Under the constant advancement of remote sensing technology, more and more high spatial and spectral resolution and high time frequency data are becoming available. The spatial resolution of satellite data is approaching 0.6m while for aerial imagery it is better than 0. 1m. The spectral resolution of data can be as high as 3-4nm. These developments not only greatly increased our capability and accuracy in information extraction and monitoring but also opened new application opportunities. Existing research activities have focused on how to increase the utilization efficiency of high resolution data, particularly for high spatial resolution data in a wide range of applications such as urban environment, precision agriculture, transportation and road infrastructure, forest inventory, artificial targets recognition and disaster risk assessment. However, the overall level of automation is still low. In this paper, we introduce some of the bottleneck problems and research questions related to information extraction from high spatial resolution imagery, and high spectral resolution and polar imagery, data fusion, and high spatial resolution image change detection. We suggest that an image base and a corresponding relevant knowledge base be built to improve data sharing and to facilitate investigations in this field.

近年来，不断发展的遥感技术使遥感数据呈现出高空间分辨率、高光谱分辨率和高时间采集频率的特点。卫星图像空间分辨率已经提高到0．6m级，而航空遥感数字影像分辨率高达0．1m以上。光谱分辨率高达3―4nm。不断发展的高分辨率遥感数据能够提高信息提取和监测精度，并拓展遥感数据的应用范围。目前，国外已经加快对高分辨率图像，特别是高空间分辨率影像，在城市环境、精准农业、交通及道路设施、林业测量、军事目标识别和灾害评估中的应用。但是总的情况是自动化程度不高。介绍高空间分辨率影像信息提取、高光谱和偏振影像信息提取、影像数据融合和高分辨率遥感变化探测等方面迫切需要研究的一些科学问题及其意义。建议建立图像知识库，改善数据共享环境，为有志于从事这方面研究的学者提供参考。

传统的基于像素的统计特征的分类方法在处理高分辨率影像的分类问题上遇到了很大的困难.本研究利用面向对象的影像分析方法对IKONOS全色影像进行了河流与道路的分类,包括利用影像对象的光谱特征的初次分类和利用子目标对象的线特征的二次分类两个过程;其中初次分类由于仅依据光谱信息,不能很好地将河流或道路与其他光谱特征相似的地物区分开,而通过引入子目标的形状特征进行二次分类,就可以准确地将河流与道路提取出来.试验结果表明,面向对象的分类方法能够满足高分辨率或纹理影像的分类需要,具有很大的应用潜力.

With the rapid increase of remote sensing image storage, it becomes more critical for the quick and effective information extraction from remote sensing imagery. As a widely-used method, object-based image analysis (OBIA) has been rapidly developed from the beginning of this century, but the automatic procedure for land use mapping is still problematic facing with geographical complexity. Regarding to the complex feature contents in the imagery of costal zones, this paper presents a method of optimal segmentation scale extraction and an object-based multi-layer classification procedure. The proposed approach mainly contains three parts: segmentation, optimal scale generation and multi-level classification. First, we select the high resolution images as the data source, segment the imagery with series of scale parameters. Then choose the appropriate scales with the curve of local variance (LV) variation. Variation in heterogeneity is explored by evaluating LV plotted against the corresponding scale in order to get different types of the landuse/cover with their own extraction scales. Finally, we classify the image with multi-features, including spectral, shape, texture and spatial relationship. This paper selects the coastal area of Zhuhai, Guangdong Province as the experiment zone, the classification results show that overall accuracy and Kappa index of the new method are better than those of the traditional pixel-based classifiers and object-oriented classifiers based on the single-level segmentation.

为了快速、准确地提取我国海岸带地区土地利用及其变化信息,选择高分辨率遥感影像作为数据源,提出了最优分割尺度下的遥感多层次地物识别分类方法。首先,通过改进的局部方差法进行最优分割尺度的确定,建立影像中各对象的方差均值与变化率随分割尺度变化曲线,确定方差均值的峰值,以变化率开始呈现下降趋势时所对应的分割值为最优分割尺度参考;然后,针对地物分类特征差异选取各自适宜的分割尺度,建立多层次地物特征表达与规则,最后,实现最优尺度分割选择下的遥感多层次识别分类,即实现较大尺度下分割形成父对象,而较小尺度下分割出其若干子对象的目标,提出了快速、自动化获取土地利用/覆盖图的策略流程。本文选取了广东省珠海市海岸带地区作为实验区,利用多层次遥感分类方法进行地物识别分类。结果表明,其目视效果以及总体精度、Kappa系数,均优于传统方法和单一分割尺度下的影像分类方法。

Segmentation is a basic and pivotal step of object-oriented remote sensing image classification, and the scale is a key problem of image segmentation. Aiming at the optimal segmentation scale selection for object-oriented remote sensing image classification, conformity degree between vectorial boundary lines of image region object after segmentation and true boundary lines of classification objects as criterion, through their multi-dimensions distance to define the conformity degree, the paper brought forward a new method of optimal segmentation scale selection for object-oriented remote sensing image classification-vector distance index method. Research verified the validity and applicability of this method through two experiments�� One experiment compared the results of optimal segmentation scale selection based on vector distance index method and ‘trial and error’ method. Results showed the vector distance index could reflect the optimal segmentation scale for seven classes exactly. The other experiment classified the segmentation result from the first experiment, through the accuracy assessment, explored the relationship of selection result based on vector distance index method and classification accuracy. Results showed water, dryland, rice, forest and resident region gained the highest accuracy on the scale that was selected by the vector distance index method, although marsh and grass didn��t gain the highest accuracy on the scale that was selected by the vector distance index method, through further analysis, the classification result tallied with the practical condition. Then both two experiment results showed that this method could realize the optimal segmentation scale selection for object-oriented remote sensing image classification, which was similar to human thought, unconfined to data source, intuitionistic, comprehensible and practical. Based on the basic theory of vector distance index method, aiming at the ‘submergence’ and ‘fragmentation’ phenomenon, research brought forward a scale index, which could reflect the small or big status for a given object type, and provided a quantitative tool to assess the conflict degree between ‘submergence’ and ‘fragmentation’, then showed its significance during the process of segmentation scale selection.

影像分割是面向对象遥感影像分类的基础步骤，而分割尺度又是影像分割的核心问题。研究针对面向对象遥感影像分类中的最优分割尺度选择问题，以分割后影像区域对象矢量边界线与欲分类目标对象真实矢量边界的吻合程度为标准，通过两者多向距离量化吻合程度，提出了一种最优分割尺度定量选择的新方法――矢量距离指数法。通过两种实验，同步验证了该方法的正确性与适用性，实验1将基于矢量距离指数法选择的最优分割尺度结果与较为成熟的人为试错法的选择结果比较，结果表明针对7种地类的矢量距离指数均可以正确反映最优分割尺度；实验2挖掘了矢量距离

由于客观现实世界中的实体、模式与过程表现为固有的等级结构，因此与等级相对应的尺度问题成为各个领域中研究的主题之一，是自然学科与人文学科中一种普遍存在的问题，也是遥感应用领域中的基础问题。遥感影像是对具有尺度依赖性的空间格局与过程的反映，不同尺度的遥感信息反映着地表特性不同层次的空间结构差异，单尺度的影像分析已不能满足反映空间格局与过程异质性与动态性的需要。面向对象影像分析采用影像分割技术生成多尺度影像对象，构建与地表实体相似的层次等级结构，不同等级的地学过程可在相应尺度的影像对象层中得到反映，实现了空间格局与过程的多尺度影像分析。面向对象影像分析中的尺度概念已不再是像元大小，而是指影像的分割尺度，是关于影像对象异质性最小的阈值。每一个遥感专题应用任务对于影像分析中尺度的需求是不一样的，要正确反映测区内格局与过程的变化特征，了解多尺度转换过程中影像对象信息的变化效应，选择特定类别信息提取的最优尺度成为面向对象影像分析中关注的重点。 本文在分析基于像元尺度问题研究方法的基础上，围绕面向对象影像分析中存在的尺度转换与最优尺度选择问题，开展了如下几个方面的研究： 1．提出面向对象影像分析中最优尺度的双重性：一是提取面积最小类别所需的最大分辨率，二是不同类别提取的最优分割尺度。前者指在保证反映地物类别空间结构特征情况下识别最小地物类别的最大分辨率，是包含所需信息而且数据量最小的分辨率。后者是针对于一种地物类别而言的，影像对象多边形既不能太破碎，也不能边界模糊的分割尺度，该尺度下对象大小与特定的地物目标大小接近，且类别内部对象的光谱变异较小。 2．提出以影像对象均值方差方法确定影像最优分割尺度。该方法能同时确定影像中多种地物类别的最优分割尺度，证明了不同类别有其相应的最优分割尺度，突破了最优尺度选择曲线只能适用于一个类别的限制，但该方法只适用于高分辨率影像信息提取最优尺度的选择，对于中低分辨率的影像效果不甚理想。 3．提出以影像对象最大面积方法确定影像最优分割尺度。影像对象最大面积随分 摘要 害」尺度变化的曲线呈阶梯状上升的趋势，每一个曲线平台对应于每种类别提取 适宜尺度的范围，该方法适用于不同分辨率影像的信息提取。通过曲线分析验 证了最优尺度或适宜尺度是一个连续的范围值，而不是一个断点值的观点。 4.提出三种不同的对象类型的概念:纯对象、混合对象与伪对象。影像信息提取 在纯对象占多数的尺度层提取能保证有较高的分类精度，因此选择最优分割尺 度也就是选择影像中纯对象最多而混合对象数目最少的分割尺度。 5.以影像对象为基本实体的尺度转换方法就是影像的多尺度分割。不同尺度层上 影像对象的属性信息是基于原始影像像元的整合，从而避免了重采样过程中损 失大量的有用信息。 6.提出可以用影像对象的均值、标准差、异质性系数与分形维数作为评价面向对 象尺度转换的指标，根据这四个指标随尺度的变化趋势分析尺度转换方法的合 理性。 7.全面介绍了面向对象影像分析技术。面向对象技术在遥感应用中有着无可比拟 的优势，是当前影像分析技术的发展趋势。本文回顾了面向对象影像分析中影 像分割技术与信息提取技术的方法与研究进展，详细讨论了以遥感影像分析为 目标的多尺度影像分割技术以及模糊数学分类技术。 8.面向对象技术中尺度问题应用:以大庆市城市绿地遥感调查为目标，采用面向 对象的信息提取技术及类别识别的最优尺度选择方法，成功提取了乔木、灌木、 人工草地与疏林地等绿地覆盖信息，展示了面向对象技术应用于遥感影像信息 提取的广阔前景

本文从面向对象的遥感信息提取中的尺度效应研究入手,对影像对象的分形维数、紧凑度、面积、均值、标准差和与邻域均值差分等特征进行了实验.在此基础上,根据"类内同质性大,类问异质性大"的最佳分类原则,提出了面向对象的RMAS方法,该方法的思想是,当对象RMAS值最大时,对象内部的异质性最小,对象外部的异质性最大,此时的分割尺度为类别提取的最优分割尺度.根据最优尺度下信息提取精度最高的原理.实验验证了该方法的可行性,且能获得较好的分类结果.

Image segmentation determines the accuracy of subsequent classification in remote sensing.In consideration of the research lacking of evaluation about segmentation technology and limitation of the mainly subjective method currently,quantitative method is used to select optimal segmentation scale in this paper.With object\|oriented segmentation algorithm of Definiens software,standard deviation of all pixels from an image object is used as the homogeneity measured criteria in object,absolute value of  mean difference to neighbors is served as the variable of heterogeneity between objects,weighting coefficient of the object area is also considered.With the three evaluation criteria,thinking of multispectral images,the average segmentation evaluation index is constructed.Based on the evaluation index,with goodness test method,QucikBird multispectral image is used to be researched,and the optimal segmentation scales of different surface features are concluded.At last,the average object fit index is used to verify evaluation results,and the feasibility of the evaluation method is discussed.

遥感影像分割决定了后续分类的精度,鉴于目前分割技术评价的研究缺乏且局限于主观判断的现状,以定量方法确定最优分割尺度。利用Definiens平台面向对象的分割算法,将组成对象的像素灰度值的标准差作为衡量对象内同质性的标准,用与邻域的平均差分的绝对值作为对象间的异质性度量变量,同时考虑面积权重的影响；根据上述3个评价指标,在考虑多光谱影像的基础上,构造了平均分割评价指数；基于该评价指数,以优度实验法对QuickBird多光谱影像进行了研究,并确定了不同地物类型的最优分割尺度。最后,利用平均对象匹配指数对评价结果进行了验证,并对评价方法的可行性进行了探讨。

随着遥感技术的飞速发展,通过各种传感器获取的影像种类越来越多.传统的面向像元的影像分析技术已难以适应高分辨率影像,面向对象的影像分析技术应运而 生.面向对象的影像分析技术所面临的首要问题是影像分割,如何选择合适的影像分割尺度已成为面向对象影像分析技术的关键.为了提高遥感影像分割尺度选择的 准确性,通过对影像分析建立了分割质量函数,针对不同分辨率的多源遥感影像进行分割实验,获取了最优分割尺度,最后提出了一种顾及面积和光谱的分割匹配指 数对最优分割尺度进行评价,并研究了分割尺度与亮度均值标准差、影像空间分辨率之间的对应关系.研究结果表明,构建的分割质量函数获取的最优分割尺度避免 了人工分割带来的主观性,建立的同时顾及面积和光谱因子的分割匹配指数验证了最优分割尺度选择的正确性.

Abstract: The growing population and accelerating urbanization have caused much illegal occupation of farmland, which seriously threat to national food security, social stability and economic development of our country. Farmland information extraction has become a hot issue in agricultural research field in the world. In addition, farmland mapping is closely related to the food security and is one of the most concerned issues of government departments. However, traditional technology of surveying and mapping is time consuming and labor costing, which is unable to adapt to the precise and effective information acquisition of farmland. The high resolution remote sensing imagery can provide more details of ground truth than low resolution imagery. However, the information mining in high resolution remote sensing imagery faces a big challenge caused by the complex ground environment. Farmland blocks in high resolution remote sensing imagery have various shapes, complicated texture and heterogeneous spectrum. The shape information is one of the most important content of farmland mapping. In this study, high resolution remote sensing imagery from QuickBird was used to precisely extract farmland in hilly area. And the method of farmland extraction combining multi-scale segmentation and optimal scale selection was put forward. Firstly, gradient image is generated by using Sobel gradient operator. In order to enhance the edge information and reduce the irrelevant information for farmland extraction, the multi-scale gradient images are filtered with anisotropic diffusion operator. Secondly, effective scale range of multi-scale gradient images is determined through the entropy difference analysis, which can reduce the amount of calculation of the multi-scale analysis in next stage. Thirdly, a marker driven watershed transform based on minima extension and minima imposition is applied to segment the multi-scale gradient images to produce multi-scale shape information of farmland with precious boundaries. Finally, the optimal scale identification for multi-scale segmentation is obtained by unsupervised segmentation evaluation method based on the global Moran'I and variance to automatically get the farmland block without manual intervention. The experimental results show that the multi-scale segmentation and optimal scale identification approach can be used to accurately discriminate farmland in hilly area. Farmland extraction accuracy of the proposed method is 73.06% which is 22.40% higher than the Mean-shift segmentation method with a better performance in farmland extraction in hilly area. The results can basically meet the requirements of drawing and revision for the large scale thematic mapping of farmland, showing that the proposed method can provide a technical assistance for the surveying and mapping of farmland.

农田测绘与粮食安全密切相关，高效经济的农田测绘是中国政府部门重点关注的工作之一。农田田块是农田测绘的基本要素，从遥感影像中提取农田田块信息是当前研究的热点。然而，丘陵地区农田形状不规则、光谱特征不明显导致农田信息提取困难，该文通过研究最优的农田分割尺度来提高农田田块信息提取的精度。首先，利用各向异性扩散算子在由Sobel得到的梯度图上生成多尺度梯度影像。然后，通过信息熵差异分析得到有效尺度范围。其次，利用标记分水岭算法对农田梯度影像进行分割获得多尺度农田信息。最后，利用非监督的全局评价方法在已得的有效尺度范围内确定农田提取的最优尺度，同时确定最优的农田提取结果。对比试验结果表明，该文方法能够有效地提取丘陵地区的农田田块，精度可以达到73.06%，比Mean-shift方法提取的精度高22.48%。该研究可为中国农田测绘提供技术参考。