基于粗糙集的K--均值聚类算法在遥感影像分割中的应用

结合粗糙集理论和K——均值聚类算法，提出一种遥感影像的粗糙聚类分割方法。根据遥感影像中特征属性的相互依赖关系，应用粗糙集理论的等价关系。求出K——均值聚类所需要的初始类的个数和均值。然后采用聚类算法对图像进行分割。实验结果表明该方法比随机选取聚类的中心点和个数减少了运算量．提高了分类精度和准确性。     

Role of the Kazerun fault system in active deformation of the Zagros fold-and-thrust belt (Iran)

Field structural and SPOT image analyses document the kinematic framework enhancing transfer of strike-slip partitioned motion from along the backstop to the interior of the Zagros fold-and-thrust belt in a context of plate convergence slight obliquity. Transfer occurs by slip on the north-trending right-lateral Kazerun Fault System (KFS) that connects to the Main Recent Fault, a major northwest-trending dextral fault partitioning oblique convergence at the rear of the belt. The KFS formed by three fault zones ended by bent orogen-parallel thrusts allows slip from along the Main Recent Fault to become distributed by transfer to longitudinal thrusts and folds. To cite this article: C. Authemayou et al., C. R. Geoscience 337 (2005).     

活断层运动区段的划分问题

断层运动是复杂而多变的,在一条巨大的断裂带上经常以区段为单位活动,每次活动的区段长度、运动量等都不相同,且往往具有一定的迁移规律等。为工程建设和强震的预测,划分活断层运动区段越来越迫切。以一次强震活动序列、跨断层大地测量网的复测、直接鉴定断层新活动、并结合地形地貌和地下水位变化等途径来划分断层运动的区段     

一种基于人工免疫的图像分割算法

图像二维熵分割,一直因耗时长而限制了实际应用。本文借鉴生物免疫思想,提出二维熵图像分割的人工免疫算法。在克隆选择算法中引入疫苗的免疫接种,用于优化最优分割阈值对的搜索过程。在遥感高分辨率图像上的实验显示,该算法不仅能准确搜索到最优阈值对,而且计算时间只有传统算法的1.8%。该算法也验证了人工免疫思想用于图像分割的可行性和有效性。     

标记分水岭算法及区域合并的遥感图像分割

传统的分水岭算法通常对梯度图像进行无标记分割，其结果是容易造成过度分割。本文采用了一种基于标记的分水岭算法，首先,利用Sobel边缘算子对原遥感图像进行梯度重建，获得梯度幅值图像，同时计算待分割区域的周长、面积和形态因子，并对其进行标记； 然后,利用距离函数图标定种子法和等值线跟踪法获得初始分割图像； 最后，利用改进的区域合并方法获得最终的分割结果。实验结果表明了该方法的有效性。     

分块DPCM与5/3整数小波变换结合的天文图像无损压缩

针对海量的天文图像数据与有限的存储空间和带宽资源之间矛盾日益突出这一问题,提出一种无损压缩方法,首先将超大天文图像分块,再使用差分脉冲编码调制和5/3整数小波变换,最后使用霍夫曼算法编码。对该方法的原理和具体实现做了详细的分析与介绍,通过实验验证该方法比天文中常用的tar、PKZip、WinZip、WinRar软件在压缩比上分别提高了30%、29%、26%、2%,压缩速度远大于WinZip和WinRar;且该算法实现简单,适合硬件实现和利于并行处理。     

An automated approach to vertical road characterisation using mobile LiDAR systems: Longitudinal profiles and cross-sections

The characterisation the vertical profiles and cross-sections of roads is important for the verification of proper construction and road safety assessment. The goal of this paper is the extraction of geometric parameters through the automatic processing of mobile LiDAR system (MLS) point clouds. Massive and complex datasets provided by the MLS are processed using a hierarchical strategy that includes segmentation, principal component analysis (PCA)-based orthogonal regression, filtering and parameter extraction procedures. Best-fit geometric parameters act as a vertical road model for both linear parameters (slope and vertical curves) and cross-sections (superelevations). The proposed automatic processing approach gives satisfactory results for the analysed scenario.     

Fusion of airborne laserscanning point clouds and images for supervised and unsupervised scene classification

Automatic urban object detection from airborne remote sensing data is essential to process and efficiently interpret the vast amount of airborne imagery and Laserscanning (ALS) data available today. This paper combines ALS data and airborne imagery to exploit both: the good geometric quality of ALS and the spectral image information to detect the four classes buildings, trees, vegetated ground and sealed ground. A new segmentation approach is introduced which also makes use of geometric and spectral data during classification entity definition. Geometric, textural, low level and mid level image features are assigned to laser points which are quantified into voxels. The segment information is transferred to the voxels and those clusters of voxels form the entity to be classified. Two classification strategies are pursued: a supervised method, using Random Trees and an unsupervised approach, embedded in a Markov Random Field framework and using graph-cuts for energy optimization. A further contribution of this paper concerns the image-based point densification for building roofs which aims to mitigate the accuracy problems related to large ALS point spacing.Results for the ISPRS benchmark test data show that to rely on color information to separate vegetation from non-vegetation areas does mostly lead to good results, but in particular in shadow areas a confusion between classes might occur. The unsupervised classification strategy is especially sensitive in this respect. As far as the point cloud densification is concerned, we observe similar sensitivity with respect to color which makes some planes to be missed out, or false detections still remain. For planes where the densification is successful we see the expected enhancement of the outline.     

A comparison between different methods for determining grain distribution in coarse channel beds

The determination of grain size distribution in alluvial channels plays a crucial role in understanding fluvial dynamics and processes (e.g., hydraulic resistance, sediment transport and erosion, and habitat suitability). However, to determine an accurate distribution, tremendous field efforts are often required. Traditionally, the grain size distribution of channel beds have been obtained by manually counting a set of randomly selected stones (the“pebble count”). Based on this elementary principle, many authors have proposed different adaptations to overcome weaknesses and problems with the original method; with the development of digital technology, photographic methods have been developed in order to sig-nificantly reduce the time spent in the field. Two of these“image-assisted”methods include Automated Grain Sizing, AGS, and Manual Photo Sieving, MPS. In this study, AGS and MPS were applied under ideal laboratory conditions, to be used as reference, and in two field conditions with different degrees of difficulty in terms of visual determination of the grain size distribution; these included an artificial unlined channel and two natural mountainous streams. The results were compared with those obtained with the pebble-count method. In general, strong agreement between the methods was found when they were applied under favorable conditions (”the laboratory”), and the differences between the image-assisted and pebble count methods were similar to those found in previous studies. Despite being more time consuming, MPS was deemed preferable to AGS when conditions are not optimal;in these cases, the time spent on image elaboration significantly increased in the AGS method (approximately three-fold), but the estimation error of the median grain size decreased by approximately 37%. The use of image-assisted analysis has proven to be robust for characterizing sediment in watercourse beds and reducing fieldwork time, but because field conditions can significantly affect the accuracy of results, the method choice must be carefully considered.     

Stability of Kerogen Classification with Regard to Image Segmentation

This paper investigates the stability of an automatic system for classifying kerogen material from images of sieved rock samples. The system comprises four stages: image acquisition, background removal, segmentation, and classification of the segmented kerogen pieces as either inertinite or vitrinite. Depending upon a segmentation parameter d, called “overlap”, touching pieces of kerogen may be split differently. The aim of this study is to establish how robust the classification result is to variations of the segmentation parameter. There are two issues that pose difficulties in carrying out an experiment. First, even a trained professional may be uncertain when distinguishing between isolated pieces of inertinite and vitrinite, extracted from transmitted-light microscope images. Second, because manual labelling of large amount of data for training the system is an arduous task, we acquired the true labels (ground truth) only for the pieces obtained at overlap d=0.5. To construct ground truth for various values of d we propose here label-inheritance trees. With thus estimated ground truth, an experiment was carried out to evaluate the robustness of the system to changes in the segmentation through varying the overlap value d. The average system accuracy across values of d spanning the range from 0 to 1 was 86.5%, which is only slightly lower than the accuracy of the system at the design value of d=0.5 (89.07%).