Discrimination and classification of mangrove forests using EO-1 Hyperion data: a case study of Indian Sundarbans

In remote sensing the identification accuracy of mangroves is greatly influenced by terrestrial vegetation. This paper deals with the use of specific vegetation indices for extracting mangrove forests using Earth Observing-1 Hyperion image over a portion of Indian Sundarbans, followed by classification of mangroves into floristic composition classes. Five vegetation indices (three new and two published), namely Mangrove Probability Vegetation Index, Normalized Difference Wetland Vegetation Index, Shortwave Infrared Absorption Index, Normalized Difference Infrared Index and Atmospherically Corrected Vegetation Index were used in decision tree algorithm to develop the mangrove mask. Then, three full-pixel classifiers, namely Minimum Distance, Spectral Angle Mapper and Support Vector Machine (SVM) were evaluated on the data within the mask. SVM performed better than the other two classifiers with an overall precision of 99.08%. The methodology presented here may be applied in different mangrove areas for producing community zonation maps at finer levels.     

Learning with transductive SVM for semisupervised pixel classification of remote sensing imagery

Land cover classification using remotely sensed data requires robust classification methods for the accurate mapping of complex land cover area of different categories. In this regard, support vector machines (SVMs) have recently received increasing attention. However, small number of training samples remains a bottleneck to design suitable supervised classifiers. On the other hand, adequate number of unlabeled data is available in remote sensing images which can be employed as additional source of information about margins. To fully leverage all of the precious unlabeled data, integration of filtering in a transductive SVM is proposed.Using two labeled image datasets of small size and two large unlabeled image datasets, the effectiveness of the proposed method is explored. Experimental results show that the proposed technique achieves average overall accuracies of around 4.5–7.8%, 0.8–2.6% and 0.9–2.2% more than the standard inductive SVM (ISVM), progressive transductive SVM (PTSVM) and low density separation (LDS) classifiers, respectively on larger domains in case of labeled datasets. Using image datasets, visual interpretation from the classified images as well as the segmentation quality reveal that the proposed method can efficiently filter informative data from the unlabeled samples.     

一种改进光谱角匹配的水体信息提取方法

快速、准确地从卫星遥感影像上获取水体信息已成为水资源调查及监测、湿地保护、洪涝灾害评估等领域的重要技术手段。本文以GF-4卫星的PMS传感器影像为数据源,提出了一种改进光谱角匹配（MSAM）的水体信息提取方法,以内蒙古中东部和长江中下游两个试验区为研究对象,对比分析了MSAM与单波段阈值法、NDWI阈值法、支持向量机和光谱角匹配等传统方法的水体信息提取结果,并进行了精度评价。试验结果表明,本文提出的MSAM方法不仅能准确地提取水体信息,而且能很好地区分水体与云阴影,对细小水体的提取也具有很好的效果。在内蒙古中部和长江中下游两个试验区的水体提取精度分别达到99.86%和98.37%,在5种水体提取方法中的精度最高,可以有效地提取水体信息。     

Analyzing fine-scale wetland composition using high resolution imagery and texture features

In order to monitor natural and anthropogenic disturbance effects to wetland ecosystems, it is necessary to employ both accurate and rapid mapping of wet graminoid/sedge communities. Thus, it is desirable to utilize automated classification algorithms so that the monitoring can be done regularly and in an efficient manner. This study developed a classification and accuracy assessment method for wetland mapping of at-risk plant communities in marl prairie and marsh areas of the Everglades National Park. Maximum likelihood (ML) and Support Vector Machine (SVM) classifiers were tested using 30.5 cm aerial imagery, the normalized difference vegetation index (NDVI), first and second order texture features and ancillary data. Additionally, appropriate window sizes for different texture features were estimated using semivariogram analysis. Findings show that the addition of NDVI and texture features increased classification accuracy from 66.2% using the ML classifier (spectral bands only) to 83.71% using the SVM classifier (spectral bands, NDVI and first order texture features).     

Deep neural network for complex open-water wetland mapping using high-resolution WorldView-3 and airborne LiDAR data

Wetland inventory maps are essential information for the conservation and management of natural wetland areas. The classification framework is crucial for successful mapping of complex wetlands, including the model selection, input variables and training procedures. In this context, deep neural network (DNN) is a powerful technique for remote sensing image classification, but this model application for wetland mapping has not been discussed in the previous literature, especially using commercial WorldView-3 data. This study developed a new framework for wetland mapping using DNN algorithm and WorldView-3 image in the Millrace Flats Wildlife Management Area, Iowa, USA. The study area has several wetlands with a variety of shapes and sizes, and the minimum mapping unit was defined as 20 m² (0.002 ha). A set of potential variables was derived from WorldView-3 and auxiliary LiDAR data, and a feature selection procedure using principal components analysis (PCA) was used to identify the most important variables for wetland classification. Furthermore, traditional machine learning methods (support vector machine, random forest and k-nearest neighbor) were also implemented for the comparison of results. In general, the results show that DNN achieved satisfactory results in the study area (overall accuracy = 93.33 %), and we observed a high spatial overlap between reference and classified wetland polygons (Jaccard index ∼0.8). Our results confirm that PCA-based feature selection was effective in the optimization of DNN performance, and vegetation and textural indices were the most informative variables. In addition, the comparison of results indicated that DNN classification achieved relatively similar accuracies to other methods. The total classification errors vary from 0.104 to 0.111 among the methods, and the overlapped areas between reference and classified polygons range between 87.93 and 93.33 %. Finally, the findings of this study have three main implications. First, the integration of DNN model and WorldView-3 image is useful for wetland mapping at 1.2-m, but DNN results did not outperform other methods in this study area. Second, the feature selection was important for model performance, and the combination of most relevant input parameters contributes to the success of all tested models. Third, the spatial resolution of WorldView-3 is appropriate to preserve the shape and extent of small wetlands, while the application of medium resolution image (30-m) has a negative impact on the accurate delineation of these areas. Since commercial satellite data are becoming more affordable for remote sensing users, this study provides a framework that can be utilized to integrate very high-resolution imagery and deep learning in the classification of complex wetland areas.     

Evaluation of different machine learning methods for land cover mapping of a Mediterranean area using multi-seasonal Landsat images and Digital Terrain Models

Land cover monitoring using digital Earth data requires robust classification methods that allow the accurate mapping of complex land cover categories. This paper discusses the crucial issues related to the application of different up-to-date machine learning classifiers: classification trees (CT), artificial neural networks (ANN), support vector machines (SVM) and random forest (RF). The analysis of the statistical significance of the differences between the performance of these algorithms, as well as sensitivity to data set size reduction and noise were also analysed. Landsat-5 Thematic Mapper data captured in European spring and summer were used with auxiliary variables derived from a digital terrain model to classify 14 different land cover categories in south Spain. Overall, statistically similar accuracies of over 91% were obtained for ANN, SVM and RF. However, the findings of this study show differences in the accuracy of the classifiers, being RF the most accurate classifier with a very simple parameterization. SVM, followed by RF, was the most robust classifier to noise and data reduction. Significant differences in their performances were only reached for thresholds of noise and data reduction greater than 20% (noise, SVM) and 25% (noise, RF), and 80% (reduction, SVM) and 50% (reduction, RF), respectively.     

Fusion of RADARSAT-2 and multispectral optical remote sensing data for LULC extraction in a tropical agricultural area

In this study, we investigated the performance of different fusion and classification techniques for land cover mapping in Hilir Perak, Peninsula Malaysia using RADAR and Landsat-8 images in a predominantly agricultural area. The fusion methods used are Brovey Transform, Wavelet Transform, Ehlers and Layer Stacking and their results classified into seven different land cover classes which include (1) pixel-based classifiers (spectral angle mapper (SAM), maximum likelihood (ML), support vector machine (SVM)) and (2) Object-based (rule-based and standard nearest neighbour (NN)) classifiers. The result shows that pixel-based classification achieved maximum accuracy of the optical data classification using SVM in Landsat-8 with 74.96% accuracy compared to SAM and ML. For multisource data classification, the highest overall accuracy recorded for layer stacking (SVM) was 79.78%, Ehlers fusion (SVM) with 45.57%, Brovey fusion (SVM) with 63.70% and Wavelet fusion (SVM) 61.16%. And for object-based classifiers, the overall classification accuracy is 95.35% for rule-based and 76.33% for NN classifier, respectively. Based on the analysis of their performances, object-based and the rule-based classifiers produced the best classification accuracy from the fused images.     

A systematic comparison of different object-based classification techniques using high spatial resolution imagery in agricultural environments

Geographic Object-Based Image Analysis (GEOBIA) is becoming more prevalent in remote sensing classification, especially for high-resolution imagery. Many supervised classification approaches are applied to objects rather than pixels, and several studies have been conducted to evaluate the performance of such supervised classification techniques in GEOBIA. However, these studies did not systematically investigate all relevant factors affecting the classification (segmentation scale, training set size, feature selection and mixed objects). In this study, statistical methods and visual inspection were used to compare these factors systematically in two agricultural case studies in China. The results indicate that Random Forest (RF) and Support Vector Machines (SVM) are highly suitable for GEOBIA classifications in agricultural areas and confirm the expected general tendency, namely that the overall accuracies decline with increasing segmentation scale. All other investigated methods except for RF and SVM are more prone to obtain a lower accuracy due to the broken objects at fine scales. In contrast to some previous studies, the RF classifiers yielded the best results and the k-nearest neighbor classifier were the worst results, in most cases. Likewise, the RF and Decision Tree classifiers are the most robust with or without feature selection. The results of training sample analyses indicated that the RF and adaboost. M1 possess a superior generalization capability, except when dealing with small training sample sizes. Furthermore, the classification accuracies were directly related to the homogeneity/heterogeneity of the segmented objects for all classifiers. Finally, it was suggested that RF should be considered in most cases for agricultural mapping.     

Tree Diversity Detection with Mid-Resolution Images and Environmental Data in a Neural Network

Biodiversity maps are crucial to conservation management. The present study assesses the accuracy of detecting tree diversity in an Italian forest site by combining mid-resolution images from Landsat-TM or Advanced Land Observation Satellite (ALOS)’s Advanced Visible and Near Infrared Radiometer type 2 (AVNIR-2) sensors with environmental data namely elevation, slope, aspect and solar radiation in an artificial Neural Network (NN) classifier. The map accuracies obtained for Landsat-TM and ALOS images are 60 % and 53 % respectively. Use of environmental data increases accuracies to 91 % and 81 % respectively. Landsat-TM detects tree diversity more accurately than ALOS. Both the coarser pixel size and finer spectral resolution of Landsat-TM contributed to its higher accuracy.     

Optical and SAR sensor synergies for forest and land cover mapping in a tropical site in West Africa

The classification of tropical fragmented landscapes and moist forested areas is a challenge due to the presence of a continuum of vegetation successional stages, persistent cloud cover and the presence of small patches of different land cover types. To classify one such study area in West Africa we integrated the optical sensors Landsat Thematic Mapper (TM) and the Advanced Visible and Near Infrared Radiometer type 2 (AVNIR-2) with the Phased Arrayed L-band SAR (PALSAR) sensor, the latter two on-board the Advanced Land Observation Satellite (ALOS), using traditional Maximum Likelihood (MLC) and Neural Networks (NN) classifiers. The impact of texture variables and the use of SAR to cope with optical data unavailability were also investigated. SAR and optical integrated data produced the best classification overall accuracies using both MLC and NN, respectively equal to 91.1% and 92.7% for TM and 95.6% and 97.5% for AVNIR-2. Texture information derived from optical images was critical, improving results between 10.1% and 13.2%. In our study area, PALSAR alone was able to provide valuable information over the entire area: when the three forest classes were aggregated, it achieved 75.7% (with MCL) and 78.1% (with NN) overall classification accuracies. The selected classification and processing methods resulted in fine and accurate vegetation mapping in a previously untested region, exploiting all available sensors synergies and highlighting the advantages of each dataset.     

Assessing the performance of two unsupervised dimensionality reduction techniques on hyperspectral APEX data for high resolution urban land-cover mapping

Despite the high richness of information content provided by airborne hyperspectral data, detailed urban land-cover mapping is still a challenging task. An important topic in hyperspectral remote sensing is the issue of high dimensionality, which is commonly addressed by dimensionality reduction techniques. While many studies focus on methodological developments in data reduction, less attention is paid to the assessment of the proposed methods in detailed urban hyperspectral land-cover mapping, using state-of-the-art image classification approaches. In this study we evaluate the potential of two unsupervised data reduction techniques, the Autoassociative Neural Network (AANN) and the BandClust method – the first a transformation based approach, the second a feature-selection based approach – for mapping of urban land cover at a high level of thematic detail, using an APEX 288-band hyperspectral dataset. Both methods were tested in combination with four state-of-the-art machine learning classifiers: Random Forest (RF), AdaBoost (ADB), the multiple layer perceptron (MLP), and support vector machines (SVM). When used in combination with a strong learner (MLP, SVM) BandClust produces classification accuracies similar to or higher than obtained with the full dataset, demonstrating the method’s capability of preserving critical spectral information, required for the classifier to successfully distinguish between the 22 urban land-cover classes defined in this study. In the AANN data reduction process, on the other hand, important spectral information seems to be compromised or lost, resulting in lower accuracies for three of the four classifiers tested. Detailed analysis of accuracies at class level confirms the superiority of the SVM/Bandclust combination for accurate urban land-cover mapping using a reduced hyperspectral dataset. This study also demonstrates the potential of the new APEX sensor data for detailed mapping of land cover in spatially and spectrally complex urban areas.     

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.     

Detecting and Downscaling Wet Areas on Boreal Landscapes

This letter presents an approach to classify wet areas from European Remote Sensing 2 (ERS-2) synthetic aperture radar (SAR)-, Landsat Thematic Mapper (TM)-, and Light Detection and Ranging (LiDAR)-derived terrain data and downscale the result from the coarse resolution of satellite images to finer resolutions needed for land managers. Using discrete wavelet transform (DWT) and support vector machines (SVM), the algorithm finds multiple relationships between the radar, optical, and terrain data and wet areas at different spatial scales. Decomposing and reconstructing processes are performed using a 2-D DWT (2D-DWT) and inverse 2D-DWT respectively. The underlying relationships between radar, optical, and terrain data and wet areas are learned by training an SVM at the coarse resolution of the wet-area map. The SVM is then applied on the predictors at a finer resolution to produce wet-area detailing images, which are needed to reconstruct a finer resolution wet-area map. The algorithm is applied to a boreal landscape in northern Alberta, Canada, characterized by many wet-area features including ephemeral and permanent streams and wetlands.     

Mapping wetland characteristics using temporally dense Sentinel-1 and Sentinel-2 data: A case study in the St. Lucia wetlands,South Africa

Wetlands have been determined as one of the most valuable ecosystems on Earth and are currently being lost at alarming rates. Large-scale monitoring of wetlands is of high importance, but also challenging. The Sentinel-1 and -2 satellite missions for the first time provide radar and optical data at high spatial and temporal detail, and with this a unique opportunity for more accurate wetland mapping from space arises. Recent studies already used Sentinel-1 and -2 data to map specific wetland types or characteristics, but for comprehensive wetland characterisations the potential of the data has not been researched yet. The aim of our research was to study the use of the high-resolution and temporally dense Sentinel-1 and -2 data for wetland mapping in multiple levels of characterisation. The use of the data was assessed by applying Random Forests for multiple classification levels including general wetland delineation, wetland vegetation types and surface water dynamics. The results for the St. Lucia wetlands in South Africa showed that combining Sentinel-1 and -2 led to significantly higher classification accuracies than for using the systems separately. Accuracies were relatively poor for classifications in high-vegetated wetlands, as subcanopy flooding could not be detected with Sentinel-1’s C-band sensors operating in VV/VH mode. When excluding high-vegetated areas, overall accuracies were reached of 88.5% for general wetland delineation, 90.7% for mapping wetland vegetation types and 87.1% for mapping surface water dynamics. Sentinel-2 was particularly of value for general wetland delineation, while Sentinel-1 showed more value for mapping wetland vegetation types. Overlaid maps of all classification levels obtained overall accuracies of 69.1% and 76.4% for classifying ten and seven wetland classes respectively.     

Evaluation of Morphological Texture Features for Mangrove Forest Mapping and Species Discrimination Using Multispectral IKONOS Imagery

This letter aims to exploit morphological textures in discriminating three mangrove species and surrounding environment with multispectral IKONOS imagery in a study area on the Caribbean coast of Panama. Morphological texture features are utilized to distinguish red (Rhizophora mangle), white (Laguncularia racemosa), and black (Avicennia germinans) mangroves and rainforest regions. Meanwhile, two fusion methods are presented, i.e., vector stacking and support vector machine (SVM) output fusion, for integrating the hybrid spectral–textural features. For comparison purposes, the object-based analysis and the gray-level co-occurrence matrix (GLCM) textures are adopted. Results revealed that the morphological feature opening by reconstruction (OBR) followed by closing by reconstruction (CBR) and its dual operator CBR followed by OBR gave very promising accuracies for both mangrove discrimination (89.1% and 91.1%, respectively) and forest mapping (91.4% and 93.7%, respectively), compared with the object-based analysis (80.5% for mangrove discrimination and 82.9% for forest mapping) and the GLCM method (81.9% and 87.2%, respectively). With respect to the spectral–textural information fusion algorithms, experiments showed that the SVM output fusion could obtain an additional 2.0% accuracy improvement than the vector-stacking approach.      

Comparative performance of convolutional neural network,weighted and conventional support vector machine and random forest for classifying tree species using hyperspectral and photogrammetric data

ABSTRACT

The classification of tree species can significantly benefit from high spatial and spectral information acquired by unmanned aerial vehicles (UAVs) associated with advanced classification methods. This study investigated the following topics concerning the classification of 16 tree species in two subtropical forest fragments of Southern Brazil: i) the potential integration of UAV-borne hyperspectral images with 3D information derived from their photogrammetric point cloud (PPC); ii) the performance of two machine learning methods (support vector machine – SVM and random forest – RF) when employing different datasets at a pixel and individual tree crown (ITC) levels; iii) the potential of two methods for dealing with the imbalanced sample set problem: a new weighted SVM (wSVM) approach, which attributes different weights to each sample and class, and a deep learning classifier (convolutional neural network – CNN), associated with a previous step to balance the sample set; and finally, iv) the potential of this last classifier for tree species classification as compared to the above mentioned machine learning methods. Results showed that the inclusion of the PPC features to the hyperspectral data provided a great accuracy increase in tree species classification results when conventional machine learning methods were applied, between 13 and 17% depending on the classifier and the study area characteristics. When using the PPC features and the canopy height model (CHM), associated with the majority vote (MV) rule, the SVM, wSVM and RF classifiers reached accuracies similar to the CNN, which outperformed these classifiers for both areas when considering the pixel-based classifications (overall accuracy of 84.4% in Area 1, and 74.95% in Area 2). The CNN was between 22% and 26% more accurate than the SVM and RF when only the hyperspectral bands were employed. The wSVM provided a slight increase in accuracy not only for some lesser represented classes, but also some major classes in Area 2. While conventional machine learning methods are faster, they demonstrated to be less stable to changes in datasets, depending on prior segmentation and hand-engineered features to reach similar accuracies to those attained by the CNN. To date, CNNs have been barely explored for the classification of tree species, and CNN-based classifications in the literature have not dealt with hyperspectral data specifically focusing on tropical environments. This paper thus presents innovative strategies for classifying tree species in subtropical forest areas at a refined legend level, integrating UAV-borne 2D hyperspectral and 3D photogrammetric data and relying on both deep and conventional machine learning approaches.     

Urban Shadow Detection and Classification Using Hyperspectral Image

Shadow is an inevitable problem in high-resolution remote sensing images. There are need and significance in extracting information from shadow-covered areas, such as in land-cover mapping. Although the illumination energy of shadow pixels is low, hyperspectral image can provides rich enough band information to differentiate various urban targets/materials and to classify them. This study firstly analyzes the spectra difference between shadow and non-shadow classes so as to detect shadow-pixel. To classify the shadow pixels, Spectral Angle Mapper (SAM) method was adopted to classify urban land-cover mapping, because it can reduce the influence resulted from different illumination intensity. Then, training samples were collected among different classes from the shadow pixels, and their Jeffries–Matusita (J–M) distance were computed to validate the spectral separability among classes, with the square distances of J–M among classes all bigger than 1.9. Finally, Maximum Likelihood Classifier (MLC) and Support Vector Machine (SVM) classifier were used to classify all the shadow pixels as different land-cover types. The results showed MLC and SVM outperform the SAM in classifying similar classes. The classification result in SVM was validated to find having conformity with ground truth.     

基于SVM的遥感影像的分类

传统的遥感图像的分类方法包括统计模式识别、句法模式识别、以及神经网络、遗传算法、模拟退火算法等。分析了统计模式识别的方法的优缺点,提出了使用SVM的方法进行遥感图像分类的设想,通过实验证明该方法是有效的和稳健的。     

Improved coastal wetland mapping using very-high 2-meter spatial resolution imagery

Accurate wetland maps are a fundamental requirement for land use management and for wetland restoration planning. Several wetland map products are available today; most of them based on remote sensing images, but their different data sources and mapping methods lead to substantially different estimations of wetland location and extent. We used two very high-resolution (2 m) WorldView-2 satellite images and one (30 m) Landsat 8 Operational Land Imager (OLI) image to assess wetland coverage in two coastal areas of Tampa Bay (Florida): Fort De Soto State Park and Weedon Island Preserve. An initial unsupervised classification derived from WorldView-2 was more accurate at identifying wetlands based on ground truth data collected in the field than the classification derived from Landsat 8 OLI (82% vs. 46% accuracy). The WorldView-2 data was then used to define the parameters of a simple and efficient decision tree with four nodes for a more exacting classification. The criteria for the decision tree were derived by extracting radiance spectra at 1500 separate pixels from the WorldView-2 data within field-validated regions. Results for both study areas showed high accuracy in both wetland (82% at Fort De Soto State Park, and 94% at Weedon Island Preserve) and non-wetland vegetation classes (90% and 83%, respectively). Historical, published land-use maps overestimate wetland surface cover by factors of 2–10 in the study areas. The proposed methods improve speed and efficiency of wetland map production, allow semi-annual monitoring through repeat satellite passes, and improve the accuracy and precision with which wetlands are identified.     

黄河三角洲地区GF-3雷达数据与Sentinel-2多光谱数据湿地协同分类研究

黄河三角洲湿地的动态变化监测对湿地资源合理利用、开发保护具有重要意义。采用C波段全极化高分三号（GF-3）合成孔径雷达数据与欧洲空间局哨兵二号（Sentinel-2B）多光谱数据,分析了黄河三角洲湿地7类地物的光谱、指数、极化散射以及纹理等特征信息,分别基于最大似然法（maximum likelihood,ML）、决策树（decision tree,DT）、支持向量机（support vector machine,SVM）方法实现了有监督分类,评估了两者协同与单独应用于湿地地物分类与识别的能力,结果表明,两者协同分类时,其总体精度分别可达90.4%、95.4%、95.7%,均明显高于两者单独分类的结果,证明了GF-3雷达数据与多光谱数据在湿地协同分类方面的可靠性和应用潜力。