessing. However, this ”conditional dependence” has long lacked systematic quantification, severely limiting the interpretability of evaluation conclusions. This study constructs a diagnostic framework to systematically assess the methodological sensitivity of cross-evaluation. Using four representative SIF products from 2018 to 2021, we isolated the independent effects of product combination, error model (Three-Cornered Hat, TCH, vs. Quadruple Collocation Analysis, QCA), spatial resolution, and seasonal-cycle treatment through controlled experiments. Direct comparison shows that the two TROPOMI products exhibit the highest consistency, whereas GOME-2 SIF shows the largest discrepancies. A key finding is that when a product combination includes two products that share observations from the same sensor, TCH produces a structural bias due to violation of the error-independence assumption: it systematically underestimates the errors of the shared-source products while overestimating the errors of the independent product. By allowing non-zero error cross-covariance between a pre-specified product pair, QCA effectively corrects this bias. Based on QCA estimates, over 97% of valid vegetated pixels for OCO-2 SIF, TROPOSIFCaltech, and TROPOSIFESA exhibit RMSE values below 0.2 W·m-2·sr-1·μm-1, whereas over 77% of GOME-2 SIF pixels are concentrated in the 0.2-0.4 W·m-2·sr-1·μm-1 range. All four products show the highest errors in evergreen broadleaf forests and the lowest errors in closed shrublands. Spatially, coarse-resolution products are more sensitive to spatial aggregation; temporally, removing the common seasonal cycle substantially reduces error estimates for all products, with OCO-2 SIF showing the largest reduction, approximately 52%. Collectively, these findings reveal that error estimates from cross-evaluation are not intrinsic product attributes but conditional outcomes of the analytical framework. The proposed diagnostic framework is transferable to uncertainty analyses of other satellite-retrieved geophysical variables, offering a rigorous methodological foundation for remote sensing product validation and fusion.

Hyperspectral images offer an efficient and reliable approach for monitoring soil heavy metal (SHM) contents over extensive regions. However, neglecting environmental covariates and limited soil samples can reduce the accuracy of SHM prediction models. Therefore, this study proposes a prediction method integrating environmental covariates and applying data augmentation to improve the reliability and accuracy of Pb, Zn, and Ni prediction. Firstly, geographic detector and Shapley additive explanations were employed to identify key environmental covariates, which were characterized by relevant spectral indices for modeling. Subsequently, a spectral smoothing-constrained Wasserstein generative adversarial network with gradient penalty (SS-WGAN-GP) was proposed to increase the sample size and enhance the reliability and realism of generated samples. The TabNet model was finally used to construct the SHM prediction models and generate SHM spatial distribution maps. The proposed method was validated using 105 topsoil samples collected from the Dongsheng coalfield in Inner Mongolia and a corresponding GF-5 hyperspectral image. The results indicate that the environmental covariates significantly improved model accuracy. Further enhancement of model performance was achieved by augmenting samples using the SS-WGAN-GP, yielding R² values of 0.70, 0.70, and 0.72 for Pb, Zn, and Ni, respectively. Spatial mapping of SHM contents and pollution levels based on the Nemerow Pollution Index suggested that elevated SHM contents were mainly distributed around mining areas and along major roads. This study provides a feasible framework for SHM prediction and pollution risk assessment under limited sample conditions.

Direct validation through in-situ measurements is a primary approach for assessing the accuracy of soil moisture (SM) products. However, validation results carry significant uncertainties due to errors inherent in ground measurements, representativeness errors, and geolocation mismatches between in-situ sites and product pixels. The mechanisms through which these factors contribute to the overall uncertainty of pixel-scale reference “truth” remain unclear, complicating uncertainty control. This study is the first to investigate the uncertainty of validation results obtained from in-situ measurements. It reveals the magnitude and influencing factors of representativeness errors, proposes a method for identifying pixel geolocation shifts in sub-pixel-scale satellite SM products, and quantifies the resulting errors. The results indicate that representativeness errors are significantly larger than errors caused by geolocation mismatches, dominating the overall uncertainty of pixel-scale reference “truth”. Representativeness errors result in an underestimation of SM product consistency and an overestimation of uncertainty. More than half of the sites exhibit representativeness errors exceeding 40%, with a maximum reaching 154%. Representativeness errors are primarily determined by spatial heterogeneity within the validation pixel and the number and spatial location of in-situ sites. The impact of spatial heterogeneity can be mitigated by optimizing site locations, and increasing the number of sites to 3 can reduce representativeness errors to below 5%. Errors due to geolocation mismatches are related to the degree of pixel shift and the spatial heterogeneity of the surrounding area. This type of error exceeds 15% approximately half of the time, with a maximum reaching 51.8%.

Thermal infrared remote sensing, capable of producing global land surface temperature (LST) datasets, is extensively utilized in geophysical, ecological, and environmental studies. However, the instantaneous and directional nature of satellite observations leads to inconsistencies in LST products due to variations in local time and viewing geometry. To enable consistent LST applications, it is valuable to generate an LST dataset at fixed local time and fixed viewing angle, which requires integrating the diurnal temperature cycle (DTC) model (temporal dimension) with a kernel-driven model (KDM, angular dimension). To this end, we developed a novel angle-evolving DTC (AEDTC) modeling framework and instantiated four AEDTC models by combining two gap-fraction kernels and two hotspot kernels. These four models were calibrated using asynchronous multi-angle LST observations within one day from current 1 km polar-orbiting satellites. Results showed that AEDTC models outperformed the traditional DTC model (which ignores angular differences), reducing RMSE from 1.58 K to 1.12 K (with a 29.1% reduction). Moreover, the calibrated AEDTC models can generate angularly normalized LST (i.e., hemispherical LST) at any local time. Using in-situ pyrgeometer measurements as reference, the RMSE (MBE) of daytime LST products decreased from 2.56 K (0.96 K) to 1.88 K (0.46 K), representing reductions of 26.6% (52.1%). At night, the normalized hemispherical LST showed similar accuracy to directional LST due to the diminished angular effect. Overall, the AEDTC models could produce temporally and angularly consistent LST products, enhancing the reliability of LST-related applications.

Accurate reconstruction of fine-resolution, spatially temporally continuous land surface albedo is critical for energy balance studies and climate change modeling. This paper proposes a dual-constrained optimal image matching strategy (DC-OPS) for reconstructing seamless 16 m daily albedo by integrating 500 m MODIS and 16 m GF-1 WFV products through the flexible spatiotemporal data fusion (FSDAF) model. The strategy combines temporal proximity and texture similarity to identify optimal reference image pairs, enhancing reconstruction accuracy. Validated in the Huailai region and the Tibetan Plateau (TP), the results show high accuracy. In Huailai, DC-OPS yielded an RMSE of 0.0304 against original GF-1 images and 0.0282 against in situ measurements. In the TP region, the method maintained a reliable RMSE of 0.0377 despite extreme data scarcity. Comparative analysis demonstrated that DC-OPS outperforms traditional methods such as harmonic analysis of time series (HANTS) and neighborhood similar pixel interpolator (NSPI), especially in capturing rapid surface changes such as snowfall. The strategy exhibits robust applicability across diverse landscapes, providing a foundation for large-scale applications by enhancing spatial temporal continuity.

Pixel-scale reference ”truth” serves as a critical link between in situ measurements and satellite observations. However, owing to surface heterogeneity, the upscaling of single-site observations to the satellite pixel scale has long been constrained by poorly understood mechanisms of heterogeneity effects, inadequate characterization of nonlinear responses, and insufficient utilization of spatial information. Existing methods, particularly machine learning approaches, generally adopt a direct point-to-pixel mapping strategy that implicitly assumes mutual independence among subpixels within a coarse pixel. This assumption conflicts with the spatial autocorrelation inherent in a spatially continuous land surface, thereby introducing systematic biases into upscaling results over heterogeneous land surfaces. To address this challenge, we propose a site-to-pixel upscaling method that integrates environmental attribute modeling with spatial context learning. The core of this method is a parallel dual-branch architecture comprising a multilayer deep neural network and a convolutional neural network. The former establishes a nonlinear mapping between environmental attributes and the target variable, whereas the latter extracts multiscale spatial heterogeneity patterns within the coarse pixel. The features from the two branches are then adaptively fused through a gating mechanism, followed by spatial aggregation using a point spread function. Evaluation across 416 globally distributed sites encompassing 15 land cover types shows that the model achieves R² values of 0.802–0.997 and RMSE values of 0.007–0.032. At sites jointly affected by spatial heterogeneity and topographic complexity, the median RMSE is more than 30% lower than that of the best-performing benchmark model. Ablation experiments show that the combined mechanism of per-subpixel modeling and residual learning is the primary source of model accuracy improvement, reducing RMSE by more than 60%. The complete dual-branch model accurately reproduces the spatial texture and spatial variability patterns within coarse pixels. Using MCD43A3 product validation as an example, the upscaled reference reduces validation RMSE by 10.2%–22.4%; at highly heterogeneous sites, Bias is reduced by 85.7%. This study demonstrates that the proposed method effectively enhances the ability of in situ observations to represent pixel-scale conditions over heterogeneous surfaces, providing a reliable and cost-effective solution for generating high-quality pixel-scale reference ”truth” and a fundamental basis for remote sensing product validation and the construction of pixel-scale training samples.

Upscaling sparse in situ soil moisture (SM) observations to the pixel scale is hindered by site representativeness error, particularly over heterogeneous surfaces. Existing methods generally treat this task as a point-to-pixel conversion problem without explicitly accounting for differences in site representativeness. This study presents a representativeness-guided upscaling method for deriving pixel-scale reference SM from sparse in situ observations. The method first quantifies site representativeness error and then generates a 500-m SM trend surface from multisource remote sensing variables using an automated machine-learning stacked ensemble to provide prior spatial information under sparse sampling conditions. On this basis, a progressively enhanced adaptive regression scheme is developed to estimate upscaling coefficients. Experiments over a dense monitoring network on the Qinghai-Tibet Plateau show that representativeness error varies substantially across sites, ranging from 11% to 154%, with more than half of the sites exceeding 40%. For sites with medium and low representativeness, the proposed method reduces median representativeness error from 50.13% to about 10% and from 129.44% to about 20%, respectively. When used to construct pixel-scale reference SM for evaluation of the SMAP-DCA ascending product, the method reduces Bias and RMSE by 59.7% and 51.2%, respectively, for a medium-representativeness site, and by 86.5% and 77.5%, respectively, for a low-representativeness site. These results indicate that site representativeness error is a major source of uncertainty in point-to-pixel upscaling of sparse in situ SM observations and that explicitly accounting for it improves the reliability of pixel-scale reference SM under heterogeneous conditions.

Ground observation data are critical for quantitative remote sensing, yet current sampling is hindered by scale mismatch, regional bias, and parameter isolation. This study proposes a multiparameter collaborative ground site optimization method to synchronously enhance representativeness at both regional and pixel scales. At the regional scale, principal component analysis (PCA) and K-means clustering are integrated to select representative pixels characterized by attribute typicality and spatial independence. At the pixel scale, an exhaustive combinatorial optimization algorithm identifies the most representative points by minimizing a comprehensive multiparameter spatiotemporal objective function. Case study results from the Heihe River Basin demonstrate that at the regional scale, the covariance determinant of the selected sample areas reaches 149.44% of the entire study area, indicating superior feature space coverage. At the pixel scale, deploying no more than three points maintains spatiotemporal representativeness errors (REs) within 3% for most parameters. Compared with single-parameter methods, this approach achieves a multifold increase in sampling efficiency while maintaining comparable accuracy and costs. The proposed method realizes an optimized configuration for one-time deployment, multiscale synergy, and multiparameter sharing. Without relying on prior spatial assumptions, it provides a universal and robust ground-truth foundation for intelligent inversion models and multisource product validation.

The photochemical reflectance index (PRI) has been proven to be a robust proxy for photosynthetic light use efficiency (LUE). Prior studies have shown that PRI alone exhibits a weak correlation with LUE in croplands, because the observed PRI is significantly influenced by canopy structure and pigment pools, whereas LUE is not driven by these factors. To address this issue, studies have separated the PRI into two components: the early morning PRI (PRI0) to capture the seasonal change of canopy structure and pigment pools, and the diurnal change of PRI (PRId) to track the xanthophyll cycle, allowing them to monitor the seasonal and diurnal change of LUE in forest stands. However, few studies have applied this method to track the LUE in croplands, particularly in paddy rice fields. Moreover, this approach requires continuous, high-frequency (<= 30-min) observations of PRI using hyperspectral cameras mounted on towers or drones, which is challenging to achieve in natural environments. To address these challenges, we developed an ultra-spatial resolution multispectral camera to obtain high-frequency PRI observations economically in a paddy rice field, and explored the diurnal relationship between LUE and PRId. Results showed that PRId effectively tracked the diurnal cycle of LUE and captured the stress of atmospheric dryness on plant photosynthesis. The coefficient of determination (R2) between PRId and LUE was 0.65, significantly higher than that between PRI and LUE (0.46) during the greening-up period with significant changes of canopy structure and pigment pools. This indicated that separating PRI into PRI0 and PRId is an effective approach for linking PRI observations to LUE in paddy rice fields.

Retrieving high-quality and temporal continuous land surface temperature (LST) over structurally complex surfaces remains difficult because satellite observations are constrained by the revisit cycle and carry their own retrieval uncertainties. This research presents a data assimilation framework that integrates a physics-based radiative transfer and surface energy balance model(STREAM)with the Ensemble Kalman Filter (EnKF) to improve LST estimation by sequentially incorporating Landsat observations. Fifteen scenes of Landsat 8/9 LST products during May to June in 2023 were assimilated to dynamically calibrate key parameters of STREAM model which can derive temporal continuous LST, and the parameters including soil moisture (SM), vegetation canopy density (VCD), and the leaf area index(LAI) conversion coefficient. Validation against 15 in-situ station measurements at the Huailai Remote Sensing Experimental Station of CAS yields two main conclusions: (1) at satellite overpass moments, assimilation reduces station-level RMSE from 4.12 K (physics-only simulation) to 3.26 K, outperforming the Landsat LST product itself (RMSE = 4.02 K); (2) over the full growing season, the proposed method estimates hourly LST only by assimilating the Landsat 8/9 LST products with around 8-day interval, and the reconstruction achieves an overall RMSE of 3.35 K with a near-zero bias of −0.09 K comparing with the in-situ measurements, the results demonstrate that the inter-overpass LST estimates kept a comparable level of accuracy with the Landsat 8/9 LST products, which was practically acceptable to some extent. These results support satellite-constrained data assimilation as a viable and effective way to achieve improved temporal continuous LST estimation.

Hyperspectral images provide an efficient means for large-scale predicting of soil organic matter (SOM) content, yet its accuracy is often hindered by soil moisture effects and limited soil sample size. To address these challenges, this study proposes a novel method that integrates soil spectrum correction and sample expansion to improve SOM content prediction accuracy. An improved orthogonal signal correction (OSC) algorithm using visible and shortwave infrared drought index (VSDI) as a reference is developed to correct soil spectra and reduce external parameter reliance. Additionally, a sample expansion algorithm is developed to enhance sample diversity and reduce overfitting, integrating the soil spectral response mechanism with the spatial autocorrelation among samples. Finally, the hybrid Back Propagation Neural Networks-Random Forest (BPNN-RF) model is applied to predict SOM content. The proposed method was validated by 80 topsoil samples and ZiYuan-1 02D (ZY1E) hyperspectral images in Nong'an County, Jilin Province, China. The results indicate that the improved OSC algorithm effectively corrected the soil moisture effects and enhanced spectral sensitivity to SOM, increasing the average absolute correlation coefficient from 0.34 to 0.41, with a maximum value exceeding 0.50. Sample expansion improved model performance (the coefficient of determination (R2) increased from 0.42 to 0.71, the root-mean-square error (RMSE) decreased from 0.34 % to 0.24 %), and combining it with soil spectral correction further raised R2 to 0.81 and reduced RMSE to 0.19 %. SHAP analysis revealed that the top 20 important bands fell within SOM-sensitive ranges. The distribution pattern of predicted SOM content map was inverse to that of the Digital Elevation Model (DEM) map yet consistent with that of the annual average precipitation map. Thus, this method improves the spatiotemporal adaptability of SOM prediction using hyperspectral images, offering a robust approach for rapid and large-scale soil monitoring.

Land surface temperature (LST) serves as a critical state variable governing Earth system interaction. Although thermal infrared remote sensing allows for global LST monitoring, most studies are based on the assumption of a homogenous surface, and most data fusion studies overlook the effect of complex surface structural characteristics, which aim for a trade-off between spatial and temporal resolutions. To address this challenge, this study introduces a radiative transfer and energy balance-coupled model (STREAM) for global simulations, which explicitly accounts for the effect of vegetation structure in surface physical processes. This enables the simulation of land surface component temperatures (LSCT) and brightness temperatures in a unified manner. The proposed model is validated against satellite directional observations from Sentinel-3 SLSTR and in-situ measurements from FLUXNET, with its performance compared to the widely used SCOPE model. Results demonstrate that: (1) Based on SLSTR observations, STREAM effectively mitigates the underestimation of LST found in the SCOPE model, reducing the overall bias against SLSTR observations from −3.03 K to −1.77 K, and the root mean squared error (RMSE) from 4.52 K to 4.06 K; (2) Based on FLUXNET measurements, STREAM reduces RMSE from 3.36 K to 2.52 K compared to the SCOPE model. This work establishes STREAM as a robust and efficient physical framework for generating global temperature, which highlights a gap in current modeling capabilities: high-fidelity three-dimensional models are too computationally intensive for global-scale applications, whereas prevalent one-dimensional models have limitations in modeling complex vegetation types.

The optimized sampling is very important for obtaining representative observations over heterogeneous surfaces. However, spatial heterogeneity (SH) is influenced by both randomness and structure factors and varies with scale. A comprehensive understanding of how the contribution of these factors to SH varies with scale is crucial for optimizing sampling. This study quantified the scale dependence of SH caused by structure and randomness factors based on the geostatistical attributes of semivariogram and explored the relationship between the optimal deployment of ground samples and SH dominated by randomness or structure factors. The results showed that as the plot size increased, the proportion of SH caused by spatial structure factors ( P} (SSF)) increased. When the plot size was larger than 20 m, the P-SSF gradually approached 80%-100%. As the plot size increased, the optimal sample plots were distributed on the typical structural features in the area. However, when the plot size was small, the optimal samples were not necessarily located on the predominant surface types. Optimizing sampling can characterize the SH, and the optimal deployment of ground samples should comprehensively consider the plot size and the number of sample plots.

Natural surface is non-isothermal frequently. In this situation unavoidable spectral bias of measured radiance will arise due to that the high-temperature components contribute radiance intensity in shorter wavelengths more than in longer wavelengths [1]. All the existing Temperature Emissivity Separation (TES) algorithms assumed that temperature was independent of wavelength[2]. The spectral bias will be residual in the retrieved emissivity[3]. To overcome the residual spectral bias in retrieved emissivity for nonisothermal surface, we accept the conception that the radiative temperature changes with the spectrum wavelength which was proposed by [4]. For the first time, this paper propose a new algorithm to retrieve both the temperature and the emissivity changing with wavelength. A group of artificial non-isothermal surface measurements were used to validate the new algorithm. The emissivities retrieved by the original algorithm had apparent spectral bias changed with component temperature difference, while the new algorithm can obtain stable emissivity spectral curves and effectively eliminates the spectral bias in emissivity.

Land surface temperature (LST) is a critical parameter in the study of land surface water and thermal environments. To better understand and address the challenges faced by satellite LST products in applications, conducting global-scale radiative transfer modeling is essential. In this study, the system for tracing radiative transfer and energy balance activities monitoring (STREAM) was developed to simulate global land surface thermal environments. Using 2019 summer data from the Huang-Huai-Hai Plain as a case study, LST results derived from the SLSTR were used as a reference to validate the simulation outcomes. The validation results indicate that, compared to ERA5 LST, STREAM simulations show significant accuracy improvements over heterogeneous surfaces, such as forest canopies, with reductions in RMSD of 1.28 K and 1.84 K in the nadir and oblique directions, respectively. The STREAM model is suitable for large-scale global land surface temperature simulations and can support global ecological and environmental monitoring efforts.

Linear Variable Filter (LVF) hyperspectral cameras possess the advantages of high spectral resolution, compact size, and light weight, making them highly suitable for unmanned aerial vehicle (UAV) platforms. However, challenges arise in data registration due to the imaging characteristics of LVF data and the instability of UAV platforms. These challenges stem from the diversity of LVF data bands and significant inter-band differences. Even after geometric processing, adjacent flight lines still exhibit varying degrees of geometric deformation. In this paper, a progressive grouping-based strategy for iterative band selection and registration is proposed. In addition, an improved Scale-Invariant Feature Transform (SIFT) algorithm, termed the Double Sufficiency–SIFT (DS-SIFT) algorithm, is introduced. This method first groups bands, selects the optimal reference band, and performs coarse registration based on the SIFT method. Subsequently, during the fine registration stage, it introduces an improved position/scale/orientation joint SIFT registration algorithm (IPSO-SIFT) that integrates partitioning and the principle of structural similarity. This algorithm iteratively refines registration based on the grouping results. Experimental data obtained from a self-developed and integrated LVF hyperspectral remote sensing system are utilized to verify the effectiveness of the proposed algorithm. A comparison with classical algorithms, such as SIFT and PSO-SIFT, demonstrates that the registration of LVF hyperspectral data using the proposed method achieves superior accuracy and efficiency.

Validation,inversion,and scale problems have been listed as the three major scientific problems in quantitative remote sensing.Validation serves as a crucial foundation for reflecting and revealing the errors in remote sensing algorithms and products.It is also an important guarantee for the continuous improvement of remote sensing product quality.After nearly 40 years of development,validation has received widespread attention in the international remote sensing community.Thus far,the theory and methods of validation have become relatively mature,and an increasing number of practical validation works have been conducted.This scenario has played a wide role in clarifying the error distribution of remote sensing products,thereby iteratively improving the quality of remote sensing products and promoting the application benefits of remote sensing products.However,given the development of remote sensing science and technology,the connotation of validation should not be limited to assessing the accuracy of remote sensing products.The in-depth expansion of various macro and micro applications related to geography objectively promotes the proactive and systematic global analysis of various uncertainties in the entire process of remote sensing information from data acquisition to application from the perspective of remote sensing science and technology disciplines.As a result,the technological conditions and advantages of the era are linked,and the iterative space observation capabilities are improved.With 40 years of academic accumulation,particularly the development of remote sensing product validation technology,the theory,method,and technology of validation have evolved into a comprehensive system from traditional comparative analysis based on statistics to simulation validation grounded in physical models.At present,we have the validation capability for the whole chain of remote sensing,including quantitative remote sensing mechanism model validation,satellite imaging data calibration,remote sensing data product assessment,application effect evaluation,and even remote sensing observation theory and methods.However,remote sensing information has multidimensional characteristics of time,space,spectrum,and events.In previous research,validation techniques and methods were developed for various stages,including primary data product processing,model/algorithm evaluation,production,and application.However,these stages were merely linked through simplistic,rigid input-output relationships.The correlations and inheritance of uncertainties across these stages were not explored.This rigid model overlooked the intrinsic connections and mutual influences among various stages.Thus,validation was often limited to individual stages rather than addressing problems systematically as a whole.Merely"evaluating"the accuracy and uncertainty of remote sensing products is far from sufficient because the ultimate goal of validation is to enhance further the quality of remote sensing products.Systematically reorganizing the connotation,methodology,and output of validation,as well as forming a working mechanism for interdisciplinary cooperation within and across disciplines,is necessary to enhance remote sensing spatial observation capabilities systematically through validation.This study provides a new interpretation of the concept and connotation of validation,analyzes and summarizes the current status of methods and technological development,and examines the key challenges that urgently need to be overcome in validation at present.Finally,this study provides a view of the specific ideas and development prospects of validation in the future.