In previous classification studies, three non-parametric classifiers, Random Forest (RF), k-Nearest Neighbor (kNN), and Support Vector Machine (SVM), were reported as the foremost classifiers at producing high accuracies. However, only a few studies have compared the performances of these classifiers with different training sample sizes for the same remote sensing images, particularly the Sentinel-2 Multispectral Imager (MSI). In this study, we examined and compared the performances of the RF, kNN, and SVM classifiers for land use/cover classification using Sentinel-2 image data. An area of 30 × 30 km2 within the Red River Delta of Vietnam with six land use/cover types was classified using 14 different training sample sizes, including balanced and imbalanced, from 50 to over 1250 pixels/class. All classification results showed a high overall accuracy (OA) ranging from 90% to 95%. Among the three classifiers and 14 sub-datasets, SVM produced the highest OA with the least sensitivity to the training sample sizes, followed consecutively by RF and kNN. In relation to the sample size, all three classifiers showed a similar and high OA (over 93.85%) when the training sample size was large enough, i.e., greater than 750 pixels/class or representing an area of approximately 0.25% of the total study area. The high accuracy was achieved with both imbalanced and balanced datasets.

Comparison of Random Forest, k-Nearest Neighbor, and Support Vector Machine Classifiers for Land Cover Classification Using Sentinel-2 Imagery

Remotely Piloted Aircraft (RPA) is presently in continuous development at a rapid pace. Unmanned Aerial Vehicles (UAVs) or more extensively Unmanned Aerial Systems (UAS) are platforms considered under the RPAs paradigm. Simultaneously, the development of sensors and instruments to be installed onboard such platforms is growing exponentially. These two factors together have led to the increasing use of these platforms and sensors for remote sensing applications with new potential. Thus, the overall goal of this paper is to provide a panoramic overview about the current status of remote sensing applications based on unmanned aerial platforms equipped with a set of specific sensors and instruments. First, some examples of typical platforms used in remote sensing are provided. Second, a description of sensors and technologies is explored which are onboard instruments specifically intended to capture data for remote sensing applications. Third, multi-UAVs in collaboration, coordination, and cooperation in remote sensing are considered. Finally, a collection of applications in several areas are proposed, where the combination of unmanned platforms and sensors, together with methods, algorithms, and procedures provide the overview in very different remote sensing applications. This paper presents an overview of different areas, each independent from the others, so that the reader does not need to read the full paper when a specific application is of interest

Overview and Current Status of Remote Sensing Applications Based on Unmanned Aerial Vehicles (UAVs)

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

Wireless Sensor Networks and Multi-UAV systems for natural disaster management

In just the past five years, the field of Earth observation has progressed beyond the offerings of conventional space agency based platforms to include a plethora of sensing opportunities afforded by CubeSats, Unmanned Aerial Vehicles (UAVs), and smartphone technologies that are being embraced by both for-profit companies and individual researchers. Over the previous decades, space agency efforts have brought forth well-known and immensely useful satellites such as the Landsat series and the Gravity Research and Climate Experiment (GRACE) system, with costs typically on the order of one billion dollars per satellite and with concept-to-launch timelines on the order of two decades (for new missions). More recently, the proliferation of smartphones has helped to miniaturise sensors and energy requirements, facilitating advances in the use of CubeSats that can be launched by the dozens, while providing ultra-high (3-5 m) resolution sensing of the Earth on a daily basis. Start-up companies that did not exist five years ago now operate more satellites in orbit than any space agency, and at costs that are a mere fraction of the cost of traditional satellite missions. With these advances come new space-borne measurements, such as real-time high-definition video for tracking air pollution, storm-cell development, flood propagation, precipitation monitoring, or even for constructing digital surfaces using structure-from-motion techniques. Closer to the surface, measurements from small unmanned drones and tethered balloons have mapped snow depths, floods, and estimated evaporation at sub-meter resolutions, pushing back on spatio-temporal constraints and delivering new process insights. At ground level, precipitation has been measured using signal attenuation between antennae mounted on cell phone towers, while the proliferation of mobile devices has enabled citizen-scientists to catalogue photos of environmental conditions, estimate daily average temperatures from battery state, and sense other hydrologically important variables such as channel depths using commercially available wireless devices. Global internet access is being pursued via high altitude balloons, solar planes, and hundreds of planned satellite launches, providing a means to exploit the Internet of Things as an entirely new measurement domain. Such global access will enable real-time collection of data from billions of smartphones or from remote research platforms. This future will produce petabytes of data that can only be accessed via cloud storage and will require new analytical approaches to interpret. The extent to which today's hydrologic models can usefully ingest such massive data volumes is unclear. Nor is it clear whether this deluge of data will be usefully exploited, either because the measurements are superfluous, inconsistent, not accurate enough, or simply because we lack the capacity to process and analyse them. What is apparent is that the tools and techniques afforded by this array of novel and game-changing sensing platforms present our community with a unique opportunity to develop new insights that advance fundamental aspects of the hydrological sciences. To accomplish this will require more than just an application of the technology: in some cases, it will demand a radical rethink on how we utilise and exploit these new observing systems to enhance our understanding of the Earth and its linked processes.

/pdf/the-future-of-earth-observation-in-hydrology-4rd0x66qs3.pdf

The Future of Earth Observation in Hydrology.

Unmanned aerial vehicle (UAV) remote sensing has great potential for vegetation mapping in complex urban landscapes due to the ultra-high resolution imagery acquired at low altitudes. Because of payload capacity restrictions, off-the-shelf digital cameras are widely used on medium and small sized UAVs. The limitation of low spectral resolution in digital cameras for vegetation mapping can be reduced by incorporating texture features and robust classifiers. Random Forest has been widely used in satellite remote sensing applications, but its usage in UAV image classification has not been well documented. The objectives of this paper were to propose a hybrid method using Random Forest and texture analysis to accurately differentiate land covers of urban vegetated areas, and analyze how classification accuracy changes with texture window size. Six least correlated second-order texture measures were calculated at nine different window sizes and added to original Red-Green-Blue (RGB) images as ancillary data. A Random Forest classifier consisting of 200 decision trees was used for classification in the spectral-textural feature space. Results indicated the following: (1) Random Forest outperformed traditional Maximum Likelihood classifier and showed similar performance to object-based image analysis in urban vegetation classification; (2) the inclusion of texture features improved classification accuracy significantly; (3) classification accuracy followed an inverted U relationship with texture window size. The results demonstrate that UAV provides an efficient and ideal platform for urban vegetation mapping. The hybrid method proposed in this paper shows good performance in differentiating urban vegetation mapping. The drawbacks of off-the-shelf digital cameras can be reduced by adopting Random Forest and texture analysis at the same time.

UAV Remote Sensing for Urban Vegetation Mapping Using Random Forest and Texture Analysis

Flooding is a severe natural hazard, which poses a great threat to human life and property, especially in densely-populated urban areas. As one of the fastest developing fields in remote sensing applications, an unmanned aerial vehicle (UAV) can provide high-resolution data with a great potential for fast and accurate detection of inundated areas under complex urban landscapes. In this research, optical imagery was acquired by a mini-UAV to monitor the serious urban waterlogging in Yuyao, China. Texture features derived from gray-level co-occurrence matrix were included to increase the separability of different ground objects. A Random Forest classifier, consisting of 200 decision trees, was used to extract flooded areas in the spectral-textural feature space. Confusion matrix was used to assess the accuracy of the proposed method. Results indicated the following: (1) Random Forest showed good performance in urban flood mapping with an overall accuracy of 87.3% and a Kappa coefficient of 0.746; (2) the inclusion of texture features improved classification accuracy significantly; (3) Random Forest outperformed maximum likelihood and artificial neural network, and showed a similar performance to support vector machine. The results demonstrate that UAV can provide an ideal platform for urban flood monitoring and the proposed method shows great capability for the accurate extraction of inundated areas.

Urban Flood Mapping Based on Unmanned Aerial Vehicle Remote Sensing and Random Forest Classifier—A Case of Yuyao, China

Coastal land cover classification is a significant yet challenging task in remote sensing because of the complex and fragmented nature of coastal landscapes. However, availability of multitemporal and multisensor remote sensing data provides opportunities to improve classification accuracy. Meanwhile, rapid development of deep learning has achieved astonishing results in computer vision tasks and has also been a popular topic in the field of remote sensing. Nevertheless, designing an effective and concise deep learning model for coastal land cover classification remains problematic. To tackle this issue, we propose a multibranch convolutional neural network (MBCNN) for the fusion of multitemporal and multisensor Sentinel data to improve coastal land cover classification accuracy. The proposed model leverages a series of deformable convolutional neural networks to extract representative features from a single-source dataset. Extracted features are aggregated through an adaptive feature fusion module to predict final land cover categories. Experimental results indicate that the proposed MBCNN shows good performance, with an overall accuracy of 93.78% and a Kappa coefficient of 0.9297. Inclusion of multitemporal data improves accuracy by an average of 6.85%, while multisensor data contributes to 3.24% of accuracy increase. Additionally, the featured fusion module in this study also increases accuracy by about 2% when compared with the feature-stacking method. Results demonstrate that the proposed method can effectively mine and fuse multitemporal and multisource Sentinel data, which improves coastal land cover classification accuracy.

Integrating Multitemporal Sentinel-1/2 Data for Coastal Land Cover Classification Using a Multibranch Convolutional Neural Network: A Case of the Yellow River Delta

Remote sensing is recognized as a valuable tool for flood mapping due to its synoptic view and continuous coverage of the flooding event. This paper proposed a hybrid approach based on multiple endmember spectral analysis (MESMA) and Random Forest classifier to extract inundated areas in Yuyao City in China using medium resolution optical imagery. MESMA was adopted to tackle the mixing pixel problem induced by medium resolution data. Specifically, 35 optimal endmembers were selected to construct a total of 3111 models in the MESMA procedure to derive accurate fraction information. A multi-dimensional feature space was constructed including the normalized difference water index (NDWI), topographical parameters of height, slope, and aspect together with the fraction maps. A Random Forest classifier consisting of 200 decision trees was adopted to classify the post-flood image based on the above multi-features. Experimental results indicated that the proposed method can extract the inundated areas precisely with a classification accuracy of 94% and a Kappa index of 0.88. The inclusion of fraction information can help improve the mapping accuracy with an increase of 2.5%. Moreover, the proposed method also outperformed the maximum likelihood classifier and the NDWI thresholding method. This research provided a useful reference for flood mapping using medium resolution optical remote sensing imagery.

/pdf/flood-mapping-based-on-multiple-endmember-spectral-mixture-1c64fgt8a1.pdf

Flood Mapping Based on Multiple Endmember Spectral Mixture Analysis and Random Forest Classifier—The Case of Yuyao, China

Wheat is a major staple food crop in China. Accurate and cost-effective wheat mapping is exceedingly critical for food production management, food security warnings, and food trade policy-making in China. To reduce confusion between wheat and non-wheat crops for accurate growth stage wheat mapping, we present a novel approach that combines a random forest (RF) classifier with multi-sensor and multi-temporal image data. This study aims to (1) determine whether an RF combined with multi-sensor and multi-temporal imagery can achieve accurate winter wheat mapping, (2) to find out whether the proposed approach can provide improved performance over the traditional classifiers, and (3) examine the feasibility of deriving reliable estimates of winter wheat-growing areas from medium-resolution remotely sensed data. Winter wheat mapping experiments were conducted in Boxing County. The experimental results suggest that the proposed method can achieve good performance, with an overall accuracy of 92.9% and a ...

Jiantao Liu

Papers

UAV Remote Sensing for Urban Vegetation Mapping Using Random Forest and Texture Analysis

Urban Flood Mapping Based on Unmanned Aerial Vehicle Remote Sensing and Random Forest Classifier—A Case of Yuyao, China

Integrating Multitemporal Sentinel-1/2 Data for Coastal Land Cover Classification Using a Multibranch Convolutional Neural Network: A Case of the Yellow River Delta

Flood Mapping Based on Multiple Endmember Spectral Mixture Analysis and Random Forest Classifier—The Case of Yuyao, China

Winter wheat mapping using a random forest classifier combined with multi-temporal and multi-sensor data