遥感图像分类方法研究综述

 综述了遥感图像监督分类和非监督分类中的各种方法，介绍了各种方法的优缺点、适用领域和应用情况，并作了简单评述，最后，展望了遥感图像分类方法研究发展方向和研究热点。     

基于多特征的遥感影像分析——一个新的视角

为克服传统基于像元的遥感影像分析局限,本文提出了基于多特征的遥感影像分析方法,并以遥感分类为例对该方法作了相应的阐述.文章的最后部分对该方法的优越性作出了总结.     

高分辨率遥感图像分类方法在LUCC中的研究进展

高空间分辨率遥感图像在土地利用/覆盖(LUCC)变化研究中的应用,促进了遥感分类技术的进一步发展,表现在遥感分类对象、分类特征和分类器3个方面。本文对其研究进展情况进行了综述,介绍了具有代表性的分类方法,并对各种方法的特点进行了分析。最后,对遥感分类的相关研究进行了总结和展望。     

Geographic Object-Based Image Analysis – Towards a new paradigm

The amount of scientific literature on (Geographic) Object-based Image Analysis – GEOBIA has been and still is sharply increasing. These approaches to analysing imagery have antecedents in earlier research on image segmentation and use GIS-like spatial analysis within classification and feature extraction approaches. This article investigates these development and its implications and asks whether or not this is a new paradigm in remote sensing and Geographic Information Science (GIScience). We first discuss several limitations of prevailing per-pixel methods when applied to high resolution images. Then we explore the paradigm concept developed by Kuhn (1962) and discuss whether GEOBIA can be regarded as a paradigm according to this definition. We crystallize core concepts of GEOBIA, including the role of objects, of ontologies and the multiplicity of scales and we discuss how these conceptual developments support important methods in remote sensing such as change detection and accuracy assessment. The ramifications of the different theoretical foundations between the ‘per-pixel paradigm’ and GEOBIA are analysed, as are some of the challenges along this path from pixels, to objects, to geo-intelligence. Based on several paradigm indications as defined by Kuhn and based on an analysis of peer-reviewed scientific literature we conclude that GEOBIA is a new and evolving paradigm.     

基于纹理特征的遥感影像监督分类

影像分类技术是遥感影像分析与解译的重要基础。纹理特征是影像的重要特征,本文主要实现基于纹理特征的遥感影像监督分类。首先对地物样本进行提取,通过样本训练统计各类地物纹理特征向量,建立纹理特征库;然后以各类地物的特征向量作为基准,采用最短距离分类器对影像进行分类;最后采用混淆矩阵对分类结果进行精度评定,并与ERDAS专业软件分类结果进行对比分析。实验证明,本分方法取得了与ERDAS软件相当的分类效果,从而验证本文方法的可靠性。     

Discrimination of agricultural crops in a tropical semi-arid region of Brazil based on L-band polarimetric airborne SAR data

Recent articles are indicating that polarimetric data provide significantly more information than conventional or multi-polarized images, particularly due to the additional phase information. The objective of this paper is to evaluate the multi-polarized and fully polarimetric L-band airborne SAR-R99B data, in terms of their capability to distinguish among different agricultural crops in the western part of Bahia State, Brazil. Emphasis was given to coffee, cotton and pasture crops which were at well developed growing stages. Discrimination among crops was carried out using graphical analysis of mean backscatter values. Crop classification was performed for single and multiple polarizations, and fully polarimetric images with a classifier that uses the contextual Iterated Conditional Modes–ICM algorithm. The investigation confirmed the potential of L-band multi-polarized and polarimetric airborne SAR-R99B data to distinguish and classify agricultural crops in the tropical condition of the test-site. In addition, it clearly indicated the gradual and considerable improvement that was achieved going from single to three polarizations and from multi-polarized to fully polarimetric images.     

初论遥感与地图的融合

本文从方法论的角度论述了地理科学中的地图方法和遥感方法的共同特征,分析了二者之间的差异和联系,阐述了二者融合的两个理论问题(地图和遥感图像的综合与图形识别)和技术基础(地理图像处理),提出了为促进二者融合需要研究的主要内容及具体建议。     

自然光辐射各分量卫星遥感图像的计算机生成原理与方法

本文讨论了利用卫星遥感数据和与之匹配的数字地面模型以及它们和光辐射各分量之间的定量关系，使用计算逐点分解遥感数字图像以生成成像瞬间直、散射辐射光照射下相应的卫星遥感图像的原理、方法。最后，阐述了生成的辐射各分量遥感图像特点和应用前景。     

Integrating conventional classifiers with a GIS expert system to increase the accuracy of invasive species mapping

Mapping the cover of invasive species using remotely sensed data alone is challenging, because many invaders occur as mid-level canopy species or as subtle understorey species and therefore contribute little to the spectral signatures captured by passive remote sensing devices. In this study, two common non-parametric classifiers namely, the neural network and support vector machine were used to map four cover classes of the invasive shrub Lantana camara in a protected game reserve and the adjacent area under communal land management in Zimbabwe. These classifiers were each combined with a geographic information system (GIS) expert system, in order to test whether the new hybrid classifiers yielded significantly more accurate invasive species cover maps than the single classifiers. The neural network, when used on its own, mapped the cover of L. camara with an overall accuracy of 71% and a Kappa index of agreement of 0.61. When the neural network was combined with an expert system, the overall accuracy and Kappa index of agreement significantly increased to 83% and 0.77, respectively. Similarly, the support vector machine achieved an overall accuracy of 64% with a Kappa index of agreement of 0.52, whereas the hybrid support vector machine and expert system classifier achieved a significantly higher overall accuracy of 76% and a Kappa index of agreement of 0.67. These results suggest that integrating conventional image classifiers with an expert system increases the accuracy of invasive species mapping.     

一种黄土区土壤侵蚀强度遥感调查新方法

通过对TM图像的线性纹理提取和密度统计,首先获得了黄土丘陵区沟谷密度图,然后由沟谷密度反演区域土壤侵蚀强度。这种方法避开了影响土壤侵蚀量的诸多复杂因子。可直接从水土流失的外在表现---沟谷密度来反推土壤侵蚀强度。研究结果证明,这种方法对快速调查大范围黄土丘陵区的土壤侵蚀强度是十分有效的。     

Spatial and semantic effects of LUCAS samples on fully automated land use/land cover classification in high-resolution Sentinel-2 data

In this study, we test the use of Land Use and Coverage Area frame Survey (LUCAS) in-situ reference data for classifying high-resolution Sentinel-2 imagery at a large scale. We compare several pre-processing schemes (PS) for LUCAS data and propose a new PS for a fully automated classification of satellite imagery on the national level. The image data utilizes a high-dimensional Sentinel-2-based image feature space. Key elements of LUCAS data pre-processing include two positioning approaches and three semantic selection approaches. The latter approaches differ in the applied quality measures for identifying valid reference points and by the number of LU/LC classes (7–12). In an iterative training process, the impact of the chosen PS on a Random Forest image classifier is evaluated. The results are compared to LUCAS reference points that are not pre-processed, which act as a benchmark, and the classification quality is evaluated by independent sets of validation points. The classification results show that the positional correction of LUCAS points has an especially positive effect on the overall classification accuracy. On average, this improves the accuracy by 3.7%. This improvement is lowest for the most rigid sample selection approach, PS₂, and highest for the benchmark data set, PS₀. The highest overall accuracy is 93.1% which is achieved by using the newly developed PS₃; all PS achieve overall accuracies of 80% and higher on average. While the difference in overall accuracy between the PS is likely to be influenced by the respective number of LU/LC classes, we conclude that, overall, LUCAS in-situ data is a suitable source for reference information for large scale high resolution LC mapping using Sentinel-2 imagery. Existing sample selection approaches developed for Landsat imagery can be transferred to Sentinel-2 imagery, achieving comparable semantic accuracies while increasing the spatial resolution. The resulting LC classification product that uses the newly developed PS is available for Germany via DOI: https://doi.org/10.15489/1ccmlap3mn39.     

一种小波包遥感图像融合方法

本文介绍了一种基于小波包的遥感图像融合方法,实验采用熵、光谱扭曲指数和边缘指数等作为评价融合质量的定量指标,表明了通过这一融合方法比传统的IHS、PCA、HPF以及小波融合具有更高的光谱信息保持和纹理信息增强能力.     

矿山地质灾害特征遥感研究

应用Quick Bird遥感数据对山西晋城煤矿开采引发的地质灾害进行调查，研究了不同类型地质灾害（塌陷坑、地面沉陷、地 裂缝）的遥感影像特征，对矿区地质灾害现状、成因、分布规律特点和调查精度进行了分析评价。     

高光谱遥感图像的监督分类

图像分类是高光谱遥感图像分析与应用的重要手段。总结了目前用于高光谱图像监督分类的主要方法,包括最小距离法、最大似然法、神经元网络法和支持向量机法,分析了上述方法的特点,并探讨了高光谱遥感图像分类方法的发展趋势。     

基于权重与混合模型的遥感图像分类方法研究

提出了一种基于权重与混合像元模型的遥感图像分类方法。该方法在现有光谱混合模型的基础上,根据实际应用需要确定地类权重,通过地类丰度与权重因子加权平均确定像元的隶属类型,从而实现遥感图像分类。以SPOT-5土地覆盖遥感分类为例,对权重与混合像元模型结合的图像分类方法进行了验证,结果表明,该方法提高了遥感图像分类精度,在一定条件下更具实际意义。     

Multi-scale solution for building extraction from LiDAR and image data

This paper presents a multi-scale solution based on mathematical morphology for extracting the building features from remotely sensed elevation and spectral data. Elevation data are used as the primary data to delineate the structural information and are firstly represented on a morphological scale-space. The behaviors of elevation clusters across the scale-space are the cues for feature extraction. As a result, a complex structure can be extracted as a multi-part object in which each part is represented on a scale depending on its size. The building footprint is represented by the boundary of the largest part. Other object attributes include the area, height or number of stories. The spectral data is used as an additional source to remove vegetation and possibly classify the building roof material. Finally, the results can be stored in a multi-scale database introduced in this paper. The proposed solution is demonstrated using the data derived from a Light Detection And Ranging (LiDAR) surveying flight over Tokyo, Japan. The results show a reasonable match with reference data and prove the capability of the proposed approach in accommodation of diverse building shapes. Higher density LiDAR is expected to produce better accuracy in extraction, and more spectral sources are necessary for further classification of building roof material. It is also recommended that parallel processing should be implemented to reduce the computation time.     

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).     

证据理论结合遥感分类数据能力定量评价研究

DS(Dempster-Shafer)证据理论具有结合多源数据的能力,在遥感分类中应用越来越广泛.然而,并不是所有数据源利用证据理论结合后都能提高目标类别的基本概率分配(Basic Probability Assignment,BPA),从而提高遥感分类效果.如何对证据结合的效果进行评价已成为应用证据理论的一个关键问题...     

概率神经网络与BP网络模型在遥感图像分类中的对比研究

通过分析概率神经网络(以下称PNN)的基本结构及其训练算法,建立了卫星图像分类的概率神经网络模型,并通过实例对比分析了概率神经网络与BP网络分类模型的分类效果。实验表明,PNN图像分类方法在分类精度上优于误差反向传播神经网络模型,且分类时间相当,是一种有效的图像分类方法。     

A GIHS-based spectral preservation fusion method for remote sensing images using edge restored spectral modulation

High spatial resolution and spectral fidelity are basic standards for evaluating an image fusion algorithm. Numerous fusion methods for remote sensing images have been developed. Some of these methods are based on the intensity–hue–saturation (IHS) transform and the generalized IHS (GIHS), which may cause serious spectral distortion. Spectral distortion in the GIHS is proven to result from changes in saturation during fusion. Therefore, reducing such changes can achieve high spectral fidelity. A GIHS-based spectral preservation fusion method that can theoretically reduce spectral distortion is proposed in this study. The proposed algorithm consists of two steps. The first step is spectral modulation (SM), which uses the Gaussian function to extract spatial details and conduct SM of multispectral (MS) images. This method yields a desirable visual effect without requiring histogram matching between the panchromatic image and the intensity of the MS image. The second step uses the Gaussian convolution function to restore lost edge details during SM. The proposed method is proven effective and shown to provide better results compared with other GIHS-based methods.