《《多时相遥感影像变化检测算法研究》-公开DOC·毕业论文》由会员分享，可在线阅读，更多相关《《多时相遥感影像变化检测算法研究》-公开DOC·毕业论文（64页珍藏版）》请在金锄头文库上搜索。
1、硕士研究生学位论文新 疆 大 学 论文题目(中文)：多时相遥感影像变化检测算法研究 论文题目(外文)：Change Detections in Multi-temporal Satellite Images研 究 生 姓 名： 学 科 、 专 业：工学、信号与信息处理研 究 方 向：数字图像处理导师 姓名 职称： 教授 论文答辩日期 2011 年 5 月20日 学位授予日期 年 月 日57摘要利用多时相遥感影像获取地物变化信息的过程称之为变化检测。根据影像分析的层次不同，变化检测算法可以分为像素级、特征级和目标级这三类；根据数据分析的机理，变化检测算法可以分为有监督和无监督两类。有监督的变化检
2、测算法是基于有监督的分类方法，这种方法要求训练网络以得到网络的参数。无监督的变化检测算法用两张不同时相的遥感影像通过直接的比较而无需附加信息就可以检测出影像的变化。本文所提众多算法都是基于像素级、无监督的变化检测算法。本文提出了一种基于主分量分析和上下截集模糊Kohonen聚类网络的无监督的不同时相的遥感影像的像素级变化检测算法。该算法首次将主分量分析和上下截集模糊Kohonen聚类网络这两种方法相结合，并将它应用于不同时相的遥感影像变化检测。该方法结合每个象素的邻域信息，利用主分量分析，产生每个象素对应的基于邻域信息的特征向量；又将变化区域检测问题转化为两类之间的分类问题；然后利用上下截集模
3、糊 Kohonen 聚类网络对每个象素所对应的特征向量进行变化类与未变化类的聚类，得到变化检测图。本文又提出了一种基于非下采样Contourlet变换和脉冲耦合神经网络的无监督的不同时相的遥感影像的变化检测算法。该算法将非下采样Contourlet变换和脉冲耦合神经网络这两种方法相结合，并首次将它应用于不同时相的遥感影像变化检测。本文首次将非下采样Contourlet变换和上下截集模糊Kohonen聚类网络相结合，提出了一种无监督的多时相遥感影像变化检测算法。该算法采用非下采样Contourlet变换提取与对数比图像中的每个象素相对应的多尺度、多方向纹理，并采用上下截集模糊Kohonen聚类网
4、络将这些多尺度、多方向纹理分为变化类与未变化类两类，最终得到变化检测图。通过三个具体的变化检测算法的研究，归纳出变化检测算法一般研究思路。关键词：主分量分析；上下截集模糊 Kohonen 聚类网络；非下采样Contourlet变换；脉冲耦合神经网络；无监督变化检测；多尺度多方向；多时相遥感影像；遥感AbstractThe process of obtaining the changed information of the earth by making use of multi-temporal satellite images is called change detection. Acc
5、ording to the level of analyzing image, the change detection algorithms can be divided into pixel level class, characteristic level one and target level one. According to the mechanism of processing data, they can be divided into supervised class and unsupervised one. The kind of the supervised chan
6、ge detection algorithms are based on method of supervised classifying and require training to get the parameters of network. While the kind of the unsupervised change detection algorithms generate the change map by making a comparison of bi-temporal satellite images automatically without manual oper
7、ation. The proposed algorithms belong to the kind of unsupervised change detection algorithms in pixel level.An unsupervised change detection algorithm in multi-temporal satellite images based on principal component analysis and up-down-set fuzzy Kohonen clustering network is proposed. This method m
8、akes a combination of both PCA and UDSFKCN initially, and applies it to change detection. This method generates eigenvector corresponding to every pixel combining itself with its neighbors using principal component analysis. At the same time, solving the detection of the changed pixel in a region is
9、 to divide the pixel into two groups, changed class and unchanged class. Since every pixel is described as a eigenvector, therefore to obtain a changed map of the changed region in pixel level, up-down-set fuzzy Kohonen clustering network is applied to divide all the eigenvectors into changed ones a
10、nd unchanged ones.An unsupervised change detection algorithm in multi-temporal satellite images based on non-sub-sampled Contourlet transform and pulse coupled neural network is proposed. This method makes a combination of both non-sub-sampled Contourlet transform and pulse coupled neural network, a
11、nd applies it to change detection initially. An unsupervised multi-scale change detection algorithm in multi-temporal satellite images is also proposed. This method makes a combination of both non-sub-sampled Contourlet transform and up-down-set fuzzy Kohonen clustering network, and applies it to ch
12、ange detection initially. For each pixel in the log-ratio image, multi-scale and multi-direction feature vector is extracted using non-sub-sampled Contourlet transform. The final change detection map is achieved by clustering the multi-scale and multi-direction feature vectors using up-down-set fuzz
13、y Kohonen clustering network into two classes: changed and unchanged. Through three specific change detection algorithms, summarized the change detection algorithm for general research ideas. Keywords：Principal Component Analysis(PCA); Up-Down-Set Fuzzy Kohonen Clustering Network(UDSFKCN); Non-sub-s
14、ampled Contourlet Transform (NSCT); Pulse Coupled Neural Network (PCNN); Unsupervised Change Detection; Multi-scale and Multi-direction; Multi-temporal Satellite Images; Remote Sensing目录第一章 引言11.1 研究背景及意义11.1.1背景11.1.2意义21.2 国内图像变化检测经典算法41.2.1主分量分析法41.2.2最大类间方差法51.2.3最小二乘图像相减法61.2.4小波与FCM结合法71.2.5综合
15、特征级和像素级的两步变化检测算法81.2.6自适应空间邻域分析和瑞利-高斯模型81.3 国外图像变化检测经典算法91.3.1 基于EM和MRF的无监督变化检测算法91.3.2 基于遗传算法的变化检测算法101.3.3 基于高斯混合模型和贝叶斯推理的变化检测算法101.3.4 基于半监督的支持向量机的变化检测算法111.3.5 基于后验概率空间的变化向量分析算法111.3.6 基于多项式回归和空间邻域的变化检测算法121.3.7 基于全色融合的变化检测算法121.4 研究方案预案131.5本论文的结构安排14第二章 变化检测算法的基本理论162.1主分量分析PCA162.2上下截集模糊Kohonen聚类网络UDSFKCN162.3非降采样Contourlet变换NSCT182.3.1非降采样塔型滤波器组182.3.2非降采样方向滤波器组192.3.3 NCST系数特点202.4 脉冲耦合神经网络PCNN212.4.1脉冲耦合神经网络的基本原理212.4.2图像中的脉冲耦合神经网络设计222.5 遥感影像变化检测的一般思路23第三章 基于PCA和UDSFKCN的变化检测算法243.1 算法概述243.2 理论模型和实现算法253.3 实验结果及讨论273.4 结论29第四章 基于N
