Doris应用大全

-Doris应用大全FFTW的安装FFTW是由MIT的Matteo Frigo和Steven G. Johnson. 开发的，可用来进展快速傅立叶变换。目前最新版本：v 相关 翻开终端，进入FFTW路径，依次执行：./configure -prefi*=pwd -enable-floatmake make install完成安装。1.1 Doris的安装目前，Doris的最新版本为v3.17，下载地址：解压后进入src路径cd /doris/src，依次执行： ./configuremakemake install完成安装。1.2 SARtools和ENVISAT_TOOLS的安装分别进入各自路径，执行：make make install完成安装1.3 getorb的安装假设linu*的版本为fedora4,则需首先安装g77利用getorb可以获得ERS/ENVISAT的准确轨道，Doris的M_PORBITS 和S_PORBITS要调用getorb 目前最新版本v 相关 进入getorb路径，依次执行：makemake install完成安装1.4 snaphu的安装Stanford开发的解缠工具，被在Doris中的UNWRAP调用，目前最新版本v 相关 进入snaphu的src路径，依次执行：makemake install完成安装。1.5 GMT的安装2. 软件运行2.1 设置环境变量在fedora默认的bash下，设置如下环境变量：e*port PAGER=lesse*port EDITOR=vie*port EDITOR=gedit /为日前方便，可直接在./etc/bashrc中参加上述两行以永久性设置环境变量。2.2 runDoris为方便用户，可用run命令生成初始化文件。2.2.1在终端输入run g在当前路径下生成如下几个文件：Input.m_inital;input.s_intial;input.resample;input.coregistration;input.products;正常时在终端返回以下信息：*-andylocalhost $ run -grun: *INFO* generation of inputfiles finished.run: *INFO* inputdir: Inputfilesrun: *INFO* outputdir (stout): Outinforun: *INFO* outputdir (matrices): Outdatarun: *INFO* Using e*ecutable in path: dorisrun: *TIP* run -e1; run -s1-*输入run e1调用gedit编辑Input.m_inital(主影像初始化文件)Input.m_inital主要实现以下四步：读取SLC文件-m_readfiles数据剪裁及格式转化m_crop过采样可选m_ovs确定准确轨道m_porbits在Input.m_inital中要给出SLC文件volume,leader、data file的路径以及Delft轨道数据文件的路径。I在图像剪裁选项卡中，需要注意：# -# CROP (master)# -M_DBOW 1 5000 1 1000 / database output window假设要处理整幅影像时需要将上面的选项注释掉。# -# OVERSAMPLE (master)# -M_OVS_OUT Outdata/master_ovs.raw / 输出文件名称M_OVS_OUT_FORMAT ci2 /过采样影像格式 ci2 | cr4.M_OVS_FACT_RNG 2 /距离向过采样率可采用其他值M_OVS_FACT_AZI 1 / 方位向过采样率可采用其他值M_OVS_KERNELSIZE 16 / 过采样所采用插值核的长度Raffaele Nutricato曾将距离向采样率设为4来进展多时相分析multitemporal anylsys 效果不错。对于PS-Insar，两个方向都设为2比拟合理Bert Kamps# -# PORBITS (master)# -从DEOS获得的轨道数据为小写的时，需要改成大写。即将odr.*换成ODR.*否则getorb将无法读取这跟系统有关。getorb提供了两种确定轨道的方法三次自然样条插值和三次多项式插值。采用三次自然样条插值时在第一行纪录前最后一行纪录后至少有3个数据点；采用三次多项式插值时，时间间隔可取20-30秒，额外时间可取200秒。Doris默认样条插值。c -for polyfit(3-c M_ORB_E*TRATIME 300 / Time before first linec M_ORB_INTERVAL 30 / 时间间隔c M_ORB_DUMP 0.05 / dump to ascii filec -for spline-M_ORB_E*TRATIME 6 / Time before first lineM_ORB_INTERVAL 1 / 时间间隔。在终端输入run s1执行Input.m_inital中设定的步骤。正常时在终端返回以下信息：*-andylocalhost $ run -s1PROGRESS: Interpretation inputoptionsfile finished.total cpu: 0 min 0 secPROGRESS: Finished initializationPROGRESS: Start M_READFILES.PROGRESS: readvolume finished.PROGRESS: readleader finished.PROGRESS: readdat (header info) finished.PROGRESS: Finished M_READFILES.total cpu: 0 min 0.01 secPROGRESS: Start M_PORBITS.(getorb: reading arclist /usr/local/doris/delftorbits/ers2/dgme04/arclist)(getorb: orbit file /usr/local/doris/delftorbits/ers2/dgme04/ODR.112)PROGRESS: getorb: program finished ok.PROGRESS: Finished M_PORBITS.total cpu: 0 min 0.01 secPROGRESS: Orbit: interpolation coefficients computed.PROGRESS: Start M_CROP.PROGRESS: Start cropping slc data.PROGRESS: WRITESLC: 0%PROGRESS: WRITESLC: 10%PROGRESS: WRITESLC: 20%PROGRESS: WRITESLC: 30%PROGRESS: WRITESLC: 40%PROGRESS: WRITESLC: 50%PROGRESS: WRITESLC: 60%PROGRESS: WRITESLC: 70%PROGRESS: WRITESLC: 80%PROGRESS: WRITESLC: 90%PROGRESS: WRITESLC: 100%PROGRESS: Finished M_CROP.total cpu: 0 min 2.21 secPROGRESS: calling preview for cropped masterPROGRESS: Start PREVIEW generation.PROGRESS: SUNraster file created of: Outdata/master.raw (see also file: ./master_mag.ras.sh)PROGRESS: Start M_OVS.PROGRESS: OVERSAMPLESLC: 0%(#)Doris software, $Revision: 3.11 $, $Author: kampes $cp*fiddle: INFO: Number of output pi*els: 500PROGRESS: OVERSAMPLESLC: 10%PROGRESS: OVERSAMPLESLC: 20%PROGRESS: OVERSAMPLESLC: 30%PROGRESS: OVERSAMPLESLC: 40%PROGRESS: OVERSAMPLESLC: 50%PROGRESS: OVERSAMPLESLC: 60%PROGRESS: OVERSAMPLESLC: 70%PROGRESS: OVERSAMPLESLC: 80%PROGRESS: OVERSAMPLESLC: 90%PROGRESS: OVERSAMPLESLC: 100%PROGRESS: Finished M_OVS.total cpu: 0 min 17.99 secPROGRESS: calling preview for oversampled masterPROGRESS: Start PREVIEW generation.PROGRESS: SUNraster file created of: Outdata/master_ovs.raw (see also file: ./master_ovs_mag.ras.sh)total cpu: 0 min 17.99 sec .Any given program, when running, is obsolete. - WARNING SUMMARY -There were no messages. view with: run -v1 NOTE: before run -s1, you can do run -ql to get a quicklook processing. which will run step 1 and 2 if not done already.-*2.2.4输入run e2调用gedit编辑Input.s_inital(副影像初始化文件)与run e1同2.2.5 输入run s2执行I与run s1同2.2.6输入run e3调用gedit编辑Input.coregistration(匹配的初始化文件)Input.coregistration主要实现以下四步：基于轨道的粗匹配coarseorb像素级匹配coarsecorr主影像方位向滤波m_filtazi副影像方位相滤波s_filtazi距离向滤波filtrange可选首先根据获取主副影像的轨道进展匹配coarseorb，没有输入参数，精度30像素左右。然后进展像素级匹配coarsecorr有两种方法可供选择，一种是在空间域space domain进展的匹配magspace。要求窗口尺寸为奇数，设为偶数时会自动转化为奇数。这样做只是为了执行起来方便，因为窗口移动的最小尺寸为1个像素。另一种是在频率域进展的匹配magfft。我们知道，空间域的卷积运算与频率域的乘法运算是等价的，通过FFT转化到频率域进展相关性的计算，效率较高。Doris缺省方法为magfftInput.coregistration中假设干参数的设定：# -# COARSE COREGISTRATION# -CC_METHOD magfft /缺省方法c CC_METHOD magspace / CC_NWIN 21 / 窗口数量至少为5CC_WINSIZE 256 256 / 窗口尺寸c CC_INITOFF 0 0 / 如果没有准确轨道数据，则启用该项设定为0 0CC_INITOFF orbit / 利用基于轨道匹配的结果。窗口数和窗口尺寸的设定，没有什么规律，按默认的也可以，但对于所处理的影像而言，往往不是最正确的。注意：这一步完成后，查看一下logfile，如果初始偏移量与估计值的差异超过1(对ERS-1/2而言，1是正常的则需要将窗口尺寸增大。# -# AZIMUTH FILTERING# -对主影像进展方位向滤波，通过滤波主副影像不相重叠的局部将被滤掉。在SAR处理过程中如果主副影像采用的多普勒质心频率不一样则会产生不相重叠的局部。本步最好在粗匹配和精匹配之间进展。这样粗匹配得到的方位向的偏移量可用来确定多普勒质心频率的多项式，并有助于精匹配获得较好的结果。c AF_METHOD /AF_BLOCKSIZE 1024 / fftlength each columnLength of fft per buffer in azimuth direction. In general, the larger the better.AF_OVERLAP 64 / hbsHalf of the overlap between consecutive bufferes in azimuth direction. Partiallythe same data is used to estimate the spectrum, which might have certainadvantages. However it has not been studied yet if taking an overlap isrequird. Setting this card to 0 is fastest.AF_HAMMING 0.75 / The weighting of the spectrum in azimuth direction. The filtered output spectrumis first de-weighted with the specified hamming flter, then re-weightedwith a (newly centered) one.AF_OUT_FORMAT cr4 /以上参数按默认值即可# -# RANGE FILTERING (master&slave)# -RF_METHOD porbits / based on orbit geometryRF_SLOPE 0.0 / deg and terrain slopeRF_FFTLENGTH 1024 / power of 2RF_HAMMING 0.75 / alphaRF_OUT_FORMAT cr4 /以上参数按默认值即可#- FINE COREGISTRATION# -该步主要在亚像素精度的水平上计算主副影像对应点位的偏移矢量。首先在分布于整幅影像中的大量窗口中，对于给定窗口，在整幅影像范围内通过计算能量影像magnitude images的相干系数来估计偏移量精度约1个像素，其次，对于上一步所计算的最大相关系数的窗口，进展过采样，以在亚像素的水平上确定相干系数最大值。重复进展上述步骤。下面举个例子说明过采样距离向过采样的实现方法：影像是按行读入的，过采样时也按行进展。定义：line对应于方位向，慢时间；pi*el对应于距离向，快时间给定输入信号，及过采样率假设取为3首先，生成一个用填充的与输入信号构造一致的复本：然后，将其与插值核sinc函数作卷积，生成输出信号：，其中表示新的采样。c FC_METHOD magfft / fastFC_METHOD oversample / best: oversample comple* datac FC_METHOD magspace / same as magfftFC_NWIN 601 / number of smaller windowsc FC_IN_POS highcorr.pos / filename with window positionsFC_WINSIZE 64 64 / size of windowsFC_ACC 8 8 / search window, 2nFC_INITOFF coarsecorr / use result of coarseFC_OSFACTOR 32 / oversampling factorFC_PLOT 0.2 NOBG / GMT plot with magnitude backgroundc /c /STOP 窗口数和窗口尺寸当采用默认值效果不理想时，应根据数据的具体情况进展调整，可以考虑将窗口数增大一倍个人经历；窗口尺寸建议使用6464FC_ACC为最大相干系数的搜索精度，建议为8，8，这样搜索区域就为8到8,采用FFT方法时必须为2的幂。FC_OSFACTOR为采样因子采样率，要想到达1/10之一的匹配精度，建议取为32FC_PLOT为GMT的绘图参数，将绘制大于所给参数的偏移矢量。2.2.6输入run s3将执行Input.coregistration中设定的步骤。正常时，在终端返回以下信息：rootlocalhost doris_tailand# run -s3PROGRESS: Interpretation inputoptionsfile finished.total cpu: 0 min 0 secPROGRESS: Finished initializationPROGRESS: Orbit: interpolation coefficients computed.PROGRESS: Orbit: interpolation coefficients computed.PROGRESS: Start COARSE_ORB.PROGRESS: Coarse precise orbits coregistration finished.PROGRESS: Finished COARSE_ORB.total cpu: 0 min 0.04 secPROGRESS: Start COARSE_CORR.PROGRESS: COARSE_CORR: 0%PROGRESS: COARSE_CORR: 10%PROGRESS: COARSE_CORR: 20%PROGRESS: COARSE_CORR: 30%PROGRESS: COARSE_CORR: 40%PROGRESS: COARSE_CORR: 50%PROGRESS: COARSE_CORR: 60%PROGRESS: COARSE_CORR: 70%PROGRESS: COARSE_CORR: 80%PROGRESS: COARSE_CORR: 90%PROGRESS: COARSE_CORR: 100%PROGRESS: Coarse coregistration based on correlation finished.PROGRESS: Finished COARSE_CORR.total cpu: 0 min 29.35 secPROGRESS: Start FILTAZI (master).PROGRESS: FILTAZI: 0%PROGRESS: FILTAZI: 10%PROGRESS: FILTAZI: 20%PROGRESS: FILTAZI: 30%PROGRESS: FILTAZI: 40%PROGRESS: FILTAZI: 50%PROGRESS: FILTAZI: 60%PROGRESS: FILTAZI: 70%PROGRESS: FILTAZI: 80%PROGRESS: FILTAZI: 90%PROGRESS: Finished FILTAZI for master.total cpu: 2 min 18.9 secPROGRESS: Start FILTAZI (slave).PROGRESS: FILTAZI: 0%PROGRESS: FILTAZI: 10%PROGRESS: FILTAZI: 20%PROGRESS: FILTAZI: 30%PROGRESS: FILTAZI: 40%PROGRESS: FILTAZI: 50%PROGRESS: FILTAZI: 60%PROGRESS: FILTAZI: 70%PROGRESS: FILTAZI: 80%PROGRESS: FILTAZI: 90%PROGRESS: Finished FILTAZI slave.total cpu: 4 min 4.84999 secPROGRESS: Start FILTRANGE (porbits).PROGRESS: filtrange: 0%PROGRESS: Finished FILTRANGE.total cpu: 6 min 57.8 secPROGRESS: Start FINE.PROGRESS: FINE: 0%PROGRESS: FINE: 10%PROGRESS: FINE: 20%PROGRESS: FINE: 30%PROGRESS: FINE: 40%PROGRESS: FINE: 50%PROGRESS: FINE: 60%PROGRESS: FINE: 70%PROGRESS: FINE: 80%PROGRESS: FINE: 90%PROGRESS: FINE: 100%PROGRESS: Fine coregistration finished.PROGRESS: Finished FINE.total cpu: 10 min 9 sectotal cpu: 10 min 9.02002 sec .% How would you rate his incompetence Unmatched . - WARNING SUMMARY -There were no messages.输入run e4调用gedit编辑(重采样的初始化文件)主要完成以下三个步骤：coregpm计算匹配参数resample副影像重采样filtrange主副影像距离向滤波# -# COREGISTRATION PARAMETERS MODELING# -该步主要根据步骤FINE估计的偏移量，利用最小二乘方法计算匹配模型该模型应当具有亚像素级的精度。可以事先对相干系数设一个阈值来滤除一些数据。计算之后，估计模型和和实际偏移量之间的残差将被绘制。由于干预图对主副影像之间的匹配误差十分敏感，这一步非常关键。需要不断地迭代以获得最正确的变换模型。Doris采用一个CPM_MA*ITER选项来设定自动迭代的次数。CPM_THRESHOLD 0.4 / higher threshold-less windowsCPM_DEGREE 2 / 2d-polynomialc CPM_WEIGHT linear / or nonec CPM_WEIGHT quadratic / or noneCPM_WEIGHT bamler / paper R.Bamler, IGARSS 2000/2004CPM_MA*ITER 20 / automated removal, ma* 20 iterationsc CPM_K_ALPHA 1.97 / automated removal crit. valueCPM_PLOT NOBG / plot w/o magnitude backgroundc CPM_DUMP OFF / no creation of filesCPM_THRESHOLD为相干系数的阈值，这一选项决定了有多少数据可以用于估计多项式的系数。该参数的设定取决于FINE中所采用窗口的大小。假设采用的为6464的窗口，设为0.2即可。当然最好还是自己试验假设干数值，以获得理想的结果。CPM_DEGREE多项式阶数，建议采用2CPM_WEIGHT最小二乘的定权方法，建议选bamler方法CPM_MA*ITER迭代次数选项。如果在FINE中定义了600个窗口，则20次即可。其它参数可以采用默认值。# -# RESAMPLING SLAVE# -根据主影像格网和COREGPM步骤中得到的变换模型对副影像重采样重采样的过程就是利用插值核在空间域重建副影像原始信号的过程。该环节非常耗时为了评价重采样的质量，可以将采样后的副影像与主影像再匹配。这时偏移矢量应根本为零。Doris提供了大量的插值核供选择，按缺省设置即可。c - Interpolation Kernel -c RS_METHOD tri / linearc RS_METHOD nearest / nearest neighbourc RS_METHOD cc4p / cubic convolutionc RS_METHOD cc6p / good choice; testedc RS_METHOD ts6p / truncated sincc RS_METHOD ts8p / truncated sincc RS_METHOD ts16p / slower but betterc RS_METHOD knab6p / better than cc6 (theory)c RS_METHOD knab8p / KNAB windowc RS_METHOD knab10p / KNAB windowc RS_METHOD knab16p / KNAB windowc RS_METHOD knab4p / KNAB windowRS_METHOD rc6p / best choice (theory 0.9999)c RS_METHOD rc12p / best choice (theory 0.9999)cc - Output file -RS_OUT_FORMAT cr4 /c RS_DBOW 2000 3000 1 5000 / database output windowRS_SHIFTAZI on / keep this on插值在距离向和方位向上是独立的，对于主副影像重叠区重采样是逐点进展的，每次匹配多项式都要重新生成一次，假设按格网进展，效率可能会提高Bert Kampes# -# RANGE FILTERING (master&slave)# -该步主要通过对主副影像距离向滤波去除不重叠区域以减少干预图中的噪声。不重叠区域产生的原因是卫星视角差异。垂直基线越长，重叠区域越小。RF_METHOD adaptive / estimate local fringe freq.c RF_FFTLENGTH 8 / 8 pi*els 160m ground rangec RF_FFTLENGTH 32 / 32 pi*els 640m ground rangeRF_FFTLENGTH 128 / 128 pi*els 2500m ground rangeRF_OVERLAP 50 / half overlapRF_OVERSAMPLE 4 / master/slave before ifg generationRF_NLMEAN 15 / must be odd (15 pi*els 60m)RF_THRESHOLD 5 / SNR spectral peak detectionRF_HAMMING 0.75 / alpha of Hamming windowRF_WEIGHTCORR OFFc RF_OUT_FORMAT ci2 / reduce output file sizec /c /STOP / mandatoryRF_METHOD为距离向滤波方法，有adaptive和porbits方法，推荐使用adaptiveRF_FFTLENGTH对于自适应滤波方法，此参数为窗口长度。RF_OVERSAMPLE 过采样因子，缺省为4RF_NLMEAN 行数方位向将数行平均为1行以减少噪声，该参数必须为奇数，缺省值为15.RF_THRESHOLD信噪比阈值，缺省为5RF_HAMMING汉明滤波权其它参数按缺省值设置。2.2.8输入run s4正常时在终端返回如下信息：rootlocalhost doris_tailand# run -s4PROGRESS: Interpretation inputoptionsfile finished.total cpu: 0 min 0 secPROGRESS: Finished initializationPROGRESS: Orbit: interpolation coefficients computed.PROGRESS: Orbit: interpolation coefficients computed.PROGRESS: Start COREGPM.PROGRESS: Start iteration 0PROGRESS: Start iteration 1PROGRESS: Start iteration 2PROGRESS: Start iteration 3PROGRESS: Start iteration 4PROGRESS: Start iteration 5PROGRESS: Start iteration 6PROGRESS: Start iteration 7PROGRESS: Start iteration 8PROGRESS: Start iteration 9PROGRESS: Start iteration 10PROGRESS: Start iteration 11PROGRESS: Start iteration 12PROGRESS: Start iteration 13PROGRESS: Start iteration 14PROGRESS: Start iteration 15PROGRESS: Start iteration 16PROGRESS: Start iteration 17PROGRESS: Start iteration 18PROGRESS: Start iteration 19PROGRESS: Start iteration 20PROGRESS: finished computation of coregistration parameters.PROGRESS: Finished COREGPM.total cpu: 0 min 0.24 secPROGRESS: Start RESAMPLE.PROGRESS: Resample: normalized lookup table created (kernel and a*is).PROGRESS: RESAMPLE: 0%PROGRESS: RESAMPLE: 10%PROGRESS: RESAMPLE: 20%PROGRESS: RESAMPLE: 30%PROGRESS: RESAMPLE: 40%PROGRESS: RESAMPLE: 50%PROGRESS: RESAMPLE: 60%PROGRESS: RESAMPLE: 70%PROGRESS: RESAMPLE: 80%PROGRESS: RESAMPLE: 90%PROGRESS: RESAMPLE: 100%PROGRESS: Finished RESAMPLE.total cpu: 9 min 1.83002 secPROGRESS: calling preview for resampled slavePROGRESS: Start PREVIEW generation.PROGRESS: SUNraster file created of: Outdata/slave.resampled (see also file: ./slave_rs_mag.ras.sh)(#)Doris software, $Revision: 3.11 $, $Author: kampes $cp*fiddle: INFO: Number of output pi*els: 4896total cpu: 9 min 1.85004 sec .The shortest path between two truths in the real domain passes through the comple* domain. - WARNING SUMMARY -There were no messages. view with: run -v42.2.9输入run e5调用gedit编辑input.products(干预图的初始化文件)input.products主要完成如下六个步骤：interfero干预图生成comprefpha计算因地球形状引起的参考相位subtrrefpha 去除因地球形状引起的参考相位去平comprefdem计算因地形引起的参考相位subtrrefdem去除因地形引起的参考相位coherence生成相干系数图 -# INTERFEROGRAM# -该步主要生成干预图，为了减少干预图的噪声，一般要进展多视处理，取line：pi*el5:1,即在方位向上5行并一行，距离向不变，显然分辨率将降低，约为20m20mcINT_METHOD old / INT_METHOD oversample / INT_OUT_CINT Outdata/master_slave.cint / optional comple* int.c INT_OUT_INT Outdata/master_slave.int / optional real int. INT_MULTILOOK 5 1 / line, pi*el factors按缺省值执行即可。# -# REFERENCE PHASE COMPUTATION# -该步需要准确的轨道数据，如果去平方法是e*act则本步不需要。FE_METHOD porbits /c FE_METHOD method2 / not implementedFE_DEGREE 5 /多项式阶数FE_NPOINTS 501 /用于最小二乘估计的点数# -# SUBTRACT REFERENCE PHASE# -去平，缺省方法为polynomialc SRP_METHOD e*act / compute foreach pi*elSRP_METHOD polynomial / evaluate comprefpha polySRP_OUT_CINT Outdata/master_slave.srp /SRP_MULTILOOK 2 2c SRP_DUMPREFPHA ONc SRP_OUT_REFPHA Outdata/refphase.cr4 / only output for debug按缺省值设置# -# REFERENCE DEM# -该步计算地形引起的干预图，采样的DEM为USGS gtopo30 DEM 精度较低，水平格网间距为30秒，相当于1公里。由于输入的DEM精度较低，可采用双线性内插使之与干预图的精度相当，譬如假设干预图精度为20m20m,则需在DEM两点间内插50个点，但内插这么多点可以会出现问题。另一种方法是通过多视处理将干预图分辨率降低，以减少内插点数。CRD_METHOD TRI_LINEAR / or NEARESTc CRD_IN_FORMAT r4 / default is short integerCRD_IN_SIZE 6000 4800 / number