Spatial application of Random Forest models for fine-scale coastal vegetation classification using object based analysis of aerial orthophoto and DEM data

High spatial resolution mapping of natural resources is much needed for monitoring and management of species, habitats and landscapes. Generally, detailed surveillance has been conducted as fieldwork, numerical analysis of satellite images or manual interpretation of aerial images, but methods of object-based image analysis (OBIA) and machine learning have recently produced promising examples of automated classifications of aerial imagery. The spatial application potential of such models is however still questionable since the transferability has rarely been evaluated.We investigated the potential of mosaic aerial orthophoto red, green and blue (RGB)/near infrared (NIR) imagery and digital elevation model (DEM) data for mapping very fine-scale vegetation structure in semi-natural terrestrial coastal areas in Denmark. The Random Forest (RF) algorithm, with a wide range of object-derived image and DEM variables, was applied for classification of vegetation structure types using two hierarchical levels of complexity. Models were constructed and validated by cross-validation using three scenarios: (1) training and validation data without spatial separation, (2) training and validation data spatially separated within sites, and (3) training and validation data spatially separated between different sites.Without spatial separation of training and validation data, high classification accuracies of coastal structures of 92.1% and 91.8% were achieved on coarse and fine thematic levels, respectively. When models were applied to spatially separated observations within sites classification accuracies dropped to 85.8% accuracy at the coarse thematic level, and 81.9% at the fine thematic level. When the models were applied to observations from other sites than those trained upon the ability to discriminate vegetation structures was low, with 69.0% and 54.2% accuracy at the coarse and fine thematic levels, respectively.Evaluating classification models with different degrees of spatial correlation between training and validation data was shown to give highly different prediction accuracies, thereby highlighting model transferability and application potential. Aerial image and DEM-based RF models had low transferability to new areas due to lack of representation of aerial image, landscape and vegetation variation in training data. They do, however, show promise at local scale for supporting conservation and management with vegetation mappings of high spatial and thematic detail based on low-cost image data.     

An easy and low-cost method for preprocessing and matching small-scale amateur aerial photography for assessing agricultural land use in Burkina Faso

In recent decades, the Kou watershed in south-western Burkina Faso has suffered from poor water management. Despite the abundance of water, most water users regularly face water shortages because of the increase in the amount of land under irrigation. To help them achieve a more equitable allocation of irrigated land, local stakeholders need an easily managed low-cost tool for monitoring and mapping these irrigated zones. The aim of this study was to develop a fast and low-cost procedure for mosaicing and geo referencing amateur small-scale aerial photographs for land-use surveys. Sets of tens (2009) and hundreds (2007) of low-altitude aerial photographs, with a resolution of 0.4 m and 0.8 m, respectively, were used to create a detailed land-cover map of typical African small-scale irrigated agriculture. A commercially available stitching tool and GIS allowed geo referenced 'mono-images’ to be constructed; both mosaics were warped on a high-resolution SPOT image with a horizontal root mean square error (RMSE) of about 11 m. The RMSE between the two image datasets was 2 m. This approach is less sensitive to atmospheric conditions that are non-predictable in programming satellite imagery.     

Assessment of Nigeriasat-1 satellite data for urban land use/land cover analysis using object-based image analysis in Abuja,Nigeria

This study assesses the usefulness of Nigeriasat-1 satellite data for urban land cover analysis by comparing it with Landsat and SPOT data. The data-sets for Abuja were classified with pixel- and object-based methods. While the pixel-based method was classified with the spectral properties of the images, the object-based approach included an extra layer of land use cadastre data. The classification accuracy results for OBIA show that Landsat 7 ETM, Nigeriasat-1 SLIM and SPOT 5 HRG had overall accuracies of 92, 89 and 96%, respectively, while the classification accuracy for pixel-based classification were 88% for Landsat 7 ETM, 63% for Nigeriasat-1 SLIM and 89% for SPOT 5 HRG. The results indicate that given the right classification tools, the analysis of Nigeriasat-1 data can be compared with Landsat and SPOT data which are widely used for urban land use and land cover analysis.     

Empirical modeling and spatio-temporal patterns of urban evapotranspiration for the Phoenix metropolitan area,Arizona

In this study, an empirical model for predicting urban evapotranspiration (ET) is examined for the Phoenix metropolitan area that is in a subtropical desert climate using in situ ET measurements from a local flux tower and remotely sensed moderate-resolution imaging spectroradiometer land products. Annual ET maps of Phoenix are then created for the period from 2001 to 2015 using the empirical model developed. A time-series trend analysis is finally performed using predicted ET maps to discover the spatio-temporal patterns of ET changes during the study period. Results suggest that blue-sky albedo and land surface temperature are two statistically significant variables explanatory to model urban ET for Phoenix. Areas that have experienced significant increases of ET are highly spatially clustered, and are mainly found on the outskirts of the city, while areas of decreasing ET are generally associated with highly developed areas, such as downtown Phoenix.     

Agricultural cropland mapping using black-and-white aerial photography,Object-Based Image Analysis and Random Forests

Land-use and land-cover (LULC) conversions have an important impact on land degradation, erosion and water availability. Information on historical land cover (change) is crucial for studying and modelling land- and ecosystem degradation. During the past decades major LULC conversions occurred in Africa, Southeast Asia and South America as a consequence of a growing population and economy. Most distinct is the conversion of natural vegetation into cropland. Historical LULC information can be derived from satellite imagery, but these only date back until approximately 1972. Before the emergence of satellite imagery, landscapes were monitored by black-and-white (B&W) aerial photography. This photography is often visually interpreted, which is a very time-consuming approach. This study presents an innovative, semi-automated method to map cropland acreage from B&W photography. Cropland acreage was mapped on two study sites in Ethiopia and in The Netherlands. For this purpose we used Geographic Object-Based Image Analysis (GEOBIA) and a Random Forest classification on a set of variables comprising texture, shape, slope, neighbour and spectral information. Overall mapping accuracies attained are 90% and 96% for the two study areas respectively. This mapping method increases the timeline at which historical cropland expansion can be mapped purely from brightness information in B&W photography up to the 1930s, which is beneficial for regions where historical land-use statistics are mostly absent.     

Classifying a high resolution image of an urban area using super-object information

In this study, a multi-scale approach was used for classifying land cover in a high resolution image of an urban area. Pixels and image segments were assigned the spectral, texture, size, and shape information of their super-objects (i.e. the segments that they are located within) from coarser segmentations of the same scene, and this set of super-object information was used as additional input data for image classification. The accuracies of classifications that included super-object variables were compared with the classification accuracies of image segmentations that did not include super-object information. The highest overall accuracy and kappa coefficient achieved without super-object information was 78.11% and 0.727%, respectively. When single pixels or fine-scale image segments were assigned the statistics of their super-objects prior to classification, overall accuracy increased to 84.42% and the kappa coefficient increased to 0.804.     

Methods for improving accuracy and extending results beyond periods covered by traditional ground-truth in remote sensing classification of a complex landscape

Successful development of approaches to quantify impacts of diverse landuse and associated agricultural management practices on ecosystem services is frequently limited by lack of historical and contemporary landuse data. We hypothesized that ground truth data from one year could be used to extrapolate previous or future landuse in a complex landscape where cropping systems do not generally change greatly from year to year because the majority of crops are established perennials or the same annual crops grown on the same fields over multiple years. Prior to testing this hypothesis, it was first necessary to classify 57 major landuses in the Willamette Valley of western Oregon from 2005 to 2011 using normal same year ground-truth, elaborating on previously published work and traditional sources such as Cropland Data Layers (CDL) to more fully include minor crops grown in the region. Available remote sensing data included Landsat, MODIS 16-day composites, and National Aerial Imagery Program (NAIP) imagery, all of which were resampled to a common 30 m resolution. The frequent presence of clouds and Landsat7 scan line gaps forced us to conduct of series of separate classifications in each year, which were then merged by choosing whichever classification used the highest number of cloud- and gap-free bands at any given pixel. Procedures adopted to improve accuracy beyond that achieved by maximum likelihood pixel classification included majority-rule reclassification of pixels within 91,442 Common Land Unit (CLU) polygons, smoothing and aggregation of areas outside the CLU polygons, and majority-rule reclassification over time of forest and urban development areas. Final classifications in all seven years separated annually disturbed agriculture, established perennial crops, forest, and urban development from each other at 90 to 95% overall 4-class validation accuracy. In the most successful use of subsequent year ground-truth data to classify prior year landuse, an overall 57-class accuracy of 75% was achieved despite the omission of 10 entire classes, most of which were annually disturbed or perennial crops grown on very few fields. Synthetic ground-truth data for the 2004 harvest year based on the most common landuse classes over the following 7 years classified 49 of 57 categories at an overall accuracy of 96% in a final version that included CLU polygon majority rule, default smoothing and aggregation, and forcing of urban development and forest from multi-year majority-rule.     

A multi-level context-guided classification method with object-based convolutional neural network for land cover classification using very high resolution remote sensing images

Classification of very high resolution imagery (VHRI) is challenging due to the difficulty in mining complex spatial and spectral patterns from rich image details. Various object-based Convolutional Neural Networks (OCNN) for VHRI classification have been proposed to overcome the drawbacks of the redundant pixel-wise CNNs, owing to their low computational cost and fine contour-preserving. However, classification performance of OCNN is still limited by geometric distortions, insufficient feature representation, and lack of contextual guidance. In this paper, an innovative multi-level context-guided classification method with the OCNN (MLCG-OCNN) is proposed. A feature-fusing OCNN, including the object contour-preserving mask strategy with the supplement of object deformation coefficient, is developed for accurate object discrimination by learning simultaneously high-level features from independent spectral patterns, geometric characteristics, and object-level contextual information. Then pixel-level contextual guidance is used to further improve the per-object classification results. The MLCG-OCNN method is intentionally tested on two validated small image datasets with limited training samples, to assess the performance in applications of land cover classification where a trade-off between time-consumption of sample training and overall accuracy needs to be found, as it is very common in the practice. Compared with traditional benchmark methods including the patch-based per-pixel CNN (PBPP), the patch-based per-object CNN (PBPO), the pixel-wise CNN with object segmentation refinement (PO), semantic segmentation U-Net (U-NET), and DeepLabV3+(DLV3+), MLCG-OCNN method achieves remarkable classification performance (> 80 %). Compared with the state-of-the-art architecture DeepLabV3+, the MLCG-OCNN method demonstrates high computational efficiency for VHRI classification (4–5 times faster).     

Optimizing support vector machine learning for semi-arid vegetation mapping by using clustering analysis

In remote sensing communities, support vector machine (SVM) learning has recently received increasing attention. SVM learning usually requires large memory and enormous amounts of computation time on large training sets. According to SVM algorithms, the SVM classification decision function is fully determined by support vectors, which compose a subset of the training sets. In this regard, a solution to optimize SVM learning is to efficiently reduce training sets. In this paper, a data reduction method based on agglomerative hierarchical clustering is proposed to obtain smaller training sets for SVM learning. Using a multiple angle remote sensing dataset of a semi-arid region, the effectiveness of the proposed method is evaluated by classification experiments with a series of reduced training sets. The experiments show that there is no loss of SVM accuracy when the original training set is reduced to 34% using the proposed approach. Maximum likelihood classification (MLC) also is applied on the reduced training sets. The results show that MLC can also maintain the classification accuracy. This implies that the most informative data instances can be retained by this approach.     

Classifiers vs. input variables—The drivers in image classification for land cover mapping

The study investigates the performance of image classifiers for landscape-scale land cover mapping and the relevance of ancillary data for the classification success in order to assess and to quantify the importance of these components in image classification. Specifically tested are the performance of maximum likelihood classification (MLC), artificial neural networks (ANN) and discriminant analysis (DA) based on Landsat7 ETM+ spectral data in combination with topographic measures and NDVI. ANN produced high accuracies of more than 75% also with limited input information, while MLC and DA produced comparable results only by incorporating ancillary data into the classification process. The superiority of ANN classification was less pronounced on the level of the single land cover classes. The use of ancillary data generally increased classification accuracy and showed a similar potential for increasing classification accuracy than the selection of the classifier. Therefore, a stronger focus on the development of appropriate and optimised sets of input variables is suggested. Also the definition and selection of land cover classes has shown to be crucial and not to be simply adaptable from existing land cover class schemes. A stronger research focus towards discriminating land cover classes by their typical spectral, topographic or seasonal properties is therefore suggested to advance image classification.     

Object-based urban land cover classification using rule inheritance over very high-resolution multisensor and multitemporal data

Very high spatial and temporal resolution remote sensing data facilitate mapping highly complex and diverse urban environments. This study analyzed and demonstrated the usefulness of combined high-resolution aerial digital images and elevation data, and its processing using object-based image analysis for mapping urban land covers and quantifying buildings. It is observed that mapping heterogeneous features across large urban areas is time consuming and challenging. This study presents and demonstrates an approach for formulating an optimal land cover classification rule set over small representative training urban area image, and its subsequent transfer to the multisensor, multitemporal images. The classification results over the training area showed an overall accuracy of 96%, and the application of rule set to different sensor images of other test areas resulted in reduced accuracies of 91% for the same sensor, 90% and 86% for the different sensors temporal data. The comparison of reference and classified buildings showed ±4% detection errors. Classification through a transferred rule set reduced the classification accuracy by about 5%–10%. However, the trade-off for this accuracy drop was about a 75% reduction in processing time for performing classification in the training area. The factors influencing the classification accuracies were mainly the shadow and temporal changes in the class characteristics.     

大载荷长航时无人机航空应急测绘系统设计与实现

无人机航空摄影是应急测绘保障的重要手段。近年来,大载荷长航时无人机已逐步进入民用领域。以往关于航空应急测绘系统的研究主要是针对轻小型无人机和有人机,缺少适用于大载荷长航时无人机性能和特点的系统设计方法。本文首先分析了不同固定翼无人机产品的性能差异,从无人机平台、任务传感器、遥感数据处理软件等的集成关系入手,提出了基于航测控制器的多载荷一体化集成的设计方案,以及涵盖航摄规划、视频实时处理、影像高精度处理的软件系统集成方案。采用该设计方案研制实现了CH-4航空应急测绘系统,并在四川广元某区域进行了试验,结果表明该系统不仅可以同时获取视频、光学影像和SAR数据,而且数据获取质量能够满足大比例尺测图的要求。     

Land cover classification using Landsat 8 Operational Land Imager data in Beijing,China

The successful launch of Landsat 8 provides a new data source for monitoring land cover, which has the potential to significantly improve the characterization of the earth’s surface. To assess data performance, Landsat 8 Operational Land Imager (OLI) data were first compared with Landsat 7 ETM + data using texture features as the indicators. Furthermore, the OLI data were investigated for land cover classification using the maximum likelihood and support vector machine classifiers in Beijing. The results indicated that (1) the OLI data quality was slightly better than the ETM + data quality in the visible bands, especially the near-infrared band of OLI the data, which had a clear improvement; clear improvement was not founded in the shortwave-infrared bands. Moreover, (2) OLI data had a satisfactory performance in terms of land cover classification. In summary, OLI data were a reliable data source for monitoring land cover and provided the continuity in the Landsat earth observation.     

Fusion of HJ1B and ALOS PALSAR data for land cover classification using machine learning methods

Image classification from remote sensing is becoming increasingly urgent for monitoring environmental changes. Exploring effective algorithms to increase classification accuracy is critical. This paper explores the use of multispectral HJ1B and ALOS (Advanced Land Observing Satellite) PALSAR L-band (Phased Array type L-band Synthetic Aperture Radar) for land cover classification using learning-based algorithms. Pixel-based and object-based image analysis approaches for classifying HJ1B data and the HJ1B and ALOS/PALSAR fused-images were compared using two machine learning algorithms, support vector machine (SVM) and random forest (RF), to test which algorithm can achieve the best classification accuracy in arid and semiarid regions. The overall accuracies of the pixel-based (Fused data: 79.0%; HJ1B data: 81.46%) and object-based classifications (Fused data: 80.0%; HJ1B data: 76.9%) were relatively close when using the SVM classifier. The pixel-based classification achieved a high overall accuracy (85.5%) using the RF algorithm for classifying the fused data, whereas the RF classifier using the object-based image analysis produced a lower overall accuracy (70.2%). The study demonstrates that the pixel-based classification utilized fewer variables and performed relatively better than the object-based classification using HJ1B imagery and the fused data. Generally, the integration of the HJ1B and ALOS/PALSAR imagery can improve the overall accuracy of 5.7% using the pixel-based image analysis and RF classifier.     

Relevance of airborne lidar and multispectral image data for urban scene classification using Random Forests

Airborne lidar systems have become a source for the acquisition of elevation data. They provide georeferenced, irregularly distributed 3D point clouds of high altimetric accuracy. Moreover, these systems can provide for a single laser pulse, multiple returns or echoes, which correspond to different illuminated objects. In addition to multi-echo laser scanners, full-waveform systems are able to record 1D signals representing a train of echoes caused by reflections at different targets. These systems provide more information about the structure and the physical characteristics of the targets. Many approaches have been developed, for urban mapping, based on aerial lidar solely or combined with multispectral image data. However, they have not assessed the importance of input features. In this paper, we focus on a multi-source framework using aerial lidar (multi-echo and full waveform) and aerial multispectral image data. We aim to study the feature relevance for dense urban scenes. The Random Forests algorithm is chosen as a classifier: it runs efficiently on large datasets, and provides measures of feature importance for each class. The margin theory is used as a confidence measure of the classifier, and to confirm the relevance of input features for urban classification. The quantitative results confirm the importance of the joint use of optical multispectral and lidar data. Moreover, the relevance of full-waveform lidar features is demonstrated for building and vegetation area discrimination.     

Modelling high arctic percent vegetation cover using field digital images and high resolution satellite data

In this study, digital images collected at a study site in the Canadian High Arctic were processed and classified to examine the spatial-temporal patterns of percent vegetation cover (PVC). To obtain the PVC of different plant functional groups (i.e., forbs, graminoids/sedges and mosses), field near infrared-green-blue (NGB) digital images were classified using an object-based image analysis (OBIA) approach. The PVC analyses comparing different vegetation types confirmed: (i) the polar semi-desert exhibited the lowest PVC with a large proportion of bare soil/rock cover; (ii) the mesic tundra cover consisted of approximately 60% mosses; and (iii) the wet sedge consisted almost exclusively of graminoids and sedges. As expected, the PVC and green normalized difference vegetation index (GNDVI; (R_NIR − R_Green)/(R_NIR + R_Green)), derived from field NGB digital images, increased during the summer growing season for each vegetation type: i.e., ∼5% (0.01) for polar semi-desert; ∼10% (0.04) for mesic tundra; and ∼12% (0.03) for wet sedge respectively. PVC derived from field images was found to be strongly correlated with WorldView-2 derived normalized difference spectral indices (NDSI; (R_x − R_y)/(R_x + R_y)), where R_x is the reflectance of the red edge (724.1 nm) or near infrared (832.9 nm and 949.3 nm) bands; R_y is the reflectance of the yellow (607.7 nm) or red (658.8 nm) bands with R²’s ranging from 0.74 to 0.81. NDSIs that incorporated the yellow band (607.7 nm) performed slightly better than the NDSIs without, indicating that this band may be more useful for investigating Arctic vegetation that often includes large proportions of senescent vegetation throughout the growing season.     

A method for monitoring building construction in urban sprawl areas using object-based analysis of Spot 5 images and existing GIS data

Urban sprawl has been identified as one of the most negative effects of global population growth on the environment and biodiversity. Frequent monitoring of urban sprawl is needed to limit the impact of this ongoing phenomenon. This paper proposes precise monitoring of building construction using an object-based classification methodology applied to Spot 5 images with a 2.5 m resolution. An application at a regional scale on Reunion Island in the Indian Ocean shows that this building extraction methodology has limitations in the production of reference urban maps because of difficulties in defining the shape and the number of buildings compared to classical photo-interpretation of aerial photography. However, these results are of great value for planning in urban sprawl areas where up-to-date information is lacking because of the rapid pace of house construction and residential development.     

A sub-pixel analysis of urbanization effect on land surface temperature and its interplay with impervious surface and vegetation coverage in Indianapolis,United States

This study developed an analytical procedure based upon a spectral unmixing model for characterizing and quantifying urban landscape changes in Indianapolis, Indiana, the United States, and for examining the environmental impact of such changes on land surface temperatures (LST). Three dates of Landsat TM/ETM+ images, acquired in 1991, 1995, and 2000, respectively, were utilized to document the historical morphological changes in impervious surface and vegetation coverage and to analyze the relationship between these changes and those occurred in LST. Three fraction endmembers, i.e., impervious surface, green vegetation, and shade, were derived with an unconstrained least-squares solution. A hybrid classification procedure, which combined maximum-likelihood and decision-tree algorithms, was developed to classify the fraction images into land use and land cover classes. Correlation analyses were conducted to investigate the changing relationships of LST with impervious surface and vegetation coverage. Results indicate that multi-temporal fraction images were effective for quantifying the dynamics of urban morphology and for deriving a reliable measurement of environmental variables such as vegetation abundance and impervious surface coverage. Urbanization created an evolved inverse relationship between impervious and vegetation coverage, and brought about new LST patterns because of LST's correlations with both impervious and vegetation coverage. Further researches should be directed to refine spectral mixture modeling by stratification, and by the use of multiple endmembers and hyperspectral imagery.     

Reducing landscape heterogeneity for improved land use and land cover (LULC) classification across the large and complex Ethiopian highlands

This paper presents a land use and land cover (LULC) classification approach that accounts landscape heterogeneity. We addressed this challenge by subdividing the study area into more homogeneous segments using several biophysical and socio-economic factors as well as spectral information. This was followed by unsupervised clustering within each homogeneous segment and supervised class assignment. Two classification schemes differing in their level of detail were successfully applied to four landscape types of distinct LULC composition. The resulting LULC map fulfills two major requirements: (1) differentiation and identification of several LULC classes that are of interest at the local, regional, and national scales, and (2) high accuracy of classification. The approach overcomes commonly encountered difficulties of classifying second-level classes in large and heterogeneous landscapes. The output of the study responds to the need for comprehensive LULC data to support ecosystem assessment, policy formulation, and decision-making towards sustainable land resources management.     

Changing landscape in the Three Gorges Reservoir Area of Yangtze River from 1977 to 2005: Land use/land cover,vegetation cover changes estimated using multi-source satellite data

The eco-environment in the Three Gorges Reservoir Area (TGRA) in China has received much attention due to the construction of the Three Gorges Hydropower Station. Land use/land cover changes (LUCC) are a major cause of ecological environmental changes. In this paper, the spatial landscape dynamics from 1978 to 2005 in this area are monitored and recent changes are analyzed, using the Landsat TM (MSS) images of 1978, 1988, 1995, 2000 and 2005. Vegetation cover fractions for a vegetation cover analysis are retrieved from MODIS/Terra imagery from 2000 to 2006, being the period before and after the rising water level of the reservoir. Several analytical indices have been used to analyze spatial and temporal changes. Results indicate that cropland, woodland, and grassland areas reduced continuously over the past 30 years, while river and built-up area increased by 2.79% and 4.45% from 2000 to 2005, respectively. The built-up area increased at the cost of decreased cropland, woodland and grassland. The vegetation cover fraction increased slightly. We conclude that significant changes in land use/land cover have occurred in the Three Gorges Reservoir Area. The main cause is a continuous economic and urban/rural development, followed by environmental management policies after construction of the Three Gorges Dam.