典型遥感数据分类方法的适用性分析——以遥感图像场景分类为例

目前普遍采用的分类器通常都是针对单一或小量任务而设计的,在小数据量的处理中能取得比较满意的结果。但对于海量遥感数据的处理,其在处理时效和分类精度方面还有待研究。本文以遥感图像场景分类任务为例,着重对遥感数据分类问题中几种典型分类方法的适用性进行比较研究,包括K近邻(KNN)、随机森林(RF),支持向量机(SVM)和稀疏表达分类器(SRC)等。分别从参数敏感性,训练样本数据量,待分类样本数据量和样本特征维数对分类器性能的影响等几个方面进行比较分析。实验结果表明:(1)KNN,RF和L0-SRC方法相比RBF-SVM,Linear-SVM和L1-SRC,受参数影响的程度更弱;(2)待分类样本固定的情况下,随着训练样本数目的增加,SRC类型分类方法的分类性能最佳,SVM类型方法次之,然后是RF和KNN,在总体分类时间上呈现出L0-SRCL1-SRCRFRBF-SVM/Linear-SVMKNN/L0-SRC-Batch的趋势;(3)训练样本固定的情况下,所有分类方法的分类精度几乎都不受待分类样本数目变化的影响,RBF-SVM方法性能最佳,其次是L1-SRC,然后是Linear-SVM,最后是RF和L0-SRC/L0-SRC-Batch,在总体分类时间上,L1-SRC和L0-SRC相比其他分类方法最为耗时;(4)样本特征维数的变化不仅影响分类器的运行效率,同时也影响其分类精度,其中SRC和KNN分类器器无需较高的特征维数即可获得较好的分类结果,SVM对高维特征具有较强的包容性和学习能力,RF分类器对特征维数增加则表现得并不敏感,特征维数的增加并不能对其分类精度的提升带来更多的贡献。总的来说,在大数据量的遥感数据分类任务中,现有分类方法具有良好的适用性,但是对于分类器的选择应当基于各自的特点和优势,结合实际应用的特点进行权衡和选择,选择参数敏感性较小,分类总体时间消耗低但分类精度相对较高的分类方法。     

天文视频图像采集系统的研究

介绍了采用数字图像处理技术实现软件积分研制成的天文视频图像采集、处理系统。利用该系统 ,我们对 1 997年 3月 9日黑龙江漠河日全食的观测资料进行了处理 ,结果证明 ,该系统完全达到预期研制的目标。     

A new Likelihood Ratio for supervised classification of fully polarimetric SAR data: An application for sea ice type mapping

One of the potential applications of polarimetric Synthetic Aperture Radar (SAR) data is the classification of land cover, such as forest canopies, vegetation, sea ice types, and urban areas. In contrast to single or dual polarized SAR systems, full polarimetric SAR systems provide more information about the physical and geometrical properties of the imaged area. This paper proposes a new Bayes risk function which can be minimized to obtain a Likelihood Ratio (LR) for the supervised classification of polarimetric SAR data. The derived Bayes risk function is based on the complex Wishart distribution. Furthermore, a new spatial criterion is incorporated with the LR classification process to produce more homogeneous classes. The application for Arctic sea ice mapping shows that the LR and the proposed spatial criterion are able to provide promising classification results. Comparison with classification results based on the Wishart classifier, the Wishart Likelihood Ratio Test Statistic (WLRTS) proposed by Conradsen et al. (2003) and the Expectation Maximization with Probabilistic Label Relaxation (EMPLR) algorithm are presented. High overall classification accuracy of selected study areas which reaches 97.8% using the LR is obtained. Combining the derived spatial criterion with the LR can improve the overall classification accuracy to reach 99.9%. In this study, fully polarimetric C-band RADARSAT-2 data collected over Franklin Bay, Canadian Arctic, is used.     

合成孔径雷达森林资源监测技术研究综述

合成孔径雷达（Synthetic Aperture Radar,SAR）技术凭借其全天时、全天候的成像能力以及对森林垂直结构信息敏感的特点,在森林资源监测中具有独特的优势,已成为当前森林资源遥感调查技术的研究热点.本文首先介绍了SAR森林资源监测技术的发展背景、发展轨迹和相关知识;然后,重点阐述了极化SAR、干涉SAR、极化干涉SAR和层析SAR在林地覆盖类型分类、变化检测以及森林参数定量化估测应用中的技术方法;最后,就SAR在森林资源监测研究和应用中存在的问题与发展趋势进行了总结与展望.     

遥感影像局部兴趣目标的可逆信息隐藏算法

提出了一种适用于遥感影像局部兴趣目标授权使用的可逆信息隐藏算法。提取遥感影像上的局部兴趣目标作为待隐藏的信息,并在原始图像上对该目标进行变形处理,然后利用可逆整数线性变换实现待隐藏信息的调制嵌入,授权用户能够完全无损重建原始遥感影像。算法为盲算法,具有较大的嵌入容量,含隐藏信息图像的峰值信噪比较高,并且可以有效抵抗剪切及椒盐噪声等攻击。     

基于RFM的高分辨率卫星遥感影像自动匹配研究

摘　要：提出一种基于有理多项式模型（RFM）进行高分辨率卫星遥感影像自动匹配的方法。首先利用RFM进行高分辨率卫星影像直接定位和同名点预测;然后基于投影轨迹建立近似核线方程,并分析了核线精度;接着采用金字塔影像策略进行核线约束的近似一维影像匹配,并经最小二乘影像匹配精化匹配结果;最后采用RANSAC算法剔除误匹配点以获取最终的匹配结果。通过与二维灰度相关方法和SIFT匹配方法的比较试验,证明本文方法可靠性好、匹配成功率高,较好地解决了多时相、大姿态角高分辨率卫星遥感影像的自动匹配难题。     

基于数字图像处理技术的冰川堆积物粒度分析——以川西贡嘎山冰川堆积物为例

冰川堆积物的粒度组成在很大程度上控制了其物理力学性质,但常规试验无法得到连续的冰川堆积物的粒度组成数据.应用数字图像处理技术和室内试验相结合的分析方法,首先借助图像处理技术,对冰川堆积物剖面的砾石形态进行分析.获取粗粒部分(>2cm)的粒度组成,然后采用室内颗分试验获得细粒(<2cm)部分的粒度组成,将2类粒度数据进行叠合即可获取连续的冰川堆积物的粒度组成.结果表明:研究区冰ill堆积物的粒度组成极不均匀且颗粒分析曲线呈现多峰性,具有良好的级配和工程性质;冰川堆积物的粒度组成、砾石形态和排列特征都有很好的分形特征.     

A Practical Procedure for the Measurement of Fragmentation by Blasting by Image Analysis

Summary. The operation of a digital image analysis system in a limestone quarry is described. The calibration of the system, required in order to obtain moderately reliable fragmentation values, is done from muckpile sieving data by tuning the image analysis software settings so that the fragmentation curve obtained matches as close as possible the sieving. The sieving data have also been used to extend the fragment size distribution curves measured to sizes below the system’s optical resolution and to process the results in terms of fragmented rock, discounting the material coming from a loose overburden (natural fines) that is cast together with the fragmented rock. Automatic and manual operation modes of the image analysis are compared. The total fragmentation of a blast is obtained from the analysis of twenty photographs; a criterion for the elimination of outlier photographs has been adopted using a robust statistic. The limitations of the measurement system due to sampling, image processing and fines corrections are discussed and the errors estimated whenever possible. An analysis of consistency of the results based on the known amount of natural fines is made. Blasts with large differences in the amount of fines require a differentiated treatment, as the fine sizes tend to be the more underestimated in the image analysis as they become more abundant; this has been accomplished by means of a variable fines adjustment factor. Despite of the unavoidable errors and the large dispersion always associated with large-scale rock blasting data, the system is sensitive to relative changes in fragmentation.     

面向对象和DEM的ZY-1 02C影像湿地提取研究

本文运用面向对象分类与DEM数据相结合的方法,对资源卫星一号02C卫星遥感影像进行湿地提取。探索了基于对象与DEM信息的提取技术,在02C影像湿地的提取应用,对研究我国国产卫星在湿地监测和保护方面有重要的意义。研究结果表明：（1）面向对象的遥感影像信息提取方法,可同时兼顾影像光谱信息及空间信息,适用于02C影像的湿地提取,精度得到明显提高;（2）基于对象与DEM信息的提取方法,使沼泽地与草地相混淆的现象明显减轻,湿地分类精度进一步提高,该方法适用于高分辨率遥感影像的湿地提取研究;（3）基于对象与DEM信息提取的水田、水体、沼泽地及河滩的精度,分别为88.46%、97.44%、86.96%和83.33%,满足资源卫星一号02C遥感影像对湿地进行监测和保护的需要。