Statistical Machine Learning Methods and Remote Sensing for Sustainable Development Goals: A Review

Holloway, J. Leith; Mengersen, Kerrie

doi:10.3390/rs10091365

Cited by 203 publications

(133 citation statements)

References 82 publications

Supporting

Mentioning

128

Contrasting

Unclassified

Order By: Relevance

“…Accordingly, it can be assumed that NRMs must be calibrated crop-specific, as is the case for approaches with the nitrogen nutrition index (NNI). Novel machine learning approaches and algorithms, summarized as artificial intelligence, should also be taken into account for estimating crop traits [93][94][95]. Approaches of estimating grass sward biomass [96], quantifying rice N status [97], or monitoring wheat leaf %N [98] can already be found.…”

Section: Discussionmentioning

confidence: 99%

Estimating Nitrogen from Structural Crop Traits at Field Scale—A Novel Approach Versus Spectral Vegetation Indices

Tilly

Bareth

2019

Remote Sensing

View full text Add to dashboard Cite

A sufficient nitrogen (N) supply is mandatory for healthy crop growth, but negative consequences of N losses into the environment are known. Hence, deeply understanding and monitoring crop growth for an optimized N management is advisable. In this context, remote sensing facilitates the capturing of crop traits. While several studies on estimating biomass from spectral and structural data can be found, N is so far only estimated from spectral features. It is well known that N is negatively related to dry biomass, which, in turn, can be estimated from crop height. Based on this indirect link, the present study aims at estimating N concentration at field scale in a two-step model: first, using crop height to estimate biomass, and second, using the modeled biomass to estimate N concentration. For comparison, N concentration was estimated from spectral data. The data was captured on a spring barley field experiment in two growing seasons. Crop surface height was measured with a terrestrial laser scanner, seven vegetation indices were calculated from field spectrometer measurements, and dry biomass and N concentration were destructively sampled. In the validation, better results were obtained with the models based on structural data (R2 < 0.85) than on spectral data (R2 < 0.70). A brief look at the N concentration of different plant organs showed stronger dependencies on structural data (R2: 0.40–0.81) than on spectral data (R2: 0.18–0.68). Overall, this first study shows the potential of crop-specific across‑season two-step models based on structural data for estimating crop N concentration at field scale. The validity of the models for in-season estimations requires further research.

show abstract

Section: Discussionmentioning

confidence: 99%

Estimating Nitrogen from Structural Crop Traits at Field Scale—A Novel Approach Versus Spectral Vegetation Indices

Tilly

Bareth

2019

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Unlike the simple one-time land-cover classification of the forest/barren areas, the classification of the never-deforested areas/ever-deforested areas is more challenging. The use of more sophisticated machine-learning approaches, such as convolutional NN [50], object-based classification [51], and algorithm improvement to describe reforestation, can mitigate the commission errors. Altered indices to be input into the NN may also change the classification accuracy.…”

Section: Accuracy Of the Integration Mapmentioning

confidence: 99%

Development of an Operational Algorithm for Automated Deforestation Mapping via the Bayesian Integration of Long-Term Optical and Microwave Satellite Data

2019

View full text Add to dashboard Cite

The frequent fine-scale monitoring of deforestation using satellite sensors is important for the sustainable management of forests. Traditional optical satellite sensors suffer from cloud interruption, particularly in tropical regions, and recent active microwave sensors (i.e., synthetic aperture radar) demonstrate the difficulty in data interpretation owing to their inherent sensor noise and complicated backscatter features of forests. Although the sensor integration of optical and microwave sensors is of compelling research interest, particularly in the conduct of deforestation monitoring, this topic has not been widely studied. In this paper, we introduce an operational algorithm for automated deforestation mapping using long-term optical and L-band SAR data, including a simple time-series analysis of Landsat stacks and a multilayered neural network with Advanced Spaceborne Thermal Emission and Reflection Radiometer and Phased Array-type L-band Synthetic Aperture Radar-2, followed by sensor integration based on the Bayesian Updating of Land-Cover. We applied the algorithm over a deciduous tropical forest in Cambodia in 2003-2018 for validation, and the algorithm demonstrated better accuracy than existing approaches, which only depend on optical data or SAR data. Owing to the cloud penetration ability of SAR, observation gaps of optical data under cloudy conditions were filled, resulting in a prompter detection of deforestation even in the tropical rainy season. We also investigated the effect of posterior probability constraints in the Bayesian approach. The land-cover maps (forest/deforestation) created by the well-tuned Bayesian approach achieved 94.0% ± 4.5%, 80.0% ± 10.1%, and 96.4% ± 1.9% for the user's accuracy, producer's accuracy, and overall accuracy, respectively. In the future, small-scale commission errors in the resultant maps should be improved by using more sophisticated machine-learning approaches and considering the reforestation effects in the algorithm. The application of the algorithm to other landscapes with other sensor combinations is also desirable.

show abstract

“…ML has become a common method in remote sensing image processing, with algorithms having high reported accuracies and high computational efficiency when working with large datasets [22,[24][25][26]. For a comprehensive review of ML approaches, see [27][28][29][30][31]. Recent studies have explored the potential of object-based ML in simple systems, such as agricultural monocultures [32].…”

Section: For Vhr and Complex Urban Areasmentioning

confidence: 99%

Mapping Informal Settlements in the Middle East Environment using an Object-Based Machine-Learning Approach

Fallatah¹,

Jones²,

Mitchell³

2018

Preprint

View full text Add to dashboard Cite

The identification of informal settlements in urban areas is an important step in developing and implementing pro-poor urban policies. Understanding when, where and who lives inside informal settlements is critical to efforts to improve their resilience. This study aims to analyse the capability of machine-learning (ML) methods to map informal areas in Jeddah, Saudi Arabia, using very-high-resolution (VHR) imagery and terrain data. Fourteen indicators of settlement characteristics were derived and mapped using an object-based ML approach and VHR imagery. These indicators were categorised according to three different spatial levels: environ, settlement and object. The most useful indicators for prediction were found to be density and texture measures, (with random forest (RF) relative importance measures of over 25% and 23% respectively). The success of this approach was evaluated using a small, fully independent validation dataset. Informal areas were mapped with an overall accuracy of 91%. Object-based ML as a hybrid approach performed better (8%) than object-based image analysis alone due to its ability to encompass all available geospatial levels.

show abstract

Statistical Machine Learning Methods and Remote Sensing for Sustainable Development Goals: A Review

Cited by 203 publications

References 82 publications

Estimating Nitrogen from Structural Crop Traits at Field Scale—A Novel Approach Versus Spectral Vegetation Indices

Estimating Nitrogen from Structural Crop Traits at Field Scale—A Novel Approach Versus Spectral Vegetation Indices

Development of an Operational Algorithm for Automated Deforestation Mapping via the Bayesian Integration of Long-Term Optical and Microwave Satellite Data

Mapping Informal Settlements in the Middle East Environment using an Object-Based Machine-Learning Approach

Contact Info

Product

Resources

About