A novel algorithm for land use and land cover classification using RADARSAT-2 polarimetric SAR data

This study proposes a new four-component algorithm for land use and land cover (LULC) classification using RADARSAT-2 polarimetric SAR (PolSAR) data. These four components are polarimetric decomposition, PolSAR interferometry, object-oriented image analysis, and decision tree algorithms. First, polarimetric decomposition can be used to support the classification of PolSAR data. It is aimed at extracting polarimetric parameters related to the physical scattering mechanisms of the observed objects. Second, PolSAR interferometry is used to extract polarimetric interferometric information to support LULC classification. Third, the main purposes of object-oriented image analysis are delineating image objects, as well as extracting various textural and spatial features from image objects to improve classification accuracy. Finally, a decision tree algorithm provides an efficient way to select features and implement classification. A comparison between the proposed method and the Wishart supervised classification which is based on the coherency matrix was made to test the performance of the proposed method. The overall accuracy of the proposed method was 86.64%, whereas that of the Wishart supervised classification was 69.66%. The kappa value of the proposed method was 0.84, much higher than that of the Wishart supervised classification, which exhibited a kappa value of 0.65. The results indicate that the proposed method exhibits much better performance than the Wishart supervised classification for LULC classification. Further investigation was carried out on the respective contribution of the four components to LULC classification using RADARSAT-2 PolSAR data, and it indicates that all the four components have important contribution to the classification. Polarimetric information has significant implications for identifying different vegetation types and distinguishing between vegetation and urban/built-up. The polarimetric interferometric information extracted from repeat-pass RADARSAT-2 images is important in reducing the confusion between urban/built-up and vegetation and that between barren/sparsely vegetated land and vegetation. Object-oriented image analysis is very helpful in reducing the effect of speckle in PolSAR images by implementing classification based on image objects, and the textural information extracted from image objects is helpful in distinguishing between water and lawn. The decision tree algorithm can achieve higher classification accuracy than the nearest neighbor classification implemented using Definiens Developer 7.0, and the accuracy of the decision tree algorithm is similar with that of the support vector classification which is implemented based on the features selected using genetic algorithms. Compared with the nearest neighbor and support vector classification, the decision tree algorithm is more efficient to select features and implement classification. Furthermore, the decision tree algorithm can provide clear classification rules that can be easily interpreted based on the physical meaning of the features used in the classification. This can provide physical insight for LULC classification using PolSAR data.     

Enhanced decision tree ensembles for land-cover mapping from fully polarimetric SAR data

Fully polarimetric synthetic aperture radar (PolSAR) Earth Observations showed great potential for mapping and monitoring agro-environmental systems. Numerous polarimetric features can be extracted from these complex observations which may lead to improve accuracy of land-cover classification and object characterization. This article employed two well-known decision tree ensembles, i.e. bagged tree (BT) and random forest (RF), for land-cover mapping from PolSAR imagery. Moreover, two fast modified decision tree ensembles were proposed in this article, namely balanced filter-based forest (BFF) and cost-sensitive filter-based forest (CFF). These algorithms, designed based on the idea of RF, use a fast filter feature selection algorithms and two extended majority voting. They are also able to embed some solutions of imbalanced data problem into their structures. Three different PolSAR datasets, with imbalanced data, were used for evaluating efficiency of the proposed algorithms. The results indicated that all the tree ensembles have higher efficiency and reliability than the individual DT. Moreover, both proposed tree ensembles obtained higher mean overall accuracy (0.5–14% higher), producer’s accuracy (0.5–10% higher), and user’s accuracy (0.5–9% higher) than the classical tree ensembles, i.e. BT and RF. They were also much faster (e.g. 2–10 times) and more stable than their competitors for classification of these three datasets. In addition, unlike BT and RF, which obtained higher accuracy in large ensembles (i.e. the high number of DT), BFF and CFF can also be more efficient and reliable in smaller ensembles. Furthermore, the extended majority voting techniques could outperform the classical majority voting for decision fusion.     

A study of land cover classification using polarimetric SAR parameters

Classification of the Earth's surface constitutes an important application of polarimetric synthetic aperture radar (SAR) data; in turn, it may represent an efficient way for investigating their different representations. The polarimetric parameters most frequently taken into account for classification have been the incoherent ones. A similar use of coherent methods appears to have been scarcely considered and remained neglected. In this contribution, we wish to address this issue, testing and comparing a wide range of polarimetric SAR parameters, coherent and incoherent. Another original aspect of this work is the study of the dependence of the classification results on the varying size of averaging windows of pixels. Such an analysis will permit us to evaluate the importance of speckle reduction and to prove if the chosen polarimetric parameters describe only point‐like physical properties of the targets or if they also contain ‘extended’, local information. The goal is to provide an objective estimate of the quality of the classification of polarimetric parameters and afford their comparison, an exercise hitherto unavailable in the literature in common knowledge.     

Investigation of the effect of the incidence angle on land cover classification using fully polarimetric SAR images

Incidence angle is one of the most important imaging parameters that affect polarimetric SAR (PolSAR) image classification. Several studies have examined the land cover classification capability of PolSAR images with different incidence angles. However, most of these studies provide limited physical insights into the mechanism how the variation of incidence angle affects PolSAR image classification. In the present study, land cover classification was conducted by using RADARSAT-2 Wide Fine Quad-Pol (FQ) images acquired at different incidence angles, namely, FQ8 (27.75°), FQ14 (34.20°), and FQ20 (39.95°). Land cover classification capability was examined for each single-incidence angle image and a multi-incidence angle image (i.e., the combination of single-incidence angle images). The multi-incidence angle image produced better classification results than any of the single-incidence angle images, and the different incidence angles exhibited different superiorities in land cover classification. The effect mechanisms of incidence angle variation on land cover classification were investigated by using the polarimetric decomposition theorem that decomposes radar backscatter into single-bounce scattering, double-bounce scattering and volume scattering. Impinging SAR easily penetrated crops to interact with the soil at a small incidence angle. Therefore, the difference in single-bounce scattering between trees and crops was evident in the FQ8 image, which was determined to be suitable for distinguishing between croplands and forests. The single-bounce scattering from bare lands increased with the decrease in incidence angles, whereas that from water changed slightly with the incidence angle variation. Consequently, the FQ8 image exhibited the largest difference in single-bounce scattering between bare lands and water and produced the fewest confusion between them among all the images. The single- and double-bounce scattering from urban areas and forests increased with the decrease in incidence angles. The increase in single- and double-bounce scattering from urban areas was more significant than that from forests because C-band SAR could not easily penetrate the crown layer of forests to interact with the trunks and ground. Therefore, the FQ8 image showed a slightly better performance than the other images in discriminating between urban areas and forests. Compared with other crops and trees, banana trees caused stronger single- and double-bounce scattering because of their large leaves. As a large incidence angle resulted in a long penetration path of radar waves in the crown layer of vegetation, the FQ20 image enhanced the single- and double-bounce scattering differences between banana trees and other vegetation. Thus, the FQ20 image outperformed the other images in identifying banana trees.     

A unified framework for crop classification in southern China using fully polarimetric,dual polarimetric,and compact polarimetric SAR data

With the development of synthetic aperture radar (SAR) techniques, various imaging modes that involve single polarimetry, dual polarimetry, full polarimetry (FP), and compact polarimetry (CP) have been proposed and applied to SAR systems. This article attempts to introduce a unified framework for crop classification in southern China using FP, coherent HH/VV, and CP data. By analysing the polarimetric response from different land-cover types (including rice, banana trees, sugarcane, eucalyptus, water, and built-up areas in the experimental site) and by exploring the similarities between data in these three modes, a knowledge-based characteristic space is created and a unified classification framework is presented. Time-series data acquired by TerraSAR-X over the Leizhou Peninsula, southern China, are used in our experiments. The overall classification accuracies for data in the FP and coherent HH/VV modes are approximately 95%, and for data in the CP mode, the accuracy is 91%, which suggest that the proposed classification scheme is effective. Compared with the Wishart Maximum Likelihood (ML) classifier, the proposed method provides approximately 5.64%, 7.30%, and 6.48% higher classification accuracies in the FP, HH/VV, and circular transmit and dual circular receive modes, respectively.     

Land cover mapping for tropical forest rehabilitation planning using remotely-sensed data

Lack of reliable and up-to-date maps relating to land cover (among other themes) constitute a weakness in land resource surveys and cause costly failures to many forest rehabilitation projects in the tropics. This study evaluated the utility of satellite imagery for land cover mapping for forest rehabilitation planning in a case study in Mindoro, Philippines. Using Landsat TM data, visual and digital image processing techniques were performed using the GRID module of ARC/INFO and the microBRIAN image processing software. Crown cover density is found as the most useful and the most important detail of information the image could provide. Detailed mapping at the species and forest type levels is unreliable, as is the delineation of water bodies and some cultural features in rugged terrain. Clustering of the NDVI image is found more applicable in producing land cover maps depicting crown cover classes than classifying raw TM-3, -4, and-5.     

Hierarchical mapping of Northern Eurasian land cover using MODIS data

The Northern Eurasian land mass encompasses a diverse array of land cover types including tundra, boreal forest, wetlands, semi-arid steppe, and agricultural land use. Despite the well-established importance of Northern Eurasia in the global carbon and climate system, the distribution and properties of land cover in this region are not well characterized. To address this knowledge and data gap, a hierarchical mapping approach was developed that encompasses the study area for the Northern Eurasia Earth System Partnership Initiative (NEESPI). The Northern Eurasia Land Cover (NELC) database developed in this study follows the FAO-Land Cover Classification System and provides nested groupings of land cover characteristics, with separate layers for land use, wetlands, and tundra. The database implementation is substantially different from other large-scale land cover datasets that provide maps based on a single set of discrete classes. By providing a database consisting of nested maps and complementary layers, the NELC database provides a flexible framework that allows users to tailor maps to suit their needs. The methods used to create the database combine empirically derived climate–vegetation relationships with results from supervised classifications based on Moderate Resolution Imaging Spectroradiometer (MODIS) data. The hierarchical approach provides an effective framework for integrating climate–vegetation relationships with remote sensing-based classifications, and also allows sources of error to be characterized and attributed to specific levels in the hierarchy. The cross-validated accuracy was 73% for the land cover map and 73% and 91% for the agriculture and wetland classifications, respectively. These results support the use of hierarchical classification and climate–vegetation relationships for mapping land cover at continental scales.     

Guidance on and comparison of machine learning classifiers for Landsat-based land cover and land use mapping

Remote sensing scientists are increasingly adopting machine learning classifiers for land cover and land use (LCLU) mapping, but model selection, a critical step of the machine learning classification, has usually been ignored in the past research. In this paper, step-by-step guidance (for classifier training, model selection, and map production) with supervised learning model selection is first provided. Then, model selection is exhaustively applied to different machine learning (e.g. Artificial Neural Network (ANN), Decision Tree (DT), Support Vector Machine (SVM), and Random Forest (RF)) classifiers to identify optimal polynomial degree of input features (d) and hyperparameters with Landsat imagery of a study region in China and Ghana. We evaluated the map accuracy and computing time associated with different versions of machine learning classification software (i.e. ArcMap, ENVI, TerrSet, and R).

The optimal classifiers and their associated polynomial degree of input features and hyperparameters vary for the two image datasets that were tested. The optimum combination of d and hyperparameters for each type of classifier was used across software packages, but some classifiers (i.e. ENVI and TerrSet ANN) were customized due to the constraints of software packages. The LCLU map derived from ENVI SVM has the highest overall accuracy (72.6%) for the Ghana dataset, while the LCLU map derived from R DT has the highest overall accuracy (48.0%) for the FNNR dataset. All LCLU maps for the Ghana dataset are more accurate compared to those from the China dataset, likely due to more limited and uncertain training data for the China (FNNR) dataset. For the Ghana dataset, LCLU maps derived from tree-based classifiers (ArcMap RF, TerrSet DT, and R RF) routines are accurate, but these maps have artefacts resulting from model overfitting problems.     

Fusion of multisensor multitemporal satellite data for land cover mapping

Efficient integration of remote sensing information with different temporal, spectral and spatial resolutions is important for accurate land cover mapping. A new temporal fusion classification (TFC) model is presented for land cover classification, based on statistical fusion of multitemporal satellite images. In the proposed model, the temporal dependence of multitemporal images is taken into account by estimating transition probabilities from the change pattern of a vegetation dynamics indicator (VDI). Extension of this model is applicable to Synthetic Aperture Radar (SAR) images and integration of multisensor multitemporal satellite images, concerning both temporal attributes and reliability of multiple data sources. The feasibility of the new method is verified using multitemporal Landsat Thematic Mapper (TM) and ERS SAR satellite images, and experimental results show improved performance over conventional methods.     

Classification of the Yellow River delta area using fully polarimetric SAR measurements

In this study, an analysis of the polarimetric synthetic aperture radar (SAR) capabilities to classify coastal areas is undertaken. The Yellow River delta (China) is selected as the test case since it represents an extraordinary environmental and economical area, which is characterized by a very heterogeneous scattering scenario, as witnessed by official reference data, provided by the Chinese government, that classified 12 different kinds of environment. Experimental results, obtained applying two well-known unsupervised classifiers, namely the H/α-based and the Freeman–Durden model-based algorithms, to a fully polarimetric SAR scene collected by Radarsat-2 in 2008 are compared and critically discussed. Both provide a satisfactory global accuracy (larger than 60% in average) with reference to the inland Yellow River delta area, but there are subareas that result in misclassifications and severe classification ambiguities. This study also suggests including single-polarization intensity information to improve the classification accuracy and to partly solve ambiguities.     

Fuzzy classification of earth terrain covers using complex polarimetric SAR data

The conventional approach of terrain image classification that assigns a specific class for each pixel is inadequate, because the area covered by each pixel may embrace more than a single class. Fuzzy set theory which has been developed to deal with imprecise information can be incorporate in the analysis for a more appropriate solution to this problem. In the current state of imaging radar technology, polarimetric synthetic aperture radar (SAR) is unique in providing complete polarization information of ground covers for more effective classification than a single polarization radar. In this paper, we use the fuzzy c-means clustering algorithm for unsupervised segmentation of multi-look polarimetric SAR images. A statistical distance measure adopted in this algorithm is derived from the complex Wishart distribution of the complex covariance matrix. In classifying polarimetric SAR imagery, each terrain class is characterized by its own feature covariance matrix. The algorithm searches for cluster centres for each class and generates a fuzzy partition for the whole image. Membership grades obtained for each pixel provide detailed information about spatial terrain variations. Classification of the image is achieved by choosing a defuzzification criterion. When the back-scattering characteristics of two or more classes are not well distinguished from each other, a divisive hierarchical clustering procedure is adopted to locate their respective feature covariance matrices. NASA/JPL AIRSAR data is used to substantiate this fuzzy classification algorithm.     

Advances in classification for land cover mapping using SPOT HRV imagery

Abstract

High-resolution data from the HRV (High Resolution Visible) sensors onboard the SPOT-1 satellite have been utilized for mapping semi-natural and agricultural land cover using automated digital image classification algorithms. Two methods for improving classification performance are discussed. The first technique involves the use of digital terrain information to reduce the effects of topography on spectral information while the second technique involves the classification of land-cover types using training data derived from spectral feature space. Test areas in Snowdonia and the Somerset Levels were used to evaluate the methodology and promising results were achieved. However, the low classification accuracies obtained suggest that spectral classification alone is not a suitable tool to use in the mapping of semi-natural cover types.     

Analysis of coastal ice cover using ERS-I SAR data

An applications demonstration of the use of Synthetic Aperture Radar (SAR) data in an operational selling is being conducted by the National Oceanic and Atmospheric Administration (NOAA) CoastWatch Program. In the development phase of this demonstration, case studies were conducted to assess the utility of SAR data for monitoring coastal ice in the Bering Sea, icebergs from calving glaciers in Prince William Sound, and lake ice in the Great Lakes. ERS-l SAR data was used in these studies. Results showed that depending on size and sea state icebergs could be detected from background and computer enhanced in the imagery, thaI SAR data can supplement and enhance the utility of satellite visible and infrared data sources for coastal ice monitoring, and that Greal Lakes ice cover can be classified by ice type and mapped in the SAR data using image processing techniques. Cloud cover was a common problem. Based on the further development of automated analysis algorithms and the increase in frequency of SAR coverage, the all-weather, day/night viewing capabilities of SAR make it a unique and valuable tool for operational ice detection and monitoring.     

Subpixel measurement of tropical forest cover using AVHRR data

Abstract

Tropical forest assessment using data from the Advanced Very High Resolution Radiometer (AVHRR) may lead to inaccurate estimates of forest cover in regions of small subpixel forest or non-forest patches and in regions where the pattern of clearance is particularly convoluted. Test sites typifying these two patterns were chosen in Ghana and Rondonia, respectively. To capture the subpixel proportions of forest cover, a linear mixture model was applied to two AVHRR test images over the test sites. The model produced image outputs in which pixel intensities indicated the proporton of forest cover per km². For comparison, supervised maximum likelihood classifications were also performed. The outputs were assessed against classified Landsat TM scenes, converted to proportions maps and coregistered to the AVHRR images. An empirical method was applied for determining the critical forest cover per km² needed for an AVHRR pixel to be classified as forest. The critical values exceeded 50 per cent, indicating a tendency for AVHRR classification to underestimate forest cover. This was confirmed by comparing estimates of total forest cover obtained from the AVHRR and TM classifications. In the case of Ghana, a more accurate estimate of forest cover was obtained from the AVHRR mixture model than from the classification. Both mixture model outputs were found to be well correlated with those from Landsat TM. Further work should test the robustness of the approach adopted here when applied to much larger areas.     

Measurement of ocean surface slopes and wave spectra using polarimetric SAR image data

Methods have been investigated which use fully polarimetric synthetic aperture radar (SAR) image data to measure ocean slopes and wave spectra. Independent techniques have been developed to measure wave slopes in the SAR azimuth and range directions. The azimuth slope technique, in particular, is a more direct measurement than conventional, intensity based, backscatter cross-section measurements.In the azimuth direction, wave-induced perturbations of the polarimetric orientation angle are used to sense the wave slopes. In the range direction, a new technique involving the alpha parameter from the Cloude-Pottier H-A-? (Entropy, Anisotropy, and (averaged) Alpha) polarimetric scattering decomposition theorem is used to measure slopes. Both measurement types are sensitive to ocean wave slopes and are directional. Taken together, they form a means of using polarimetric SAR (POLSAR) image data to make complete measurements of either ocean wave slopes, or directional wave spectra.These measurements must still contend with fundamental nonlinearities in the SAR image processing (i.e., azimuth direction “velocity bunching”) that are due to wave velocity and acceleration effects.NASA/JPL/AIRSAR L-, and P-band data from California coastal waters were used in the studies. Wave parameters measured using the new methods are compared with those developed using both conventional SAR intensity based methods, and with in situ NOAA National Data Center buoy measurement products.     

Deriving terrain and textural information from stereo RADARSAT data for mountainous land cover mapping

A new procedure is proposed for land cover classification in a mountainous area using stereo RADARSAT-1 data. The method integrates a few types of information that can be extracted from the same stereo RADARSAT images: (1) the Digital Elevation Model (DEM) generated from the stereo RADARSAT images; (2) terrain information (elevation, slope and aspect) extracted from the derived DEM; and (3) textural information derived from the same RADARSAT images. An Artificial Neural Network (ANN) classifier is applied for the land cover classification. Performance of the proposed method is evaluated using a mountainous study area in Southern Argentina, where there is a lack of up-to-date information for environmental monitoring. The results show that the integration of textural and terrain information can greatly improve the accuracy of the classification using the ANN classifier. It demonstrates that stereo RADARSAT images provide valuable data sources for land cover mapping, especially in mountainous areas where cloud cover is a problem for optical data collection and topographical data are not always available.     

Effects of minimum mapping unit on land cover data spatial configuration and composition

A key issue when generating a land cover map from remotely sensed data is the selection of the minimum mapping unit (MMU) to be employed, which determines the extent of detail contained in the map. This study analyses the effects of MMU in land cover spatial configuration and composition, by using simulated landscape thematic patterns generated by the Modified Random Clusters method. This approach allows a detailed control of the different factors influencing landscape metrics behaviour, as well as taking into account a wide range of land cover pattern possibilities. Land cover classes that are sparse and fragmented can be considerably misrepresented in the final map when increasing MMU, while the classes that occupy a large percentage of map area tend to become more dominant. Mean Patch Size and Number of Patches are very poor indicators of pattern fragmentation in this context. In contrast, Landscape Division (LD) and related indices (Splitting Index and Effective Mesh Size) are clearly suitable for comparing the fragmentation of landscape data with different MMUs. We suggest that the Mean Shape Index, the most sensitive to MMU of those considered in this study, should not be used in further landscape studies if land cover data with different MMU or patch size frequency distribution are to be compared. In contrast, the Area Weighted Mean Shape Index presents a very robust behaviour, which advocates the use of this index for the quantification of the overall irregularity of patch shapes in landscape spatial patterns. The results presented allow quantifying the biases resulting from selecting a certain MMU when generating a land cover dataset. In general, a bigger MMU implies underestimating landscape diversity and fragmentation, as well as over-estimating animal population dispersal success. Guidelines are provided for the proper use and comparison of spatial pattern indices measured in maps with different MMUs.     

Structured approach to land cover mapping or the Atlantic zone of Costa Rica using single date TM data

Abstract

A structured approach to land cover mapping, involving different stages of field work and processing of remote sensing data, is presented. The Processing of TM data of one acquisition date was done by Analysing the Digital data-Structure (PAD) to produce optimum imagery for land-cover mapping of the Atlantic zone in Costa Rica.

Three stages were evaluated:

1. Image-processing in the pre-fieldwork stage to obtain insight into the overall variation of the scene.

2. Small scale reconnaissance fieldwork, and processing thereafter, directed towards the production of thematic imagery guided by the properties of objects and features.

3. Medium scale reconnaissance fieldwork and classification.

With this method we made use of statistical data, such as standard deviations and correlation coefficients, graphic presentations of mean values for the evaluation of ratios as well as variance percentages expressed by principal components. The selection of training fields for statistic calculation was considered to be essential for the final result.     

High performance computing algorithms for land cover dynamics using remote sensing data

Global and regional land cover studies need to apply complex models on selected subsets of large volumes of multi-sensor and multi-temporal data sets that have been derived from raw instrument measurements using widely accepted pre-processing algorithms. The computational and storage requirements of most of these studies far exceed what is possible on a single workstation environment. We have been pursuing a new approach that couples scalable and open distributed heterogeneous hardware with the development of high performance software for processing, indexing and organizing remotely sensed data. Hierarchical data management tools are used to ingest raw data, create metadata and organize the archived data so as to automatically achieve computational load balancing among the available nodes and minimize input/output overheads. We illustrate our approach with four specific examples. The first is the development of the first fast operational scheme for the atmospheric correction of Landsat Thematic Mapper scenes, while the second example focuses on image segmentation using a novel hierarchical connected components algorithm. Retrieval of the global Bidirectional Reflectance Distribution Function in the red and near-infrared wavelengths using four years (1983 to 1986) of Pathfinder Advanced Very High Resolution Radiometer (AVHRR) Land data is the focus of our third example. The fourth example is the development of a hierarchical data organization scheme that allows on-demand processing and retrieval of regional and global AVHRR data sets. Our results show that substantial reductions in computational times can be achieved by the high performance computing technology.     

The suitability of decadal image data sets for mapping tropical forest cover change in the Democratic Republic of Congo: implications for the global land survey

Landsat remote sensing of the central African humid tropics is confounded by persistent cloud cover and, since 2003, missing data due to the Landsat‐7 Enhanced Thematic Mapper Plus (ETM+) scan line corrector (SLC) malfunction. To quantify these limitations and their effects on contemporary forest cover and change characterization, a comparison was made of multiple Landsat‐7 image mosaics generated for a six Landsat path/row study site in central Africa for 2000 and 2005. Epoch 2000 mosaics were generated by compositing (i) two to three Landsat acquisitions per path/row, (ii) using the best single GeoCover 2000 acquisition for each path/row. Epoch 2005 composites were generated by compositing SLC‐off data using (iii) five to seven acquisitions per path/row, (iv) three acquisitions per path/row. Eighty per cent of pixels were of suitable quality for change detection between (ii) and (iv), emulating that which is possible with current GeoCover and planned Global Land Survey (GLS) inputs. In a more data intensive change detection analysis using mosaics (i) and (iii), 96% of pixels had suitable quality. Compositing more acquisitions per path/row for the study area systematically reduced the percentage of SLC‐off gaps and, when more than three acquisitions were composited, reduced the percentage of pixels with high likelihood of cloud, haze or shadow. The results indicate that additional input imagery to augment both the Geocover and GLS data may be required to enable forest cover and change analyses for regions of the humid tropics.