scholarly journals Multilayer Soil Moisture Mapping at a Regional Scale from Multisource Data via a Machine Learning Method

2019 ◽  
Vol 11 (3) ◽  
pp. 284 ◽  
Author(s):  
Linglin Zeng ◽  
Shun Hu ◽  
Daxiang Xiang ◽  
Xiang Zhang ◽  
Deren Li ◽  
...  
Keyword(s):  
Machine Learning ◽  
Soil Moisture ◽  
Regional Scale ◽  
Remotely Sensed ◽  
Temporal Variations ◽  
Training Data ◽  
Estimation Accuracy ◽  
Learning Approaches ◽  
Remotely Sensed Data ◽  
Deep Soil

Soil moisture mapping at a regional scale is commonplace since these data are required in many applications, such as hydrological and agricultural analyses. The use of remotely sensed data for the estimation of deep soil moisture at a regional scale has received far less emphasis. The objective of this study was to map the 500-m, 8-day average and daily soil moisture at different soil depths in Oklahoma from remotely sensed and ground-measured data using the random forest (RF) method, which is one of the machine-learning approaches. In order to investigate the estimation accuracy of the RF method at both a spatial and a temporal scale, two independent soil moisture estimation experiments were conducted using data from 2010 to 2014: a year-to-year experiment (with a root mean square error (RMSE) ranging from 0.038 to 0.050 m3/m3) and a station-to-station experiment (with an RMSE ranging from 0.044 to 0.057 m3/m3). Then, the data requirements, importance factors, and spatial and temporal variations in estimation accuracy were discussed based on the results using the training data selected by iterated random sampling. The highly accurate estimations of both the surface and the deep soil moisture for the study area reveal the potential of RF methods when mapping soil moisture at a regional scale, especially when considering the high heterogeneity of land-cover types and topography in the study area.

scholarly journals Effects of Training Set Size on Supervised Machine-Learning Land-Cover Classification of Large-Area High-Resolution Remotely Sensed Data

2021 ◽  
Vol 13 (3) ◽  
pp. 368
Author(s):  
Christopher A. Ramezan ◽  
Timothy A. Warner ◽  
Aaron E. Maxwell ◽  
Bradley S. Price
Keyword(s):  
Machine Learning ◽  
Sample Size ◽  
Remotely Sensed ◽  
Training Data ◽  
Supervised Machine Learning ◽  
Sample Sizes ◽  
Remotely Sensed Data ◽  
Large Area ◽  
Training Set ◽  
Set Size

The size of the training data set is a major determinant of classification accuracy. Nevertheless, the collection of a large training data set for supervised classifiers can be a challenge, especially for studies covering a large area, which may be typical of many real-world applied projects. This work investigates how variations in training set size, ranging from a large sample size (n = 10,000) to a very small sample size (n = 40), affect the performance of six supervised machine-learning algorithms applied to classify large-area high-spatial-resolution (HR) (1–5 m) remotely sensed data within the context of a geographic object-based image analysis (GEOBIA) approach. GEOBIA, in which adjacent similar pixels are grouped into image-objects that form the unit of the classification, offers the potential benefit of allowing multiple additional variables, such as measures of object geometry and texture, thus increasing the dimensionality of the classification input data. The six supervised machine-learning algorithms are support vector machines (SVM), random forests (RF), k-nearest neighbors (k-NN), single-layer perceptron neural networks (NEU), learning vector quantization (LVQ), and gradient-boosted trees (GBM). RF, the algorithm with the highest overall accuracy, was notable for its negligible decrease in overall accuracy, 1.0%, when training sample size decreased from 10,000 to 315 samples. GBM provided similar overall accuracy to RF; however, the algorithm was very expensive in terms of training time and computational resources, especially with large training sets. In contrast to RF and GBM, NEU, and SVM were particularly sensitive to decreasing sample size, with NEU classifications generally producing overall accuracies that were on average slightly higher than SVM classifications for larger sample sizes, but lower than SVM for the smallest sample sizes. NEU however required a longer processing time. The k-NN classifier saw less of a drop in overall accuracy than NEU and SVM as training set size decreased; however, the overall accuracies of k-NN were typically less than RF, NEU, and SVM classifiers. LVQ generally had the lowest overall accuracy of all six methods, but was relatively insensitive to sample size, down to the smallest sample sizes. Overall, due to its relatively high accuracy with small training sample sets, and minimal variations in overall accuracy between very large and small sample sets, as well as relatively short processing time, RF was a good classifier for large-area land-cover classifications of HR remotely sensed data, especially when training data are scarce. However, as performance of different supervised classifiers varies in response to training set size, investigating multiple classification algorithms is recommended to achieve optimal accuracy for a project.

scholarly journals Integrating Remote Sensing, Machine Learning, and Citizen Science in Dutch Archaeological Prospection

2019 ◽  
Vol 11 (7) ◽  
pp. 794 ◽  
Author(s):  
Karsten Lambers ◽  
Wouter Verschoof-van der Vaart ◽  
Quentin Bourgeois
Keyword(s):  
Machine Learning ◽  
Remote Sensing ◽  
Object Detection ◽  
Citizen Science ◽  
Remotely Sensed ◽  
Training Data ◽  
Remotely Sensed Data ◽  
Study Region ◽  
Archaeological Prospection ◽  
Archaeological Object

Although the history of automated archaeological object detection in remotely sensed data is short, progress and emerging trends are evident. Among them, the shift from rule-based approaches towards machine learning methods is, at the moment, the cause for high expectations, even though basic problems, such as the lack of suitable archaeological training data are only beginning to be addressed. In a case study in the central Netherlands, we are currently developing novel methods for multi-class archaeological object detection in LiDAR data based on convolutional neural networks (CNNs). This research is embedded in a long-term investigation of the prehistoric landscape of our study region. We here present an innovative integrated workflow that combines machine learning approaches to automated object detection in remotely sensed data with a two-tier citizen science project that allows us to generate and validate detections of hitherto unknown archaeological objects, thereby contributing to the creation of reliable, labeled archaeological training datasets. We motivate our methodological choices in the light of current trends in archaeological prospection, remote sensing, machine learning, and citizen science, and present the first results of the implementation of the workflow in our research area.

scholarly journals Evaluating Downscaling Factors of Microwave Satellite Soil Moisture Based on Machine Learning Method

2021 ◽  
Vol 13 (1) ◽  
pp. 133
Author(s):  
Hao Sun ◽  
Yajing Cui
Keyword(s):  
Machine Learning ◽  
Soil Moisture ◽  
Land Surface ◽  
Regional Scale ◽  
Support Vector ◽  
Machine Learning Method ◽  
Learning Method ◽  
Feed Forward Neural Network ◽  
Comparative Performance ◽  
Geographical Factors

Downscaling microwave remotely sensed soil moisture (SM) is an effective way to obtain spatial continuous SM with fine resolution for hydrological and agricultural applications on a regional scale. Downscaling factors and functions are two basic components of SM downscaling where the former is particularly important in the era of big data. Based on machine learning method, this study evaluated Land Surface Temperature (LST), Land surface Evaporative Efficiency (LEE), and geographical factors from Moderate Resolution Imaging Spectroradiometer (MODIS) products for downscaling SMAP (Soil Moisture Active and Passive) SM products. This study spans from 2015 to the end of 2018 and locates in the central United States. Original SMAP SM and in-situ SM at sparse networks and core validation sites were used as reference. Experiment results indicated that (1) LEE presented comparative performance with LST as downscaling factors; (2) adding geographical factors can significantly improve the performance of SM downscaling; (3) integrating LST, LEE, and geographical factors got the best performance; (4) using Z-score normalization or hyperbolic-tangent normalization methods did not change the above conclusions, neither did using support vector regression nor feed forward neural network methods. This study demonstrates the possibility of LEE as an alternative of LST for downscaling SM when there is no available LST due to cloud contamination. It also provides experimental evidence for adding geographical factors in the downscaling process.

scholarly journals 3D Convolutional Neural Networks and a CrossDocked Dataset for Structure-Based Drug Design

2020 ◽  
Author(s):  
Paul Francoeur ◽  
Tomohide Masuda ◽  
David R. Koes
Keyword(s):  
Machine Learning ◽  
Ligand Binding ◽  
Binding Affinity ◽  
Mean Squared Error ◽  
Comprehensive Evaluation ◽  
Training Data ◽  
Learning Approaches ◽  
Neural Network Models ◽  
Structure Based Drug Design ◽  
Affinity Prediction

One of the main challenges in drug discovery is predicting protein-ligand binding affinity. Recently, machine learning approaches have made substantial progress on this task. However, current methods of model evaluation are overly optimistic in measuring generalization to new targets, and there does not exist a standard dataset of sufficient size to compare performance between models. We present a new dataset for structure-based machine learning, the CrossDocked2020 set, with 22.5 million poses of ligands docked into multiple similar binding pockets across the Protein Data Bank and perform a comprehensive evaluation of grid-based convolutional neural network models on this dataset. We also demonstrate how the partitioning of the training data and test data can impact the results of models trained with the PDBbind dataset, how performance improves by adding more, lower-quality training data, and how training with docked poses imparts pose sensitivity to the predicted affinity of a complex. Our best performing model, an ensemble of 5 densely connected convolutional newtworks, achieves a root mean squared error of 1.42 and Pearson R of 0.612 on the affinity prediction task, an AUC of 0.956 at binding pose classification, and a 68.4% accuracy at pose selection on the CrossDocked2020 set. By providing data splits for clustered cross-validation and the raw data for the CrossDocked2020 set, we establish the first standardized dataset for training machine learning models to recognize ligands in non-cognate target structures while also greatly expanding the number of poses available for training. In order to facilitate community adoption of this dataset for benchmarking protein-ligand binding affinity prediction, we provide our models, weights, and the CrossDocked2020 set at https://github.com/gnina/models.

scholarly journals Effects of two typical revegetation methods on soil moisture in the semi-arid Loess Plateau, China

2019 ◽  
Vol 50 (5) ◽  
pp. 1453-1462
Author(s):  
Qian Zhao ◽  
Lei Yang ◽  
Xin Wang ◽  
Runcheng Bi ◽  
Qindi Zhang
Keyword(s):  
Soil Moisture ◽  
Loess Plateau ◽  
Temporal Variations ◽  
Chinese Loess Plateau ◽  
Deep Soil ◽  
Chinese Loess ◽  
Soil Desiccation ◽  
Native Grasslands ◽  
First Choice ◽  
Natural Revegetation

Abstract Understanding the effects of vegetation on soil moisture is vital to the ecosystem restoration in water-restricted areas. For this study, the effects of introduced revegetation and natural revegetation on soil water (0–1.8 m) were investigated in the Chinese Loess Plateau, which was based on an in situ vegetation removal experiment and two years of soil moisture monitoring. The results indicated that under introduced revegetation, pasture grassland had lower soil moisture but higher temporal variations over the growing season. Compared with abandoned farmlands and native grasslands under natural revegetation, pasture grasslands revealed greater negative effects on deep soil moisture (1–1.8 m), which was difficult to recover following soil desiccation. In contrast, for abandoned farmlands and native grasslands, the surface soil moisture (0–0.4 m) was mainly impacted, which was easily replenished through rainfall events. These outcomes implied that natural revegetation, rather than introduced revegetation, should be the first choice in water-limited regions toward the rehabilitation of degraded ecosystems.

scholarly journals Monitoring Water Quality of Valle de Bravo Reservoir, Mexico, Using Entire Lifespan of MERIS Data and Machine Learning Approaches

2020 ◽  
Vol 12 (10) ◽  
pp. 1586
Author(s):  
Leonardo F. Arias-Rodriguez ◽  
Zheng Duan ◽  
Rodrigo Sepúlveda ◽  
Sergio I. Martinez-Martinez ◽  
Markus Disse
Keyword(s):  
Machine Learning ◽  
Remote Sensing ◽  
Water Quality ◽  
Water Quality Monitoring ◽  
Temporal Variations ◽  
Quality Parameters ◽  
Parameters Estimation ◽  
Learning Approaches ◽  
Effects Of Drought

Remote-sensing-based machine learning approaches for water quality parameters estimation, Secchi Disk Depth (SDD) and Turbidity, were developed for the Valle de Bravo reservoir in central Mexico. This waterbody is a multipurpose reservoir, which provides drinking water to the metropolitan area of Mexico City. To reveal the water quality status of inland waters in the last decade, evaluation of MERIS imagery is a substantial approach. This study incorporated in-situ collected measurements across the reservoir and remote sensing reflectance data from the Medium Resolution Imaging Spectrometer (MERIS). Machine learning approaches with varying complexities were tested, and the optimal model for SDD and Turbidity was determined. Cross-validation demonstrated that the satellite-based estimates are consistent with the in-situ measurements for both SDD and Turbidity, with R2 values of 0.81 to 0.86 and RMSE of 0.15 m and 0.95 nephelometric turbidity units (NTU). The best model was applied to time series of MERIS images to analyze the spatial and temporal variations of the reservoir’s water quality from 2002 to 2012. Derived analysis revealed yearly patterns caused by dry and rainy seasons and several disruptions were identified. The reservoir varied from trophic to intermittent hypertrophic status, while SDD ranged from 0–1.93 m and Turbidity up to 23.70 NTU. Results suggest the effects of drought events in the years 2006 and 2009 on water quality were correlated with water quality detriment. The water quality displayed slow recovery through 2011–2012. This study demonstrates the usefulness of satellite observations for supporting inland water quality monitoring and water management in this region.

Sign in / Sign up

Export Citation Format

Share Document