Genetic feature selection for texture classification

This paper presents a novel approach to feature subset selection using genetic algorithms. This approach has the ability to accommodate multiple criteria such as the accuracy and cost of classification into the process of feature selection and finds the effective feature subset for texture classification. On the basis of the effective feature subset selected, a method is described to extract the objects which are higher than their surroundings, such as trees or forest, in the color aerial images. The methodology presented in this paper is illustrated by its application to the problem of trees extraction from aerial images.     

A hopfield neural network algorithm for automatic name placement for point feature

This paper presents a method of adding label to the map especially for the point feature. This method overcomes the shortcoming of traditional methods, e.g.Conflict-Backtracking method. Its kernel algorithm use the hopfield neural network to find the best label position for point feature. The experimental resuits proves that this algorithm has good permanence and high speed.     

Automatic Road Change Detection and GIS Updating from High Spatial Remotely-Sensed Imagery

This paper presents a tramework for road network change detectlon In order to upctate the Canadian National Topographic DataBase (NTDB). The methodology has been developed on the basis of road extraction from IRS-pan images (with a 5.8 m spatial resolution) by using a wavelet approach. The feature matching and conflation techniques are used to road change detection and updating. Elementary experiments have showed that the proposed framework could be used for developing an operational road database updating system.     

基于改进的模糊c-均值聚类方法遥感影像分类研究(英文)

Classification is always the key point in the field of remote sensing. Fuzzy c-Means is a traditional clustering algorithm that has been widely used in fuzzy clustering. However, this algorithm usually has some weaknesses, such as the problems of falling into a local minimum, and it needs much time to accomplish the classification for a large number of data. In order to overcome these shortcomings and increase the classification accuracy, Gustafson-Kessel （GK） and Gath-Geva （GG） algorithms are proposed to improve the traditional FCM algorithm which adopts Euclidean distance norm in this paper. The experimental result shows that these two methods are able to detect clusters of varying shapes, sizes and densities which FCM cannot do. Moreover, they can improve the classification accuracy of remote sensing images.     

数字地形建模的真三维TIN算法研究(英文)

The problem of taking an unorganized point cloud in 3D space and fitting a polyhedral surface to those points is both important and difficult. Aiming at increasing applications of full three dimensional digital terrain surface modeling, a new algorithm for the automatic generation of three dimensional triangulated irregular network from a point cloud is proposed. Based on the local topological consistency test, a combined algorithm of constrained 3D Delaunay triangulation and region-growing is extended to ensure topologically correct reconstruction. This paper also introduced an efficient neighboring triangle location method by making full use of the surface normal information. Experimental results prove that this algorithm can efficiently obtain the most reasonable reconstructed mesh surface with arbitrary topology, wherein the automatically reconstructed surface has only small topological difference from the true surface. This algorithm has potential applications to virtual environments, computer vision, and so on.     

Unsupervised Change Detection in Multitemporal SAR Images Using MRF Models

An unsupervised change-detection method that considers the spatial contextual information in a log-ratio difference image generated from multitemporal SAR images is proposed. A Markov random filed (MRF) model is particularly employed to exploit statistical spatial correlation of intensity levels among neighboring pixels. Under the assumption of the independency of pixels and mixed Gaussian distribution in the log-ratio difference image, a stochastic and iterative EM-MPM change-detection algorithm based on an MRF model is developed. The EM-MPM algorithm is based on a maximiser of posterior marginals (MPM) algorithm for image segmentation and an expectation-maximum (EM) algorithm for parameter estimation in a completely automatic way. The experiment results obtained on multitemporal ERS-2 SAR images show the effectiveness of the proposed method.     

Model of Land Suitability Evaluation Based on Computational Intelligence

A novel model of land suitability evaluation is built based on computational intelligence (CI). A fuzzy neural network (FNN) is constructed by the integration of fuzzy logic and artificial neural network (ANN). The structure and process of this network is clear. Fuzzy rules (knowledge) are expressed in the model explicitly, and can be self-adjusted by learning from samples. Genetic algorithm (GA) is employed as the learning algorithm to train the network, and makes the training of the model efficient. This model is a self-learning and self-adaptive system with a rule set revised by training.     

Raster voronoi tessellation and its application to emergency modeling

Due to the advances in Web technologies,various raster maps are available through Web Map Services such as Google maps and Yahoo maps.These online maps are used to visualize diverse types of disasters.Understanding disasters with these online maps has become an important research issue.In this article,we propose a map-based general-purpose emergency management support system based on a computational model of generalized(multiplicatively weighted,order-k,and Minkowski-metric)Voronoi diagrams.The proposed system tessellates Web maps and models disasters(or emergency response units)having different weights in the complete order from 1 to k-1 in the three popular Minkowski metrics(Euclidean,Manhattan,and Maximum distance)pro-vide insightful information for various what-if emergency scenarios.The proposed map-based emergency management support system systematically supports neighboring queries,districting queries,location optimization queries,and routing queries.We pro-vide specific examples to illustrate how our system supports these queries.     

A spatially weighted degree model for network vulnerability analysis

Using degree distribution to assess network vulnerability represents a promising direction of network analysis.However,the traditional degree distribution model is inadequate for analyzing the vulnerability of spatial networks because it does not take into consideration the geographical aspects of spatial networks.This paper proposes a spatially weighted degree model in which both the functional class and the length of network links are considered to be important factors for determining the node degrees of spatial networks.A weight coefficient is used in this new model to account for the contribution of each factor to the node degree.The proposed model is compared with the traditional degree model and an accessibility-based vulnerability model in the vulnerabil-ity analysis of a highway network.Experiment results indicate that,although node degrees of spatial networks derived from the tra-ditional degree model follow a random distribution,node degrees determined by the spatially weighted model exhibit a scale-free distribution,which is a common characteristic of robust networks.Compared to the accessibility-based model,the proposed model has similar performance in identifying critical nodes but with higher computational efficiency and better ability to reveal the overall vulnerability of a spatial network.     

A 3-dimensional data model for visualizing cloverleaf junction in a city model

The research work has been seldom done about cloverleaf junction expression in a 3-dimensional city model (3DCM). The main reason is that the cloverleaf junction is often in a complex and enormous construction. Its main body is bestraddle in air,and has aerial intersections between its parts. This complex feature made cloverleaf junction quite different from buildings and terrain, therefore, it is difficult to express this kind of spatial objects in the same way as for buildings and terrain. In this paper,authors analyze spatial characteristics of cloverleaf junction, propose an all-constraint points TIN algorithm to partition cloverleaf junction road surface, and develop a method to visualize cloverleaf junction road surface using TIN. In order to manage cloverleaf junction data efficiently, the authors also analyzed the mechanism of 3DCM data management, extended BLOB type in relational database, and combined R-tree index to manage 3D spatial data. Based on this extension, an appropriate data structur     

A fast clonal selection algorithm for feature selection in hyperspectral imagery

Clonal selection feature selection algorithm (CSFS) based on clonal selection algorithm (CSA), a new computational intelligence approach, has been proposed to perform the task of dimensionality reduction in high-dimensional images, and has better performance than traditional feature selection algorithms with more computational costs. In this paper, a fast clonal selection feature selection algorithm (FCSFS) for hyperspectral imagery is proposed to improve the convergence rate by using Cauchy mutation instead of non-uniform mutation as the primary immune operator. Two experiments are performed to evaluate the performance of the proposed algorithm in comparison with CSFS using hyperspectral remote sensing imagery acquired by the pushbroom hyperspectral imager (PHI) and the airborne visible/infrared imaging spectrometer (AVIRIS), respectively. Experimental results demonstrate that the FCSFS converges faster than CSFS, hence providing an effective new option for dimensionality reduction of hyperspectral remote sensing imagery.     

我国省级空间数据库的一种构建方案(英文)

In order to provide a provincial spatial database, this paper presents a scheme for spatial database construction to meet the needs of China. The objective and overall technical route of spatial database construction are described. The logical and physical database models are designed. Key issues are addressed, such as integration of multi-scale heterogeneous spatial databases, spatial data version management based on metadata and integrative management of map cartography and spatial database. Supported by the 863 High Technology Program of China (No. 2007AA12Z214), the National Natural Science Foundation of China (No. 40601083) and the National Key Basic Research and Development Program of China ( No. 2004CB318206).     

Theory and realization of GPS orbit integration

The algorithm of using one-day-arc to integrate n-days-arc by orbit overlaying is discussed in detail. An example is given, which proves that the orbit integration method can improve the precision of the orbit efficiently, especially in the determination of a local area＇s orbit.     

高光谱矿物填图技术与应用研究

回顾和总结了"九五"以来,中国国土资源航空物探遥感中心在高光谱矿物识别和矿物填图领域所取得的成果,包括主要岩矿光谱特征与影响因素分析;矿物的种类识别、丰度反演和成分识别;中—热红外多/高光谱的数据处理和矿物识别;高光谱矿物识别的地质应用模式、矿物填图的工作方法和技术流程、矿物填图的技术体系;矿物填图在区域地质找矿、矿区勘查和成矿与找矿模型、植物地球化学探测、矿山环境监测以及月球和行星探测等方面的应用示范。最后,提出了高光谱矿物填图近期的研究方向。     

利用流形学习进行高光谱遥感影像的降维与特征提取

基于最新的非线性降维方法——流形学习的理论,从高光谱遥感数据内在的非线性结构出发,采用全局化的等距映射(Isomap)方法进行降维,取得了优于常用的MNF方法的结果。把光谱角和光谱信息散度与测地距离相结合用于Isomap算法,结果在冗余方差和光谱规范化特征值方面优于采用传统欧氏距离计算邻域的Isomap方法。实验表明,流形学习是一种有效的高光谱遥感数据特征提取方法。     

Band Selection and Dimension Estimation for Hyperspectral Imagery—a New Approach Based on Invasive Weed Optimization

Currently, hyperspectral images have potential applications in many scientific areas due to the high spectral resolution. Extracting suitable and adequate bands/features from high dimensional data is a crucial task to classify such data. To overcome this issue, dimension reduction techniques have direct effects to improve the efficiency of classifiers on hyperspectral images. One common approach for decreasing the dimensionality is the feature/band selection by considering the optimum dimensionality of the hyperspectral imagery. In this paper, a new method was proposed to select optimal band for classification application, based on a metaheuristic Invasive Weed Optimization (IWO) algorithm. In this regard, the K-nearest neighbour (K-NN) technique was used as the classifier. Moreover, as a by-product of our band selection method, a new method was proposed to estimate an optimum dimension of the reduced hyperspectral images for better classification. Experimental results over three real-world hyperspectral datasets clearly showed that the proposed IWO-based band selection algorithm of this study led to the significant progress in selecting suitable bands for classification applications and estimation of optimum dimensionality of these datasets. In this regard, the overall accuracy (OA) of classification of the proposed IWO-based band selection algorithm was 92.02, 93.57, and 89.72 % for each dataset, respectively. Moreover, results reveal the superiority of the proposed IWO-based band selection algorithm against the other algorithms including GA, SA, ACO, and PSO for band selection purpose.     

一种面向网络发布的海洋数据库引擎(英文)

To meet the requirements of efficient management and web publishing for marine remote sensing data, a spatial database engine, named MRSSDE, is designed independently. The logical model, physical model, and optimization method of MRSSDE are discussed in detail. Compared to the ArcSDE, which is the leading product of Spatial Database Engine, the MRSSDE proved to be more effective.     

粗糙集理论进行遥感图像监督分类的样本质量评价的研究(英文)

In the supervised classification process of remotely sensed imagery, the quantity of samples is one of the important factors affecting the accuracy of the image classification as well as the keys used to evaluate the image classification. In general, the samples are acquired on the basis of prior knowledge, experience and higher resolution images. With the same size of samples and the same sampling model, several sets of training sample data can be obtained. In such sets, which set reflects perfect spectral characteristics and ensure the accuracy of the classification can be known only after the accuracy of the classification has been assessed. So, before classification, it would be a meaningful research to measure and assess the quality of samples for guiding and optimizing the consequent classification process. Then, based on the rough set, a new measuring index for the sample quality is proposed. The experiment data is the Landsat TM imagery of the Chinese Yellow River Delta on August 8th, 1999. The experiment compares the Bhattacharrya distance matrices and purity index zl and △x based on rough set theory of 5 sample data and also analyzes its effect on sample quality.     

基于混合式特征选择的高分五号影像农田识别

精准农田识别是农作物估产和粮食安全评估的基础。遥感数据作为农田识别的重要数据源,可提供动态、快速的监测结果。高光谱数据在农田识别分类方面具有巨大的应用潜力,但其中的冗余波段影响了分类效率和分类精度。因此,本研究提出了一种适用于高光谱数据农田分类的混合式特征选择算法。首先,基于变量的重要性排序或约束程度,按步长逐步进行降维;其次,寻找分类精度骤减的转折点,并将其对应的变量作为特征子集;最后,利用序列后向选择SBS（Sequential Backward Selection）方法搜索最优分类特征子集。本研究利用GF-5高光谱数据,共研究了3种降维方法（随机森林RF（Random Forest）、互信息MI（Multi-Information）和L1正则化（L1 regularization））和3种分类算法（随机森林、支持向量机SVM（Support Vector Machine）和K近邻KNN（K-Nearest Neighbor））的组合在农田分类中的表现。结果表明,基于L1正则化法得到的特征子集自相关性较低,并且包含的红边和近红外波段有效提高了农田、森林和裸土的区分度。在不同分类模型比较中发现,SVM在高维空间中表现出非常好的抗噪能力,分类精度高于RF和KNN。而RF在低维空间中的泛化能力要高于SVM和KNN。相比于第一步降维得到的特征子集,使用SBS搜索得到的最优特征子集均提高了分类精度。最终,具有23维输入的L1-SVM-SBS分类模型得到了最高的总体分类精度（94.64%）和农田召回率（95.83%）。本研究为高光谱数据特征优选提供了一种新思路,筛选出了更具代表性的特征波段,提高了农田分类精度,对高光谱遥感分类研究具有参考价值。     

基于距离限制的地面点提取滤波器(英文)

The extraction of points on the bare earth from point clouds acquired by airborne laser scanning is one of the most important steps for the generation of digital terrain models (DTM). This process is called “filtering”. However, most of the current filters erode the bare earth in steep sloped landscapes and at discontinuities, and they retain low vegetation. Therefore, a new filtering method for extracting ground points based on a distance limit is proposed in this paper. The angle criterion is used to assure the robustness of the algorithm. The experimental results show that the proposed filtering method can effectively derive the ground points from point clouds in complex urban areas. Supported by the Program for Changjiang Scholars and Innovative Research Team in University (No.0438), the National 863 Program of China (No. 2006AA12Z151).