Rapid and uncontrolled population growth along with economic and industrial development, especially in developing countries during the late twentieth and early twenty-first centuries, have increased the rate of land-use/land-cover (LULC) change many times. Since quantitative assessment of changes in LULC is one of the most efficient means to understand and manage the land transformation, there is a need to examine the accuracy of different algorithms for LULC mapping in order to identify the best classifier for further applications of earth observations. In this article, six machine-learning algorithms, namely random forest (RF), support vector machine (SVM), artificial neural network (ANN), fuzzy adaptive resonance theory-supervised predictive mapping (Fuzzy ARTMAP), spectral angle mapper (SAM) and Mahalanobis distance (MD) were examined. Accuracy assessment was performed by using Kappa coefficient, receiver operational curve (RoC), index-based validation and root mean square error (RMSE). Results of Kappa coefficient show that all the classifiers have a similar accuracy level with minor variation, but the RF algorithm has the highest accuracy of 0.89 and the MD algorithm (parametric classifier) has the least accuracy of 0.82. In addition, the index-based LULC and visual cross-validation show that the RF algorithm (correlations between RF and normalised differentiation water index, normalised differentiation vegetation index and normalised differentiation built-up index are 0.96, 0.99 and 1, respectively, at 0.05 level of significance) has the highest accuracy level in comparison to the other classifiers adopted. Findings from the literature also proved that ANN and RF algorithms are the best LULC classifiers, although a non-parametric classifier like SAM (Kappa coefficient 0.84; area under curve (AUC) 0.85) has a better and consistent accuracy level than the other machine-learning algorithms. Finally, this review concludes that the RF algorithm is the best machine-learning LULC classifier, among the six examined algorithms although it is necessary to further test the RF algorithm in different morphoclimatic conditions in the future.

Land-Use Land-Cover Classification by Machine Learning Classifiers for Satellite Observations—A Review

Abstract This article aims at finding efficient hyperspectral indices for the estimation of forest sun leaf chlorophyll content (CHL, µg cm leaf − 2 ), sun leaf mass per area (LMA, g dry matter m leaf − 2 ), canopy leaf area index (LAI, m 2 leaf m soil − 2 ) and leaf canopy biomass (B leaf , g dry matter m soil − 2 ). These parameters are useful inputs for forest ecosystem simulations at landscape scale. The method is based on the determination of the best vegetation indices (index form and wavelengths) using the radiative transfer model PROSAIL (formed by the newly-calibrated leaf reflectance model PROSPECT coupled with the multi-layer version of the canopy radiative transfer model SAIL). The results are tested on experimental measurements at both leaf and canopy scales. At the leaf scale, it is possible to estimate CHL with high precision using a two wavelength vegetation index after a simulation based calibration. At the leaf scale, the LMA is more difficult to estimate with indices. At the canopy scale, efficient indices were determined on a generic simulated database to estimate CHL, LMA, LAI and Bleaf in a general way. These indices were then applied to two Hyperion images (50 plots) on the Fontainebleau and Fougeres forests and portable spectroradiometer measurements. They showed good results with an RMSE of 8.2 µg cm − 2 for CHL, 9.1 g m − 2 for LMA, 1.7 m 2 m − 2 for LAI and 50.6 g m − 2 for Bleaf. However, at the canopy scale, even if the wavelengths of the calibrated indices were accurately determined with the simulated database, the regressions between the indices and the biophysical characteristics still had to be calibrated on measurements. At the canopy scale, the best indices were: for leaf chlorophyll content: ND chl  = ( ρ 925  −  ρ 710 )/( ρ 925  +  ρ 710 ), for leaf mass per area: ND LMA  = ( ρ 2260  −  ρ 1490 )/( ρ 2260  +  ρ 1490 ), for leaf area index: D LAI  =  ρ 1725  −  ρ 970 , and for canopy leaf biomass: ND Bleaf  = ( ρ 2160  −  ρ 1540 )/( ρ 2160  +  ρ 1540 ).

Calibration and validation of hyperspectral indices for the estimation of broadleaved forest leaf chlorophyll content, leaf mass per area, leaf area index and leaf canopy biomass

Remote sensing is a useful tool for monitoring spatio-temporal variations of crop morphological and physiological status and supporting practices in precision farming. In comparison with multispectral imaging, hyperspectral imaging is a more advanced technique that is capable of acquiring a detailed spectral response of target features. Due to limited accessibility outside of the scientific community, hyperspectral images have not been widely used in precision agriculture. In recent years, different mini-sized and low-cost airborne hyperspectral sensors (e.g., Headwall Micro-Hyperspec, Cubert UHD 185-Firefly) have been developed, and advanced spaceborne hyperspectral sensors have also been or will be launched (e.g., PRISMA, DESIS, EnMAP, HyspIRI). Hyperspectral imaging is becoming more widely available to agricultural applications. Meanwhile, the acquisition, processing, and analysis of hyperspectral imagery still remain a challenging research topic (e.g., large data volume, high data dimensionality, and complex information analysis). It is hence beneficial to conduct a thorough and in-depth review of the hyperspectral imaging technology (e.g., different platforms and sensors), methods available for processing and analyzing hyperspectral information, and recent advances of hyperspectral imaging in agricultural applications. Publications over the past 30 years in hyperspectral imaging technology and applications in agriculture were thus reviewed. The imaging platforms and sensors, together with analytic methods used in the literature, were discussed. Performances of hyperspectral imaging for different applications (e.g., crop biophysical and biochemical properties’ mapping, soil characteristics, and crop classification) were also evaluated. This review is intended to assist agricultural researchers and practitioners to better understand the strengths and limitations of hyperspectral imaging to agricultural applications and promote the adoption of this valuable technology. Recommendations for future hyperspectral imaging research for precision agriculture are also presented.

Recent Advances of Hyperspectral Imaging Technology and Applications in Agriculture

Zoning eco-environmental vulnerability for environmental management and protection

Object-based image analysis (OBIA) has gained widespread popularity for creating maps from remotely sensed data. Researchers routinely claim that OBIA procedures outperform pixel-based procedures; however, it is not immediately obvious how to evaluate the degree to which an OBIA map compares to reference information in a manner that accounts for the fact that the OBIA map consists of objects that vary in size and shape. Our study reviews 209 journal articles concerning OBIA published between 2003 and 2017. We focus on the three stages of accuracy assessment: (1) sampling design, (2) response design and (3) accuracy analysis. First, we report the literature’s overall characteristics concerning OBIA accuracy assessment. Simple random sampling was the most used method among probability sampling strategies, slightly more than stratified sampling. Office interpreted remotely sensed data was the dominant reference source. The literature reported accuracies ranging from 42% to 96%, with an average of 85%. A third of the articles failed to give sufficient information concerning accuracy methodology such as sampling scheme and sample size. We found few studies that focused specifically on the accuracy of the segmentation. Second, we identify a recent increase of OBIA articles in using per-polygon approaches compared to per-pixel approaches for accuracy assessment. We clarify the impacts of the per-pixel versus the per-polygon approaches respectively on sampling, response design and accuracy analysis. Our review defines the technical and methodological needs in the current per-polygon approaches, such as polygon-based sampling, analysis of mixed polygons, matching of mapped with reference polygons and assessment of segmentation accuracy. Our review summarizes and discusses the current issues in object-based accuracy assessment to provide guidance for improved accuracy assessments for OBIA.

https://par.nsf.gov/servlets/purl/10064148

A review of accuracy assessment for object-based image analysis: From per-pixel to per-polygon approaches

Cambodia is one of the most flood-prone countries in Southeast Asia. It is geographically situated in the downstream region of the Mekong River with a lowland floodplain in the middle, surrounded by plateaus and high mountains. It usually experiences devastating floods induced by an overwhelming concentration of rainfall water over the Tonle Sap Lake’s and Mekong River’s banks during monsoon seasons. Flood damage assessment in the rice ecosystem plays an important role in this region as local residents rely heavily on agricultural production. This study introduced an object-based approach to flood mapping and affected rice field estimation in central Cambodia. In this approach, image segmentation processing was conducted with optimal scale parameter estimation based on the variation of objects’ local variances. The inundated area was identified by using Landsat 8 images with an overall accuracy of higher than 95% compared to those derived from finer spatial resolution images. Moderate Resolution Imaging Spectroradiometer (MODIS) vegetation index products were utilized to identify the paddy rice field based on seasonal inter-variation between vegetation and water index during the transplanting stage. The rice classification result was well correlated with the statistical data at a commune level (R2 = 0.675). The flood mapping and affected rice estimation results are useful to provide local governments with valuable information for flooding mitigation and post-flooding compensation and restoration.

Object-based flood mapping and affected rice field estimation with landsat 8 OLI and MODIS data

Multispectral and hyperspectral data have been used to investigate various land cover characteristics. Hyperspectral data have more potential to retrieve information of ground features than multispectral data; however, their limited availability leads to fewer studies in the literature. This research aims to acquire both multispectral and hyperspectral images and compare their performance for estimating vegetation properties (i.e., chlorophyll content). A hyperspectral image (with 325 bands in the visible near infrared (NIR) range) was obtained using a compact hyperspectral sensor mounted on a manned helicopter. A modified camera-based three-band image (with blue, green, and NIR) and a RedEdge sensor-based five-band image (with blue, green, red, red edge, and NIR) were simulated using the hyperspectral image. These three images were compared for the estimation of vegetation chlorophyll content. Partial least square (PLS) regression and random forest regression (RFR) were both applied to estimate chlorophyll using image-derived variables, including vegetation indices and imagery textures. Results show that the RedEdge image achieved good accuracy (R2 ∼ 0.80, RMSE ∼ 14 μg/cm2), close to the accuracy of using the hyperspectral image (R2 ∼ 0.81, RMSE ∼ 13 μg/cm2). The extra bands in the hyperspectral image did not substantially improve chlorophyll estimation. The three-band multispectral image yielded the lowest accuracy (e.g., R2 ∼ 0.42, RMSE ∼ 24 μg/cm2). The RFR performed consistently better than the PLS, owing to its use of randomly-selected training data and predictor variables to build regression trees. These results are expected to provide insights into future studies on the selection of remote sensing images for different monitoring needs.

Comparing the Performance of Multispectral and Hyperspectral Images for Estimating Vegetation Properties

Geothermal energy is an increasingly important component of green energy in the globe. A prerequisite for geothermal energy development is to acquire the local and regional geothermal prospects. Existing geophysical methods of estimating the geothermal potential are usually limited to the scope of prospecting because of the operation cost and site reachability in the field. Thus, explorations in a large-scale area such as the surface temperature and the thermal anomaly primarily rely on satellite thermal infrared imagery. This study aims to apply and integrate thermal infrared (TIR) remote sensing technology with existing geophysical methods for the geothermal exploration in Taiwan. Landsat 7 (L7) Enhanced Thematic Mapper Plus (ETM+) imagery is used to retrieve the land surface temperature (LST) in Ilan plain. Accuracy assessment of satellite-derived LST is conducted by comparing with the air temperature data from 11 permanent meteorological stations. The correlation coefficient of linear regression between air temperature and LST retrieval is 0.76. The MODIS LST product is used for the cross validation of Landsat derived LSTs. Furthermore, Landsat ETM+ multi-temporal brightness temperature imagery for the verification of the LST anomaly results were performed. LST Results indicate that thermal anomaly areas appear correlating with the development of faulted structure. Selected geothermal anomaly areas are validated in detail by field investigation of hot springs and geothermal drillings. It implies that occurrences of hot springs and geothermal drillings are in good spatial agreement with anomaly areas. In addition, the significant low-resistivity zones observed in the resistivity sections are echoed with the LST profiles when compared with in the Chingshui geothermal field. Despite limited to detecting the surficial and the shallow buried geothermal resources, this work suggests that TIR remote sensing is a valuable tool by providing an effective way of mapping and quantifying surface features to facilitate the exploration and assessment of geothermal resources in Taiwan.

/pdf/geothermal-anomaly-mapping-using-landsat-etm-data-in-ilan-3ig361p54x.pdf

Geothermal Anomaly Mapping Using Landsat ETM+ Data in Ilan Plain, Northeastern Taiwan

Interferometric synthetic aperture radar (InSAR), one of the most commonly used remote sensing methods for observing and monitoring land subsidence, has been applied in Hanoi, Vietnam in several st...

Phuong D. Dao

Papers

Recent Advances of Hyperspectral Imaging Technology and Applications in Agriculture

Object-based flood mapping and affected rice field estimation with landsat 8 OLI and MODIS data

Comparing the Performance of Multispectral and Hyperspectral Images for Estimating Vegetation Properties

Geothermal Anomaly Mapping Using Landsat ETM+ Data in Ilan Plain, Northeastern Taiwan

Recent land deformation detected by Sentinel-1A InSAR data (2016–2020) over Hanoi, Vietnam, and the relationship with groundwater level change