CN103970932A - High-resolution permanent scatterer modeling method for separation of building and background - Google Patents
High-resolution permanent scatterer modeling method for separation of building and background Download PDFInfo
- Publication number
- CN103970932A CN103970932A CN201410072738.9A CN201410072738A CN103970932A CN 103970932 A CN103970932 A CN 103970932A CN 201410072738 A CN201410072738 A CN 201410072738A CN 103970932 A CN103970932 A CN 103970932A
- Authority
- CN
- China
- Prior art keywords
- building
- points
- deformation
- phase
- elevation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000000926 separation method Methods 0.000 title claims abstract description 11
- 238000012545 processing Methods 0.000 claims description 9
- 238000010276 construction Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 238000012417 linear regression Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 230000003778 catagen phase Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 description 13
- 238000011439 discrete element method Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 5
- 238000005305 interferometry Methods 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000002085 persistent effect Effects 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012731 temporal analysis Methods 0.000 description 2
- 238000000700 time series analysis Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Landscapes
- Radar Systems Or Details Thereof (AREA)
Abstract
本发明公开了一种高分辨率的建筑物和背景分离的永久散射体建模方法，其包括以下步骤：S01、将建筑物和背景地物上的PS点进行分离；S02、将建筑物和背景地物上的PS点进行独立构网；S03、对建筑物上的PS点的高程和形变参数进行估算，获得最终形变图。本方案得到的结果高程误差较小，形变速率数值较小，有效降低了高相位梯度对建筑物高程和形变解算的影响，适用于高分辨率下城市环境中的遥测建模。
The invention discloses a high-resolution building and background separation permanent scatterer modeling method, which includes the following steps: S01, separating the PS points on the building and the background object; S02, separating the building and the background The PS points on the background feature are independently constructed; S03, the elevation and deformation parameters of the PS points on the building are estimated to obtain the final deformation map. The results obtained by this scheme have small elevation errors and small deformation rate values, which effectively reduces the influence of high phase gradients on building elevation and deformation calculations, and are suitable for telemetry modeling in high-resolution urban environments.
Description
技术领域technical field
本发明涉及微波雷达遥感成像领域，尤其是涉及一种高分辨率条件下城市环境中建筑物和背景地物分离的永久散射体建模和参数估计方法。The invention relates to the field of microwave radar remote sensing imaging, in particular to a permanent scatterer modeling and parameter estimation method for separating buildings and background objects in an urban environment under high-resolution conditions.
背景技术Background technique
合成孔径雷达干涉测量(InSAR)技术是近年来发展起来的一项微波雷达遥感技术。D-InSAR作为InSAR技术的延伸，即差分合成孔径雷达干涉测量，是利用卫星两次经过同一区域获取的雷达影像进行差分干涉，以提取地面沉降信息。D-InSAR技术的优势在于能够全天候的采集雷达影像，理论上D-InSAR技术地面沉降监测的精度可以达到毫米级别。D-InSAR技术的主要局限性是时间失相干、空间失相干及大气延迟相位。Synthetic Aperture Radar Interferometry (InSAR) technology is a microwave radar remote sensing technology developed in recent years. As an extension of InSAR technology, D-InSAR, that is, differential synthetic aperture radar interferometry, uses radar images acquired by satellites passing through the same area twice to perform differential interference to extract land subsidence information. The advantage of D-InSAR technology is that it can collect radar images around the clock. Theoretically, the accuracy of D-InSAR technology for land subsidence monitoring can reach millimeter level. The main limitations of D-InSAR technology are time decoherence, space decoherence and atmospheric delay phase.
为了克服D-InSAR技术中存在的大气相位延迟及其失相干等，主要的方法是D-InSAR时序分析方法。D-InSAR时序分析方法首先基于大量的SAR（一般大于25景），通过放弃失相关严重的像元并对稳定的像元集进行时间序列分析，然后通过相关的统计理论可以最大程度地减弱时空去相干和大气延迟影响，提高地表形变监测的精度和可靠性。目前主要有以下几种D-InSAR时序分析方法：永久散射体干涉测量技术，短基线集干涉测量技术(Short BaselinesInterferometry,SBAS)，StaMPS(Stanford Method for Persistent Scatterer)，干涉点目标分析(Interferometric point target analysis)(Werner et al.,2003)，Squeeze SAR(SqueeSAR)。In order to overcome the atmospheric phase delay and decoherence in D-InSAR technology, the main method is the D-InSAR timing analysis method. The D-InSAR time series analysis method is first based on a large number of SAR (generally greater than 25 scenes), by discarding the pixels with serious loss of correlation and performing time series analysis on the stable pixel set, and then through the relevant statistical theory, the time series can be weakened to the greatest extent. Decoherence and atmospheric delay effects improve the accuracy and reliability of surface deformation monitoring. At present, there are mainly the following D-InSAR timing analysis methods: permanent scatterer interferometry, short baseline set interferometry (Short Baselines Interferometry, SBAS), StaMPS (Stanford Method for Persistent Scatterer), Interferometric point target analysis (Interferometric point target analysis) (Werner et al., 2003), Squeeze SAR (SqueeSAR).
D-InSAR时序分析方法都是以PSI（Persistent scattererinterferometric，永久散射体干涉）技术为理论基础的。PSI技术的核心思想是对PS（Persistent scatterer，永久散射体）点的干涉相位进行时间序列分析，根据各相位分量的时空特征，估算大气波动，DEM误差以及噪声等，将其从差分干涉相位中逐个分离，最终获取每个PS点的线性和非线形形变速率、大气延迟量(Atmosphere Phase Screen,APS)以及DEM误差。PSI技术与D-InSAR技术相比，有如下特点：首先需要大量的SAR影像，通常认为需要大于25景，Hooper等认为可以将SAR影像数降至12景；其研究对象不再是整景影像，而是从中筛选出具有稳定散射特性的PS点，构成离散点观测网络(较之常规的变形监测网密度更高)；其次，不能对PS点进行距离向和方位向的光谱滤波；其次，由于将PS点作为观测对象，降低了空间基线对相干性的影响，即使在临界基线的条件下，仍然可以通过分析PS差分干涉相位的变化反演形变信息；另外，在DEM误差、大气相位估计以及非线性形变等的解算上有明确的阐述，可以利用低精度数字高程模型对PS点处的DEM改正值进行估计，还可以获得主、辅影像的大气延迟相位；经过PSI方法处理，对形变时间序列估计达到亚毫米级的精度。The D-InSAR timing analysis methods are all based on the PSI (Persistent scattererinterferometric, permanent scatterer interference) technology. The core idea of PSI technology is to analyze the interferometric phase of PS (Persistent scatterer, permanent scatterer) points in time series, estimate atmospheric fluctuations, DEM errors and noise, etc. Separate one by one, and finally obtain the linear and nonlinear deformation rate, atmospheric delay (Atmosphere Phase Screen, APS) and DEM error of each PS point. Compared with D-InSAR technology, PSI technology has the following characteristics: First, it needs a large number of SAR images, which are generally considered to require more than 25 scenes. Hooper et al. believe that the number of SAR images can be reduced to 12 scenes; the research object is no longer the whole scene image , but select PS points with stable scattering characteristics to form a discrete point observation network (higher density than the conventional deformation monitoring network); secondly, spectral filtering in the range and azimuth directions cannot be performed on PS points; secondly, Since the PS point is used as the observation object, the influence of the spatial baseline on the coherence is reduced. Even under the condition of a critical baseline, the deformation information can still be retrieved by analyzing the change of the PS differential interferometric phase; in addition, the DEM error, atmospheric phase estimation And the calculation of nonlinear deformation is clearly stated. The low-precision digital elevation model can be used to estimate the DEM correction value at the PS point, and the atmospheric delay phase of the main and auxiliary images can also be obtained; after processing by the PSI method, the Deformation time series estimation achieves submillimeter accuracy.
PSI技术的数据处理流程包括：首先将覆盖同一地区的M幅SAR影像组成时间序列，选取1幅影像作为公共主影像，其余所有影像都配准到主影像上，生成M-1干涉图；然后提取SAR影像上保持高相干性的PS点目标；选取合适的区域DEM，对所有DEM进行坐标转换，将其模拟成雷达坐标系下的相位图，所有干涉对逐一与模拟的DEM进行相位差分处理，得到M-1差分干涉图；利用短距离相邻PS点对的空间自相关特性建模，邻近PS点的差分干涉相位之差可以表达成形变相位、DEM误差、大气延迟影响和噪声等的函数形式；选定其中一个高质量的PS点作为相位解缠参考基准点，利用回归分析求得各个PS点的线性形变速率和高程误差，对PS点的残余相位进行分离，包括非线性形变和大气延迟相位；最后对形变结果进行地理编码，获得覆盖区的形变速率和形变序列。The data processing process of PSI technology includes: first, M SAR images covering the same area are composed into a time series, one image is selected as the common main image, and all other images are registered to the main image to generate M-1 interferogram; then Extract the PS point target that maintains high coherence on the SAR image; select a suitable area DEM, perform coordinate transformation on all DEMs, and simulate it into a phase map in the radar coordinate system, and perform phase difference processing on all interference pairs one by one with the simulated DEM , to obtain the M-1 differential interferogram; using the spatial autocorrelation characteristics of short-distance adjacent PS point pairs to model, the differential interferometric phase difference of adjacent PS points can be expressed as deformation phase, DEM error, atmospheric delay effect and noise, etc. Functional form; select one of the high-quality PS points as the reference point for phase unwrapping, use regression analysis to obtain the linear deformation rate and elevation error of each PS point, and separate the residual phase of the PS point, including nonlinear deformation and Atmospheric delay phase; finally, the deformation results are geocoded to obtain the deformation rate and deformation sequence of the coverage area.
常规的PSI方法在处理过程中需要对PS点进行统一构网，由弧段连接形成网络，然后解算其高程和形变参数。然而在高分辨率条件下，城市环境中建筑物在长基线条件下，其干涉相位上的PS点与背景地物上的PS点之间的相位差异得以放大，在没有精确城市地表模型(DSM)的条件下，使得连接建筑物和背景地物的弧段形成类似于非连续的陡坡相位，因此该弧段无法满足相应的阈值条件。PSI的处理方法主要是剔除这种无法满足阈值条件弧段上的PS点。同时，考虑到建筑物高程相位和形变相位容易混为一体，在不能准确解算建筑物高程的情况下，将直接影响建筑物形变信息的提取精度。因此，如何解决高分辨率条件下，城市建筑物和背景地物之间在长基线条件下形成的高相位梯度问题，从而提高建筑高程形变信息的估计精度是高分辨率PSI技术面临的主要难题。The conventional PSI method needs to uniformly construct a network of PS points in the processing process, and form a network by connecting arc segments, and then calculate its elevation and deformation parameters. However, under high-resolution conditions, the phase difference between the PS points on the interferometric phase and the PS points on the background objects under the long-baseline condition of the buildings in the urban environment is amplified, and there is no accurate urban surface model (DSM ), the arc segment connecting the building and the background features forms a similar discontinuous steep slope phase, so the arc segment cannot meet the corresponding threshold condition. The processing method of PSI is mainly to eliminate the PS points on the arc segment that cannot meet the threshold condition. At the same time, considering that the building elevation phase and deformation phase are easily mixed together, if the building elevation cannot be accurately calculated, it will directly affect the extraction accuracy of building deformation information. Therefore, how to solve the high phase gradient problem formed between urban buildings and background objects under long baseline conditions under high resolution conditions, so as to improve the estimation accuracy of building elevation deformation information is the main problem faced by high resolution PSI technology .
发明内容Contents of the invention
常规的PSI方法完成差分处理（包括平地相位去除和高程相位补偿）后，处理过程中需要对PS点进行统一构网，形成由弧段连接的网络，进而利用二维周期估计的方法对相邻点的差分相位之差的序列进行求解，进而满足阈值条件的弧段积分解算整个网的形变值。在高分辨率SAR条件下，城市建筑物在斜距成像后投影到二维平面上形成了多个PS点与背景地物上的PS点混叠在一起，在PSI统一构网的方式下，建筑物上PS点与背景地物上的PS点不可避免地连接在一起，对于长基线永久散射体，它们之间干涉相位的差异得以放大，使得相邻点间形成高相位梯度，容易使后续的永久散射体参数估计产生解算错误。After the conventional PSI method completes the differential processing (including flat ground phase removal and elevation phase compensation), it is necessary to uniformly construct a network of PS points during the processing to form a network connected by arc segments, and then use the two-dimensional period estimation method to analyze the adjacent Solve the sequence of point difference phase difference, and then solve the deformation value of the whole network by arc integration satisfying the threshold condition. Under high-resolution SAR conditions, urban buildings are projected onto a two-dimensional plane after oblique distance imaging, forming multiple PS points mixed with PS points on background objects. In the way of PSI unified network construction, The PS points on the building and the PS points on the background objects are inevitably connected together. For the long-baseline permanent scatterers, the difference in the interference phase between them is amplified, so that a high phase gradient is formed between adjacent points, which is easy to make the follow-up The parameter estimation of the permanent scatterers for produces solution errors.
本发明主要致力于研究新的高分辨率条件下城市环境中建筑物和背景地物分离的永久散射体建模和参数估计算法，通过分离建筑物和背景地物上的PS点，即给定解缠路径来解算独立建筑物高程和形变信息。该方法能够有效避免建筑物和背景地物上的PS点之间产生弧段连接，保证在对各弧段的相邻连接点建立相位差分模型时，弧段上的相位增量上产生相位不连续，避免高相位梯度的问题。The present invention is mainly dedicated to researching new permanent scatterer modeling and parameter estimation algorithms for separating buildings and background objects in urban environments under high-resolution conditions. By separating PS points on buildings and background objects, the given Unwrap paths to resolve independent building elevation and deformation information. This method can effectively avoid arc connection between buildings and PS points on the background object, and ensure that when the phase difference model is established for the adjacent connection points of each arc, the phase difference will be generated on the phase increment of the arc. continuous, avoiding the problem of high phase gradients.
本发明针对上述技术问题主要是通过下述技术方案得以解决的：一种高分辨率的建筑物和背景分离的永久散射体建模方法，包括以下步骤：The present invention mainly solves the above-mentioned technical problems through the following technical solutions: a high-resolution building and background-separated permanent scatterer modeling method, comprising the following steps:
S01、将建筑物和背景地物上的PS点进行分离；S01, separating the PS points on the buildings and background features;
S02、将建筑物和背景地物上的PS点进行独立构网；S02, the PS points on the buildings and background features are independently constructed;
S03、对建筑物上的PS点的高程和形变参数进行估算，获得最终形变图。S03. Estimating the elevation and deformation parameters of PS points on the building to obtain a final deformation map.
作为优选，所述步骤S01中，将建筑物和背景地物上的PS点进行分离具体为：首先，根据临近SAR图像上邻近像素灰度、纹理等对影像进行分割；其次，利用Full Lambda-Schedule算法，在结合灰度和空间信息的基础上迭代合并邻近的小斑块；接着，计算建筑物类别的属性；然后，采用K邻近法依据待分类数据与训练区元素在N维空间的欧几里得距离来对SAR影像进行分类，获得不同建筑物的掩膜文件；最后，根据不同建筑物的掩膜文件对PS点进行分类，将PS点归类到不同的建筑物上，获得最终的分类结果。Preferably, in the step S01, the separation of the PS points on the buildings and the background features is as follows: firstly, the image is segmented according to the grayscale and texture of adjacent pixels on the adjacent SAR image; secondly, the Full Lambda- The Schedule algorithm, iteratively merges adjacent small patches on the basis of combining grayscale and spatial information; then, calculates the attributes of the building category; then, uses the K-nearby method based on the Euclidean relationship between the data to be classified and the elements in the training area in the N-dimensional space Classify the SAR images with a distance of a few miles to obtain the mask files of different buildings; finally, classify the PS points according to the mask files of different buildings, classify the PS points to different buildings, and obtain the final classification results.
作为优选，所述步骤S02中，将建筑物和背景地物上的PS点进行独立构网具体为：首先将建筑物上的PS点构成一个独立的Delaunay三角网，然后将背景地物上的PS点构成一个独立Delaunay三角网，并在每栋建筑物的三角网和背景地物的三角网之间建立一个连接点，通过连接点将两个独立的网络连接起来。As preferably, in the step S02, the PS points on the building and the background feature are independently constructed as follows: first, the PS points on the building are formed into an independent Delaunay triangular network, and then the PS points on the background feature are The PS points constitute an independent Delaunay triangular network, and a connection point is established between the triangular network of each building and the triangular network of the background object, and the two independent networks are connected through the connecting point.
作为优选，所述步骤S03中，对建筑物上的PS点的高程和形变参数进行估算，获得最终形变图具体为：采用多幅干涉对得到的高程相位结果加权平均的方法以减弱噪声影响，提高高程相位估计的可靠性，具体公式为：As a preference, in the step S03, the elevation and deformation parameters of the PS points on the building are estimated, and the final deformation map is obtained specifically as follows: using a method of weighted average of the elevation phase results obtained by multiple interference to weaken the influence of noise, To improve the reliability of elevation phase estimation, the specific formula is:
式中，φj为单幅干涉对的解缠相位，为高程相位的最佳估值，t为干涉图数量，pj为每幅干涉图对应的权；然后将建筑物上的PS点的组合高程相位从干涉图中去除建筑高程相位贡献以提取建筑物形变相位信息，接着以建筑物上PS点的形变相位为处理对象，利用二维线性回归相位模型，通过若干次迭代运算逐步修正高程值，解算形变速率，最终实现对形变序列的提取；最后将独立建筑物的形变结果与背景地物的形变结果，通过共同的连接点相连接形成较大范围的形变图。where φ j is the unwrapped phase of a single interference pair, is the best estimate of the elevation phase, t is the number of interferograms, and p j is the weight corresponding to each interferogram; then the combined elevation phase of the PS points on the building is removed from the interferogram to extract the building elevation phase contribution The deformation phase information of the object, and then take the deformation phase of the PS point on the building as the processing object, use the two-dimensional linear regression phase model, and gradually correct the elevation value through several iterative operations, solve the deformation rate, and finally realize the deformation sequence Extraction; finally, the deformation results of independent buildings and the deformation results of background objects are connected through common connection points to form a large-scale deformation map.
本发明带来的实质性效果是，首先常规方法得到的PS点，建筑物顶部的PS点由于地形残余相位过大都被剔除了，造成PS点稀少，本发明提出的方法建筑物上PS点数量比较多，基本覆盖每栋建筑物；其次，常规方法得到的PS点形变速率明显偏大，个别PS点每年形变甚至达到厘米级，该结果明显带有残留的建筑物高程相位，随着建筑物高度的变化，影响了建筑物上形变速率的估算，使得建筑物的形变速率结果偏大。本发明提出的改进的PSI解算方法得到的形变速率结果中残余的高程误差明显较小，形变速率数值较小，有效降低了高相位梯度对建筑物高程和形变解算的影响，可见本方案在建筑物形变监测中的优越性。The substantive effect that the present invention brings is, at first conventional method obtains the PS point, and the PS point on the top of the building has all been rejected because the landform residual phase is too large, causes PS point to be scarce, and the PS point quantity on the method building that the present invention proposes There are quite a few, basically covering every building; secondly, the deformation rate of PS points obtained by the conventional method is obviously too large, and the annual deformation of individual PS points even reaches the centimeter level. This result obviously has residual building elevation phases. The change of height affects the estimation of the deformation rate of the building, which makes the result of the deformation rate of the building too large. The residual elevation error in the deformation rate result obtained by the improved PSI calculation method proposed by the present invention is obviously smaller, and the deformation rate value is smaller, which effectively reduces the influence of high phase gradient on the building elevation and deformation calculation. It can be seen that this scheme Superiority in building deformation monitoring.
附图说明Description of drawings
图1是本发明的一种流程示意图。Fig. 1 is a kind of flow diagram of the present invention.
具体实施方式Detailed ways
下面通过实施例，并结合附图，对本发明的技术方案作进一步具体的说明。The technical solutions of the present invention will be further specifically described below through the embodiments and in conjunction with the accompanying drawings.
实施例：本实施例的一种高分辨率的建筑物和背景分离的永久散射体建模方法，如图1所示，包括以下步骤：Embodiment: A kind of high-resolution building and background separation permanent scatterer modeling method of the present embodiment, as shown in Figure 1, comprises the following steps:
S01、将建筑物和背景地物上的PS点进行分离；S01, separating the PS points on the buildings and background features;
S02、将建筑物和背景地物上的PS点进行独立构网；S02, the PS points on the buildings and background features are independently constructed;
S03、对建筑物上的PS点的高程和形变参数进行估算，获得最终形变图。S03. Estimating the elevation and deformation parameters of PS points on the building to obtain a final deformation map.
1、建筑物和背景地物上PS点的分离1. Separation of PS points on buildings and background features
本发明采用了面向对象的分类方法实现了建筑物和背景地物上PS点的分离，整个分类方法的如下：首先，根据临近SAR图像上邻近像素灰度、纹理等对影像进行分割，使用了一种基于边缘的分割算法，这种算法计算很快，并且只要输入参数，就能产生多尺度分割结果。通过不同尺度上边界的差异控制，从而产生从细到粗的多尺度分割；其次，影像分割时，由于阈值过低，一些特征会被错分，一个特征也有可能被分成很多部分。我们利用了FullLambda-Schedule算法，结合灰度和空间信息的基础上迭代合并邻近的小斑块；接着，计算建筑物类别的属性，主要是根据建筑物的灰度特征（建筑物的图像灰度值较高，而背景地物的图像灰度值较低），计算类别的灰度信息；然后，采用K邻近法依据待分类数据与训练区元素在N维空间的欧几里得距离来对SAR影像进行分类，获得不同建筑物的掩膜文件；最后，根据不同建筑物的掩膜文件对PS点进行分类，将PS点归类到不同的建筑物上，获得最终的分类结果。The present invention adopts an object-oriented classification method to realize the separation of PS points on buildings and background features. The whole classification method is as follows: first, the image is segmented according to the adjacent pixel grayscale and texture on the adjacent SAR image, using An edge-based segmentation algorithm that is fast to calculate and can produce multi-scale segmentation results as long as the input parameters are required. Through the difference control of boundaries at different scales, multi-scale segmentation from fine to coarse is produced; secondly, when the image is segmented, due to the low threshold, some features will be misclassified, and a feature may also be divided into many parts. We use the FullLambda-Schedule algorithm to iteratively merge adjacent small patches based on grayscale and spatial information; then, calculate the attributes of the building category, mainly based on the grayscale characteristics of the building (the image grayscale of the building value is high, while the image gray value of the background feature is low), calculate the gray information of the category; then, use the K proximity method to compare the Euclidean distance between the data to be classified and the elements in the training area in N-dimensional space Classify the SAR images to obtain the mask files of different buildings; finally, classify the PS points according to the mask files of different buildings, classify the PS points to different buildings, and obtain the final classification results.
2、建筑物和背景地物上PS点的独立构网2. Independent network construction of PS points on buildings and background features
获得独立建筑物上的PS点后，即可进行PS点构网。常规的PS构网一般选择Delaunay不规则三角网，即将所有的离散的PS点连接起来构成无重叠的三角形网络。但该构网方式是在全局条件下的统一的点连接方式，使得建筑物与背景地物上的PS点不可避免地连接在一起。在没有精确DSM进行高程相位补偿条件下，由于点间的相位不连续难以应用二维周期图正确估计点间形变速率和高程误差改正。若我们将识别出的建筑物和背景地物上的PS点进行独立构网，能够有效避免建筑物和背景地物上PS点间弧段的连接。因此，我们采用建筑物和背景地物分别独立构网方式。即首先将建筑物上的PS点构成一个独立的Delaunay三角网，然后将背景地物上的PS点构成一个独立Delaunay三角网，并在每栋建筑物的三角网和背景地物的三角网之间建立一个连接点，通过连接点我们可以将两个独立的网络连接起来。After obtaining the PS points on the independent buildings, the PS point network can be constructed. The conventional PS network generally chooses the Delaunay irregular triangular network, that is, all the discrete PS points are connected to form a non-overlapping triangular network. However, this network construction method is a unified point connection method under global conditions, so that the PS points on the buildings and background objects are inevitably connected together. In the absence of accurate DSM for elevation phase compensation, it is difficult to correctly estimate the deformation rate and elevation error correction between points by using the two-dimensional periodogram due to the phase discontinuity between points. If we independently construct a network for the PS points on the recognized buildings and background objects, the connection between the arcs between the buildings and the PS points on the background objects can be effectively avoided. Therefore, we adopt the independent network construction method of buildings and background features. That is, firstly, the PS points on the buildings form an independent Delaunay triangular network, and then the PS points on the background features form an independent Delaunay triangular network, and between the triangular network of each building and the triangular network of the background features Establish a connection point between them, through which we can connect two independent networks.
3、建筑物上PS点高程和形变参数估计3. Estimation of PS point elevation and deformation parameters on the building
我们采用多干涉图组合的方式来进行独立建筑物的高程相位估计，以此来消除大气延迟误差和相位噪声误差对单幅干涉对得到的高程相位结果的影响。具体方法是采用多幅干涉对得到的高程相位结果加权平均的方法以减弱噪声影响，提高高程相位估计的可靠性：We use the combination of multiple interferograms to estimate the elevation phase of independent buildings, so as to eliminate the influence of atmospheric delay errors and phase noise errors on the obtained elevation phase results of single interferograms. The specific method is to use the method of weighted average of the elevation phase results obtained by multi-frame interference to reduce the influence of noise and improve the reliability of elevation phase estimation:
式中，φj为单幅干涉对的解缠相位，为高程相位的最佳估值，t为干涉图数量，pj为每幅干涉图对应的权。在一定范围内，干涉基线的垂直分量越大，高程结果越可靠。因此，在定权时必须考虑垂直基线，另外需要考虑的因素包括相位噪声强度和大气延迟影响。考虑到建筑物的相干性较好且范围较小（小于1km2）受大气延迟影响较小，本发明采用仅考虑基线的定权方法，即具有相对大的垂直基线B⊥的干涉对得到的高程结果更为可靠。一般选择权重因子与B⊥ 2成正比，即p=B⊥ 2。定权公式可以改写成：where φ j is the unwrapped phase of a single interference pair, is the best estimate of the elevation phase, t is the number of interferograms, and p j is the weight corresponding to each interferogram. Within a certain range, the larger the vertical component of the interferometric baseline, the more reliable the elevation result. Therefore, the vertical baseline must be considered when determining the weight, and other factors that need to be considered include the phase noise intensity and the influence of atmospheric delay. Considering that the coherence of the building is good and the range is small (less than 1km 2 ) and is less affected by atmospheric delay, the present invention adopts a weighting method that only considers the baseline, that is, the interference pair with a relatively large vertical baseline B ⊥ is obtained Elevation results are more reliable. Generally, the selection weight factor is proportional to B ⊥ 2 , that is, p=B ⊥ 2 . The fixed weight formula can be rewritten as:
我们将建筑物上的PS点的组合高程相位从干涉图中去除建筑高程相位贡献以提取建筑物形变相位信息，然后以建筑物上PS点的形变相位为处理对象，利用二维线性回归相位模型，通过多次迭代运算逐步修正高程值，解算形变速率，最终实现对形变序列的提取。最后将独立建筑物的形变结果与背景地物的形变结果，通过共同的连接点相连接形成较大范围的形变图。We remove the combined elevation phase contribution of the PS points on the building from the interferogram to extract the building deformation phase information, and then take the deformation phase of the PS points on the building as the processing object, using two-dimensional linear regression The phase model gradually corrects the elevation value through multiple iterative operations, calculates the deformation rate, and finally realizes the extraction of the deformation sequence. Finally, the deformation results of independent buildings and the deformation results of background objects are connected through common connection points to form a large-scale deformation map.
本文中所描述的具体实施例仅仅是对本发明精神作举例说明。本发明所属技术领域的技术人员可以对所描述的具体实施例做各种各样的修改或补充或采用类似的方式替代，但并不会偏离本发明的精神或者超越所附权利要求书所定义的范围。The specific embodiments described herein are merely illustrative of the spirit of the invention. Those skilled in the art to which the present invention belongs can make various modifications or supplements to the described specific embodiments or adopt similar methods to replace them, but they will not deviate from the spirit of the present invention or go beyond the definition of the appended claims range.
尽管本文较多地使用了独立构网、多幅干涉等术语，但并不排除使用其它术语的可能性。使用这些术语仅仅是为了更方便地描述和解释本发明的本质；把它们解释成任何一种附加的限制都是与本发明精神相违背的。Although terms such as independent network construction and multi-frame interference are frequently used in this paper, the possibility of using other terms is not excluded. These terms are used only for the purpose of describing and explaining the essence of the present invention more conveniently; interpreting them as any kind of additional limitation is against the spirit of the present invention.
Claims (4)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410072738.9A CN103970932B (en) | 2014-02-28 | 2014-02-28 | A kind of Permanent scatterers modeling method of high-resolution building and background separation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410072738.9A CN103970932B (en) | 2014-02-28 | 2014-02-28 | A kind of Permanent scatterers modeling method of high-resolution building and background separation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103970932A true CN103970932A (en) | 2014-08-06 |
| CN103970932B CN103970932B (en) | 2017-10-10 |
Family
ID=51240423
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410072738.9A Active CN103970932B (en) | 2014-02-28 | 2014-02-28 | A kind of Permanent scatterers modeling method of high-resolution building and background separation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103970932B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104268440A (en) * | 2014-10-28 | 2015-01-07 | 铁道第三勘察设计院集团有限公司 | PSI settlement monitoring result processing method on basis of high-speed rail structural body |
| CN104765026A (en) * | 2015-04-29 | 2015-07-08 | 天津市测绘院 | Method for extracting ground attribute data in interferometry synthetic aperture radar data |
| CN104793207A (en) * | 2015-04-29 | 2015-07-22 | 天津市测绘院 | Method for extracting building attribute data in synthetic aperture radar interferometry data |
| CN106772377A (en) * | 2017-01-18 | 2017-05-31 | 深圳市路桥建设集团有限公司 | A kind of building deformation monitoring method based on InSAR |
| CN108132468A (en) * | 2017-12-25 | 2018-06-08 | 中南大学 | A kind of more baseline polarimetric SAR interferometry depth of building extracting methods |
| CN108872989A (en) * | 2018-07-16 | 2018-11-23 | 北京航空航天大学 | A kind of PS-InSAR precise search method based on maximum cycle figure |
| CN108983233A (en) * | 2018-06-13 | 2018-12-11 | 四川大学 | The PS point of GB-InSAR Data processing combines choosing method |
| CN109709550A (en) * | 2019-01-17 | 2019-05-03 | 武汉大学 | A method for monitoring and processing reservoir bank slope deformation based on InSAR image data |
| CN116049929A (en) * | 2022-10-26 | 2023-05-02 | 马培峰 | Urban building risk level InSAR evaluation and prediction method |
| CN116106904A (en) * | 2023-02-22 | 2023-05-12 | 深圳大学 | Object-oriented MT-InSAR facility deformation monitoring method and equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103514606A (en) * | 2013-10-14 | 2014-01-15 | 武汉大学 | Heterology remote sensing image registration method |
| CN103559721A (en) * | 2013-11-25 | 2014-02-05 | 中国科学院自动化研究所 | Triangulation rapid image cloning method based on image gradient |
-
2014
- 2014-02-28 CN CN201410072738.9A patent/CN103970932B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103514606A (en) * | 2013-10-14 | 2014-01-15 | 武汉大学 | Heterology remote sensing image registration method |
| CN103559721A (en) * | 2013-11-25 | 2014-02-05 | 中国科学院自动化研究所 | Triangulation rapid image cloning method based on image gradient |
Non-Patent Citations (5)
| Title |
|---|
| LIANGMING LIU 等: ""An object-oriented daytime land-fog-detection approach based on the mean-shift and full lambda-schedule algorithm using EOS/MODIS data"", 《INTERNATIONAL JOURNAL OF REMOTE SENSING》 * |
| S.GERNHARDT 等: "Potential of very high resolution SAR for persistent scatterer interferometry in urban areas", 《ANNALS OF GIS》 * |
| 兰恒星 等: "城市建筑物沉降永久散射体干涉雷达监测", 《天津大学学报》 * |
| 李曼 等: "基于精细DEM的InSAR大气相位改正试验研究", 《国土资源遥感》 * |
| 贾洪果: "高分辨率永久散射体雷达干涉及其应用于高速铁路沉降监测", 《中国博士学位论文全文数据库 工程科技Ⅱ辑（月刊）》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104268440B (en) * | 2014-10-28 | 2017-04-19 | 铁道第三勘察设计院集团有限公司 | PSI settlement monitoring result processing method on basis of high-speed rail structural body |
| CN104268440A (en) * | 2014-10-28 | 2015-01-07 | 铁道第三勘察设计院集团有限公司 | PSI settlement monitoring result processing method on basis of high-speed rail structural body |
| CN104765026A (en) * | 2015-04-29 | 2015-07-08 | 天津市测绘院 | Method for extracting ground attribute data in interferometry synthetic aperture radar data |
| CN104793207A (en) * | 2015-04-29 | 2015-07-22 | 天津市测绘院 | Method for extracting building attribute data in synthetic aperture radar interferometry data |
| CN106772377A (en) * | 2017-01-18 | 2017-05-31 | 深圳市路桥建设集团有限公司 | A kind of building deformation monitoring method based on InSAR |
| CN108132468A (en) * | 2017-12-25 | 2018-06-08 | 中南大学 | A kind of more baseline polarimetric SAR interferometry depth of building extracting methods |
| CN108983233A (en) * | 2018-06-13 | 2018-12-11 | 四川大学 | The PS point of GB-InSAR Data processing combines choosing method |
| CN108872989A (en) * | 2018-07-16 | 2018-11-23 | 北京航空航天大学 | A kind of PS-InSAR precise search method based on maximum cycle figure |
| CN108872989B (en) * | 2018-07-16 | 2022-04-12 | 北京航空航天大学 | An accurate search method for PS-InSAR based on maximum periodogram |
| CN109709550A (en) * | 2019-01-17 | 2019-05-03 | 武汉大学 | A method for monitoring and processing reservoir bank slope deformation based on InSAR image data |
| CN116049929A (en) * | 2022-10-26 | 2023-05-02 | 马培峰 | Urban building risk level InSAR evaluation and prediction method |
| CN116049929B (en) * | 2022-10-26 | 2023-09-29 | 马培峰 | Urban building risk level InSAR evaluation and prediction method |
| US12055624B2 (en) | 2022-10-26 | 2024-08-06 | Peifeng MA | Building risk monitoring and predicting based on method integrating MT-InSAR and pore water pressure model |
| CN116106904A (en) * | 2023-02-22 | 2023-05-12 | 深圳大学 | Object-oriented MT-InSAR facility deformation monitoring method and equipment |
| CN116106904B (en) * | 2023-02-22 | 2023-07-18 | 深圳大学 | Object-oriented MT-InSAR facility deformation monitoring method and equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103970932B (en) | 2017-10-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103970932B (en) | A kind of Permanent scatterers modeling method of high-resolution building and background separation | |
| CN111059998B (en) | High-resolution-based time sequence InSAR deformation monitoring method and system | |
| CN109782282B (en) | Time series InSAR analysis method integrating troposphere atmospheric delay correction | |
| Schumann et al. | The utility of spaceborne radar to render flood inundation maps based on multialgorithm ensembles | |
| CN104360332B (en) | Atmospheric phase screen extraction method based on ground-based SAR (synthetic aperture radar) interference | |
| Pipaud et al. | Evaluation of TanDEM-X elevation data for geomorphological mapping and interpretation in high mountain environments—A case study from SE Tibet, China | |
| US10325349B2 (en) | Method and apparatus for enhancing 3D model resolution | |
| Gallant | Adaptive smoothing for noisy DEMs | |
| CN105954747A (en) | Tower foundation stability analyzing method based on three-dimensional deformation monitoring of unfavorable geologic body of power grid | |
| CN107341781B (en) | SAR image correction method based on improved phase consistency feature vector base map matching | |
| Zhu et al. | Robust registration of aerial images and LiDAR data using spatial constraints and Gabor structural features | |
| Zhang et al. | Satellite SAR geocoding with refined RPC model | |
| CN108663678B (en) | Multi-baseline InSAR phase unwrapping algorithm based on mixed integer optimization model | |
| CN104794730B (en) | SAR image segmentation method based on super-pixel | |
| CN106204539A (en) | A kind of method of inverting urban architecture thing based on Morphological Gradient sedimentation | |
| CN103559506A (en) | Sub-pixel drawing method based on vector boundaries | |
| CN103268496A (en) | SAR Image Target Recognition Method | |
| CN109212522A (en) | A kind of method and apparatus obtaining numerical map | |
| CN111458691B (en) | Building information extraction method, device and computer equipment | |
| CN110703252B (en) | Correction Method of Digital Elevation Model in Shadow Region of Interferometric Synthetic Aperture Radar | |
| CN106842199A (en) | It is a kind of to merge the method that different resolution SAR data monitors Ground Deformation | |
| CN103454636A (en) | Differential interferometric phase estimation method based on multi-pixel covariance matrixes | |
| CN105678716B (en) | A kind of ground SAR atmospheric interference method for correcting phase and device | |
| CN115951350A (en) | Permanent scatterer point extraction method, device, equipment and medium | |
| Yao et al. | A continuous digital elevation representation model for DEM super-resolution |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |