A new technique for the segmentation of single- and multiresolution (MR) remote sensing images is proposed. To guarantee the preservation of details at fine scales, edge-based watershed is used, with automatically generated markers that help in limiting oversegmentation. For MR images, the panchromatic and multispectral components are processed independently, extracting both the edge maps and the morphological and spectral markers that are eventually fused at the highest resolution, thus avoiding any information loss induced by pansharpening. Numerical results on object layer extraction and simple classification tasks prove the proposed techniques to provide accurate segmentation maps, which preserve fine details and, contrary to state-of-the-art products, can single out objects equally well at very different scales.

<p>在覆盖玉米田的遥感图像中,玉米地块边缘区域存在大量的同物异谱现象,利用传统分割方法进行玉米地块分割时,会造成边缘出现许多非玉米小块区域,因而导致玉米种植面积统计错误。根据大面积玉米种植区域的形状分布特点,提出一种类矩形引导的玉米田种植区分割方法。首先采用最小核值相似区(smallest univale segment assimilating nucleus,SUSAN)边缘检测算子对融合后的高分一号(GF-1)卫星遥感图像进行边缘提取,然后根据闭合区域与外接类矩形的关系构建类矩形引导的相关函数,最后将类矩形阈值函数引入基于图的分割算法中实现特定形状的地块分割。将分割结果分别与基于图的分割方法、分水岭分割方法和人工解译样本进行实验比较,结果表明:本文方法能有效地分割出玉米田目标,减少了同物异谱带来的影响,分割结果更加符合玉米田实际分布特征和实际统计面积。</p>

<p>为解决地震灾区震后空间数据难以及时获取的问题,考虑到无人机(unmanned aerial vehicle,UAV)遥感影像自身特点,提出了一种改进的均值漂移(Mean Shift)分割算法。按照影像特征将UAV影像划分为纹理区和匀色区,对匀色区直接进行Mean Shift算法分割得到初分结果; 对纹理区综合提取颜色、纹理、形状信息构造高维特征空间,并根据归一化分布密度值求得其合适的带宽,再使用Mean Shift算法对特征空间进行模式分类得到分割结果; 通过构造代价函数进行区域合并,消除过分割区域,得到最终分割结果。针对芦山地震后获取的高空间分辨率UAV影像进行分割实验,并提出一种考虑到面积和光谱的分割匹配指数对分割结果进行评价。实验结果表明: 所提出的改进的Mean Shift算法的分割精度优于传统的Mean Shift算法,为后续的震害信息提取提供了数据保障。</p>

This paper investigates the problem of remote sensed image segmentation using multiple kinds ofinformation. We present a region-level tree-structured Markov random field (RTS-MRF) model forthis task, which is an improved version of the tree structured Markov Random Field (TS-MRF) modeldefined on pixels. It inherits the advantages of the TS-MRF model, i.e., the modeling power and thetechnique to use hierarchical information embedded in the image, and compensates its deficiency tomake use of the structural information by modeling image on region level. In RTS-MRF, the MRFsassociated with the tree nodes are hierarchically defined on the region adjacency graph (RGA) to makeuse of the statistical attributes of regions and their boundaries. Based on RTS-MRF, a supervised imagesegmentation algorithm is carried out, and experimental results demonstrate the better performancethan some competing algorithms.

Image analysis that produces an image-like array of symbolic or numerical elements (such as edge finding or depth map reconstruction) can be formulated as a labeling problem in which each element is to be assigned a label from a discrete or continuous label set. This formulation lends itself to algorithms, based on Bayesian probability theory, that support the combination of disparate sources of information, including prior knowledge.In the approach described here, local visual observations for each entity to be labeled (e.g., edge sites, pixels, elements in a depth array) yield label likelihoods. Likelihoods from several sources are combined consistently in abstraction-hierarchical label structures using a new, computationally simple procedure. The resulting label likelihoods are combined with a priori spatial knowledge encoded in a Markov random field (MRF). From the prior probabilities and the evidence-dependent combined likelihoods, the a posteriori distribution of the labelings is derived using Bayes' theorem.A new inference method, Highest Confidence First (HCF) estimation, is used to infer a unique labeling from the a posteriori distribution that is consistent with both prior knowledge and evidence. HCF compares favorably to previous techniques, all equivalent to some form of energy minimization or optimization, in finding a good MRF labeling. HCF is computationally efficient and predictable and produces better answers (lower energies) while exhibiting graceful degradation under noise and least commitment under inaccurate models. The technique generalizes to higher-level vision problems and other domains, and is demonstrated on problems of intensity edge detection and surface depth reconstruction.

该文提出了一种基于马尔可夫随机场(MRF)的合成孔径雷达(SAR)图像分割新方法。在传统MRF的邻域基团势函数基础上，引入了图像邻域中各个像素的强度差值以及像素之间的距离因子，使SAR图像中空间上下文信息得到了更加充分的利用。根据贝叶斯定理将图像分割问题转化为最大后验概率的求取问题，运用迭代条件模型(ICM)算法求得最大后验概率的解。在实验中，将该文提出的方法、传统上使用ICM以及模拟退火(SA)优化方法的MRF分割运用于模拟的SAR图像以及真实SAR图像。比较结果证明，该文的方法在误分率以及抗噪性上更具优势。

模糊C均值聚类是一种经典的非监督聚类模型，成功地应用于遥感影像分类。但是该方法对初始值敏感，容易陷入局部最优解；同时聚类时仅考虑光谱信息，忽略了空间信息。本文提出了一种新的基于马尔科夫随机场的模糊C均值聚类方法，该方法把马尔科夫随机场和模糊C均值结合在一起。初始值依据第一主成分的密度函数确定，这样克服了对初始值的依赖性，又在聚类的时候考虑了空间信息。通过实例数据验证，所提出的方法分类精度优于传统的模糊C均值模型。

针对图像聚类问题,提出了一种基于图像空间关系的聚类方法,采用 场模型描述图像之间的空间关系,利用K-近邻思想构建图像邻域系统,聚类过程中无需手动标记特征表示的图像类别信息,只需要给定初始类别数,通过条件迭代 算法(ICM)对图像进行聚类.该文通过实验分析了图像样本大小、图像特征维数、图像特征类型、初始类别标签对聚类结果的影响,通过与多种经典聚类算法进 行对比,实验结果充分验证了该方法的有效性.

Segmentation of multispectral remote sensing images is a key competence for a great variety of applications. Many of the applied segmentation algorithms are generative models based on Markov random fields. These approaches are generally limited to multivariate probability densities such as the normal distribution. In addition, it is usually impossible to adjust the contextual parameters separately for each frequency band. In this letter, we present a new segmentation algorithm that avoids the aforementioned problems and allows the use of any univariate density function as emission probability in each band. The approach consists of three steps: first, calculate feature vectors for every frequency band; second, estimate contextual parameters for every band and apply local smoothing; and third, merge the feature vectors of the frequency bands to obtain final segmentation. This procedure can be iterated; however, experiments show that after the first iteration, most of the pixels are already in their final state. We call our approach successive band merging (SBM). To evaluate the performance of SBM, we segment a Landsat 8 and an AVIRIS image. In both cases, the k coefficients show that SBM outperforms the benchmark algorithms.

We make an analogy between images and statistical mechanics systems. Pixel gray levels and the presence and orientation of edges are viewed as states of atoms or molecules in a lattice-like physical system. The assignment of an energy function in the physical system determines its Gibbs distribution. Because of the Gibbs distribution, Markov random field (MRF) equivalence, this assignment also determines an MRF image model. The energy function is a more convenient and natural mechanism for embodying picture attributes than are the local characteristics of the MRF. For a range of degradation mechanisms, including blurring, non-linear deformations, and multiplicative or additive noise, the posterior distribution is an MRF with a structure akin to the image model. By the analogy, the posterior distribution defines another (imaginary) physical system. Gradual temperature reduction in the physical system isolates low-energy states (090004annealing090005), or what is the same thing, the most probable states under the Gibbs distribution. The analogous operation under the posterior distribution yields the maximum a posteriori (MAP) estimate of the image given the degraded observations. The result is a highly parallel 090004relaxation090005 algorithm for MAP estimation. We establish convergence properties of the algorithm and we experiment with some simple pictures, for which good restorations are obtained at low signal-to-noise ratios.

An unsupervised segmentation approach to classification of multispectral image is suggested here in Markov random field (MRF) frame work. This work generalizes the work of Sarkar et al. [32] on gray value images for multispectral images and is extended for landuse classification. The essence of this approach is based on capturing intrinsic characters of tonal and textural regions of any multispectral image. The approach takes an initially oversegmented image and the original multispectral image as the input and defines a MRF over region adjacency graph (RAG) of the initially segmented regions. Energy function minimization associated with the MRF is carried out by applying a multivariate statistical test. A cluster validation scheme is outlined after obtaining optimal segmentation. Quantitative evaluation of classification accuracy of test data for three illustrations are shown and compared with conventional maximum likelihood procedure. Comparison of the proposed methodology with a recent work of texture segmentation in the literature has also been provided. The findings of the proposed method are found to be encouraging.

In this paper, we present three optimisation techniques, Deterministic Pseudo-Annealing (DPA), Game Strategy Approach (GSA), and Modified Metropolis Dynamics (MMD), in order to carry out image classification using a Markov random field model. For the first approach (DPA), the a posteriori probability of a tentative labelling is generalised to a continuous labelling. The merit function thus defined has the same maxima under constraints yielding probability vectors. Changing these constraints convexifies the merit function. The algorithm solves this unambiguous maximisation problem, and then tracks down the solution while the original constraints are restored yielding a good, even if suboptimal, solution to the original labelling assignment problem. In the second method (GSA), the maximisation problem of the a posteriori probability of the labelling is solved by an optimisation algorithm based on game theory. A non-cooperative n-person game with pure strategies is designed such that the set of Nash equilibrium points of the game is identical to the set of local maxima of the a posteriori probability of the labelling. The algorithm converges to a Nash equilibrium. The third method (MMD) is a modified version of the Metropolis algorithm: at each iteration the new state is chosen randomly, but the decision to accept it is purely deterministic. This is also a suboptimal technique but it is much faster than stochastic relaxation. These three methods have been implemented on a Connection Machine CM2. Experimental results are compared to those obtained by the Metropolis algorithm, the Gibbs sampler and ICM (Iterated Conditional Mode).

马尔可夫随机场方法是图像分割中一个极为活跃的研究方向。本文介绍了基于马尔可夫随机场模型的一般理论与图像的关系，给出它在图像分割中的通用框架：包括空域和小波域图像模型的建立、最优准则的选取、标号数的确定、图像模型参数的估计和图像分割的实现，评述了其在图像分割中的应用，展望其发展的方向。