论文发表 | 郭善昕 Shanxin Guo

2025

A knowledge-informed machine learning (KIML) approach to disentangle single- and double-season rice yields and attainable yield gaps in the mixed cropping systems

Jingwen Wang, Feng Maoyuan, Guo Shanxin, and 3 more authors

Computers and Electronics in Agriculture, 2025

Abs

Understanding and quantifying the yield gap between current and potential yields provides a foundation for improving crop productivity and guiding agricultural policies to enhance food security. However, the yield gaps of single- and double- season rice are often entangled in mixed cropping systems such that their estimates remain highly uncertain. This study developed a knowledge-informed machine learning (KIML) method to disentangle the yields and yield gaps of single and double rice in mixed cropping systems. Specifically, we incorporated satellite-derived knowledge of cropping intensity into the extreme gradient boosting (XGBoost) model as feature inputs and a constrained loss function. Our model was applied to a typical mixed cropping region in South China (Hubei, Hunan, Jiangxi, and Guangdong Provinces). Results show that the KIML model outperforms pure ML in capturing the nonlinear relationship between predictors and county-scale rice yields, with a correlation (r) of 0.88–0.91 and relative root mean squared error (RRMSE) of 9.1 %–10.3 %. Compared to the pure ML model, the KIML model effectively disentangled the yield estimates of single and double rice, reducing their biases from − 9.2 % to − 2.6 % and from 3.7 % to 1.4 %, respectively. Moreover, the yield gap estimates from KIML are consistent with the Global Agro-Ecological Zones (GAEZ) datasets from the Food and agricultural Organization (FAO) (r = 0.65, RRMSE = 20.8 %), while the pure ML model exhibits a lower correlation (r = 0.37) and higher RRMSE (32.9 %). In South China, our estimates indicate a potential increase in rice yields by 17.1 % and a production gap of 12.35 million tons (Mt), with single and double rice contributed by 46.2 % (5.70 Mt) and 53.8 % (6.65Mt), respectively. Previous aggregate estimates overstated the yield gap and production gap by 34.1 % and 40.0 %, respectively. Overall, this study provides an effective tool to better quantify the current yields and yield gaps in mixed cropping systems, which is critical to policy making and agriculture management for the purpose of food security.
Incorporating Physical Constraint in Three-Dimensional Time Series InSAR Inversion for Urban Deformation Monitoring

Luyi Sun, Hongzhong Li, Shanxin Guo, and 3 more authors

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2025

Abs DOI

Spaceborne Interferometric Synthetic Aperture Radar (InSAR) has become a key method for remote sensing monitoring of surface deformations, providing millimeter-level accuracy along the satellite’s Line-of-Sight (LOS). However, precise understanding of deformation mechanisms requires threedimensional (3D) measurements. Challenges in obtaining long time series and high-density in-situ measurements, along with the limited sensitivity of polar-orbiting SAR satellites to north-south displacements, often lead to suboptimal horizontal accuracy in conventional 3D inversion. This study proposes a 3D inversion method for time series InSAR measurements, specifically designed for scenarios with single-track SAR data, common in many areas where free data is limited. The method introduces a physical constraint between the horizontal deformation and the spatial gradients of vertical displacement, concentrated on urban areas where deformations are typically slow, long-term, and small in magnitude. Considering the localized and different drivers of urban deformations, such as diverse settlement patterns from tunneling activities and ocean reclamation, we propose a pixel-bypixel estimation approach for the proportionality factor 𝜷, the key parameter in the physical constraint. This approach recognizes that 𝜷 varies across locations, contrasting with the conventional assumption of a uniform 𝜷 for the entire study area. An optimization process with gradient descent is introduced to iteratively refine 𝜷 , accelerating convergence and achieving a robust solution. Experiments were conducted in representative scenarios, including ocean-reclaimed and metro-tunneling areas, with validation through leveling measurements. The results demonstrate that the proposed method achieves precise 3D inversion and effectively captures the complex dynamics of urban surface deformation.
Field-scale evaluation of a satellite-based terrestrial biosphere model for estimating crop response to management practices and productivity

Jingwen Wang, Jose Luis Pancorbo, Miguel Quemada, and 5 more authors

ISPRS Journal of Photogrammetry and Remote Sensing, 2025

Abs DOI

Timely and accurate information on crop productivity is essential for characterizing crop growing status and guiding adaptive management practices to ensure food security. Terrestrial biosphere models forced by satellite observations (satellite-TBMs) are viewed as robust tools for understanding large-scale agricultural productivity, with distinct advantages of generalized input data requirement and comprehensive representation of carbon–water-energy exchange mechanisms. However, it remains unclear whether these models can maintain consistent accuracy at field scale and provide useful information for farmers to make site-specific management decisions. This study aims to investigate the capability of a satellite-TBM to estimate crop productivity at the granularity of individual fields using harmonized Sentinel-2 and Landsat-8 time series. Emphasis was placed on evaluating the model performance in: (i) representing crop response to the spatially and temporally varying field management practices, and (ii) capturing the variation in crop growth, biomass and yield under complex interactions among crop genotypes, environment, and management conditions. To achieve the first objective, we conducted on-farm experiments with controlled nitrogen (N) fertilization and irrigation treatments to assess the efficacy of using satellite-retrieved leaf area index (LAI) to reflect the effect of management practices in the TBM. For the second objective, we integrated a yield formation module into the satellite-TBM and compared it with the semi-empirical harvest index (HI) method. The model performance was then evaluated under varying conditions using an extensive dataset consisting of observations from four crop species (i.e., soybean, wheat, rice and maize), 42 cultivars and 58 field-years. Results demonstrated that satellite-retrieved LAI effectively captured the effects of N and water supply on crop growth, showing high sensitivity to both the timing and quantity of these inputs. This allowed for a spatiotemporal representation of management impacts, even without prior knowledge of the specific management schedules. The TBM forced by satellite LAI produced consistent biomass dynamics with ground measurements, showing an overall correlation coefficient (R) of 0.93 and a relative root mean square error (RRMSE) of 31.4 %. However, model performance declined from biomass to yield estimation, with the HI-based method (R = 0.80, RRMSE = 23.7 %) outperforming mechanistic modeling of grain filling (R = 0.43, RRMSE = 43.4 %). Model accuracy for winter wheat was lower than that for summer crops such as rice, maize and soybean, suggesting potential underrepresentation of the overwintering processes. This study illustrates the utility of satellite-TBMs in crop productivity estimation at the field level, and identifies existing uncertainties and limitations for future model developments.

2024

The Improved U-STFM : A Deep Learning-Based Nonlinear Spatial-Temporal Fusion Model for Land Surface Temperature Downscaling

Shanxin Guo, Min Li, Yuanqing Li, and 6 more authors

Remote Sensing, 2024
Forest Ecology and Management Forest fire susceptibility mapping based on precipitation-constrained cumulative dryness status information in Southeast China : A novel machine learning modeling approach

Longlong Zhao, Yuankai Ge, Shanxin Guo, and 3 more authors

Forest Ecology and Management, 2024

DOI

2023

Stability Analysis of Unmixing ‐ Based Spatiotemporal Fusion Model : A Case of Land Surface Temperature Product Downscaling

Min Li, Shanxin Guo, Jinsong Chen, and 5 more authors

Remote Sensing, 2023
Statistical Assessments of InSAR Tropospheric Corrections : Applicability and Limitations of Weather Model Products and Spatiotemporal Filtering

Luyi Sun, Jinsong Chen, Hongzhong Li, and 2 more authors

Remote Sensing, 2023
Object-Oriented Remote Sensing Image Change Detection Based on Color Co-Occurrence Matrix

Zhu Zhu, Tinggang Zhou, Jinsong Chen, and 4 more authors

Applied Sciences, 2023

Abs DOI

Featured Application: Change detection for remote sensing images. Aiming at the problem of misdetection caused by the traditional texture characteristic extraction model, which does not describe the correlation among multiple bands, an object-oriented remote sensing image change detection method based on a color co-occurrence matrix is proposed. First, the image is divided into multi-scale objects by graph-based superpixel segmentation, and the optimal scale is determined by the overall goodness F-measure (OGF). Then, except for the extraction of the spectral features, the multi-channel texture features based on the color co-occurrence matrix (CCM) are extracted to consider the correlation among multiple bands. To accurately find the representative features to overcome the impact of feature redundancy, a cumulative backward search strategy (CBSS) is further designed. Finally, the change detection is completed by inputting the difference image of dual time points to the trained random forest model. Taking Shenzhen and Dapeng as the study areas, with Sentinel-2 and Skysat images under different spatial resolutions, and the forest–bareland change type as an example, the effectiveness of the proposed algorithm is verified by qualitative and quantitative analyses. They show that the proposed algorithm can obtain higher detection accuracy than the texture features without band correlation.
Superpixel Segmentation Based on Anisotropic Diffusion Model for Object-oriented Remote Sensing Image Classification

Xiaoli Li, Jinsong Chen, Longlong Zhao, and 6 more authors

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023

Abs DOI

Superpixel segmentation is an essential step of object-oriented remote sensing image classification, the accuracy of the superpixel segmentation boundary will directly affect the classification result. Most of the traditional superpixel segmentation algorithms rely on spectral similarity and spatial connectivity to construct superpixels. They can not find the accurate boundary in the complex scenes, such as the spatial distribution of ground features being relatively broken, and large differences in the size and shape, especially long-thin shape and circular shape. Aiming at this problem, a superpixel segmentation algorithm based on anisotropic diffusion model named ADS is proposed and applied to image classification. The anisotropic diffusion model originated in thermodynamics has excellent properties in that the diffusion is continuous and smooth and its diffusion speed depends on the medium, which provides convenience for smoothing homogeneous regions and establishing boundary constraints for different ground objects. With this advantage, the diffusion flux model is established to consider the influence of boundary factors and used to simulate the dissimilarity measure with boundary constraints between pixels and seed points by combining the traditional spectral and spatial distance. Then, the seed points of superpixel are optimized under the K-means framework. The effectiveness of the proposed algorithm is tested and verified with different spatial resolutions, such as Landsat8 with 30m, Sentinel-2 with 10m, and Skysat with 0.5m. A large number of experiments show that the proposed algorithm can better correct the superpixel boundary fitting deviation problem in complex scenes and effectively promote the improvement of image classification accuracy.

2022

Hyperspectral Image Classification with Imbalanced Data Based on Semi-Supervised Learning

Xiaorou Zheng, Jianxin Jia, Jinsong Chen, and 4 more authors

Applied Sciences, 2022
High Quality Object Detection for Multiresolution Remote Sensing Imagery Using Cascaded Multi-Stage Detectors

Binglong Wu, Yuan Shen, Shanxin Guo, and 3 more authors

Remote Sensing, 2022

2021

PyCLiPSM: Harnessing heterogeneous computing resources on CPUs and GPUs for accelerated digital soil mapping

Guiming Zhang, A-Xing Zhu, Jing Liu, and 2 more authors

Transactions in GIS, 2021
Deep Learning Network Intensification for Preventing Noisy-Labeled Samples for Remote Sensing Classification

Chuang Lin, Shanxin Guo, Jinsong Chen, and 4 more authors

Remote Sensing, 2021
Full Parameter Time Complexity (FPTC): A Method to Evaluate the Running Time of Machine Learning Classifiers for Land Use/Land Cover Classification

Xiaorou Zheng, Jianxin Jia, Shanxin Guo, and 4 more authors

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021

Abs DOI

In emergency responses to natural disasters, actionable information provided by remote sensing images is crucial to help emergency managers become aware of the situation and assess the magnitude of the damage. Without the accurate prediction of time consumption, choosing an algorithm for land use/land cover (LULC) classification under these emergency circumstances could be blind and subjective. Here, we proposed a full parameter time complexity (FPTC) analysis and the corresponding coefficient ωto estimate the actual running time of the LULC classification without actually running the code. The FPTC of five general algorithms is derived in this article. After derivation, the FPTC of k-nearest neighbors (kNN) is F(nv+n{\text{log}}_2 u), the FPTC of logistic regression (LR) is F(Qm2vn), the FPTC of classification and regression tree (CART) is F((m+1)nv{\text{log}}_2n), the FPTC of random forest (RF) is F(s(m+1)nv{\text{log}}_2n), and the FPTC of support vector machine (SVM) is F(m2Qv (n+k)). The results show a strong linear relationship between the actual running time and FPTC [R-squared: kNN (0.991), LR (0.997), CART (0.999), RF (1.000), and SVM (0.999)], with different data size. The average root-mean-squared error between the real running time and the estimated running time is 3.34 s, which demonstrates the effectiveness of FPTC. Combining FPTC with the corresponding coefficient ω, the running time of the classification can be precisely predicted, which will help emergency managers quickly choose algorithms in response to natural disasters with available remote sensing data and limited time.

2020

Identification of Short-Rotation Eucalyptus Plantation at Large Scale Using Multi-Satellite Imageries and Cloud Computing Platform

Xinping Deng, Shanxin Guo, Luyi Sun, and 1 more author

Remote Sensing, 2020

DOI
High-Resolution U-Net : Preserving Image Details for Extraction Cultivated Land

Wenna Xu, Xinping Deng, Shanxin Guo, and 6 more authors

Sensors, 2020
Removing Stripe Noise Based on Improved Statistics for Hyperspectral Images

Jianxin Jia, Xiaorou Zheng, Shanxin Guo, and 2 more authors

IEEE Geoscience and Remote Sensing Letters, 2020
Adaptive Distance-Weighted Voronoi Tessellation for Remote Sensing Image Segmentation

Xiaoli Li, Jinsong Chen, Longlong Zhao, and 3 more authors

Remote Sensing, 2020

DOI
Improved SRGAN for Remote Sensing Image Super-Resolution Across Locations and Sensors

Yingfei Xiong, Shanxin Guo, Jinsong Chen, and 2 more authors

Remote Sensing, 2020

DOI
Integration of Time Series Sentinel-1 and Sentinel-2 Imagery for Crop Type Mapping over Oasis Agricultural Areas

Luyi Sun, Jinsong Chen, Shanxin Guo, and 2 more authors

Remote Sensing, 2020

2018

MODIS ocean color product downscaling via spatio-temporal fusion and regression: The case of chlorophyll-a in coastal waters

Shanxin Guo, Bo Sun, Hankui K. Zhang, and 5 more authors

International Journal of Applied Earth Observation and Geoinformation, 2018

important paper

DOI

2016

Unification of soil feedback patterns under different evaporation conditions to improve soil differentiation over flat area

Shanxin Guo, A-Xing Zhu, Lingkui Meng, and 4 more authors

International Journal of Applied Earth Observation and Geoinformation, 2016

DOI

2015

Data-Gap Filling to Understand the Dynamic Feedback Pattern of Soil

Shanxin Guo, Lingkui Meng, A-Xing Zhu, and 4 more authors

Remote Sensing, Sep 2015

DOI