浅海水下地形雷达成像理论研究及应用

发布时间：2017-12-27 11:15

本文关键词：浅海水下地形雷达成像理论研究及应用　出处：《华东师范大学》2017年博士论文　论文类型：学位论文

【摘要】：海岸带水深及其变化对通航安全、渔业养殖、军事以及其它近海和离岸作业安全等都具有重要的意义。随着全球气候变化和人类活动的日益增强,海岸带的水文和生态环境正面临威胁,未来的海岸带地貌演变趋势也变得越发的不确定。因此科学家需要准确的水深数据来验证相关科学理论,而海岸带管理人员则需要根据水深情况制定相对应的发展策略。相比于传统的船载测深手段,遥感技术拥有大范围采样以及短时间成像的优势。而SAR(合成孔径雷达,Synthetic Aperture Radar)不仅兼顾了遥感技术的优势,还具备全天时和全天候的对地观测能力,是未来监测海岸带地貌演变的一种潜在有效工具。雷达传感器的一个重要应用就是可以观测到水下地形特征的变化,原因是水深变化可以生成海表面流场的辐聚和辐散区进而改变海表面的粗糙度。这种复杂现象是由水下地形与流场以及海表波浪之间的相互作用所形成的,并可以通过流场模型和雷达成像模型的结合(水下地形雷达成像模型)来解释。论文主要基于实测资料,利用遥感技术和数值模拟的手段研究浅海水下地形的2维雷达成像理论,并基于流场数值模型和RIM(Radar Imaging Model)模型的集成,建立了长江河口水下地形雷达成像模型。分析并解释了长江河口 SAR影像上出现的由复杂地形和水动力过程所导致的亮暗条纹特征。为了进一步验证该水下地形雷达成像模型的性能,研究采用控制变量的方法定量分析了水深、风场以及相关雷达参数对模拟的海表面粗糙度的影响。此外,还建立了一个理想模型来探究网格分辨率与模拟SAR影像之间的关系。研究的主要成果有:(1)建立了可以对拥有复杂水下地形的水域(如航道附近)进行任意时刻的模拟和雷达成像的2维长江口水下地形雷达成像模型。长江口 SAR图像中垂直于流场方向出现的亮-暗-亮条纹无法由传统的1维雷达成像模型进行解释,1维的流场模型或雷达成像模型都会限制成像模型对复杂水下地形特征的描述。将2维流场模型应用于长江河口,可以模拟河口区域复杂的水动力环境以及由水深变化所导致的流场变化,与实测水文资料的良好对比结果证明了该模型手段的可靠性。由模拟流场计算得到的辐聚和辐散区与SAR影像上出现的海表面不规则2维亮暗条纹的位置相吻合,通过输入风场以及相关雷达参数,2维雷达成像模型模拟得到的SAR图像与实测ERS-2和Sentinel-1 SAR图像也较为一致,这表明该模型具备较为可靠的水下地形雷达成像能力,这也为后续对不同模拟参数的敏感度分析测试奠定了基础。(2)模拟和分析了水深、风速、风向以及雷达视向等影响水下地形雷达成像的主要因素和条件。基于建立的雷达成像模型,通过控制变量的方法,改变水深、风速、风向以及雷达视向等部分雷达参数,模拟和比较了其对雷达后向散射信号的影响。结果表明,海表面粗糙度受整体水深值变化影响较小,而受到水下地形坡度变化的影响较大。根据综合分析的结果,顺风(或逆风)的流场、相对低风速和垂直于风向的雷达视向条件更适合浅海水下地形雷达成像。(3)在水下地形雷达成像模型中考虑了破碎波的贡献。输入相同的流场数据,采用不同的雷达成像模型模拟和比较了沿水下地形斜坡断面方向的雷达后向散射信号,并与实测SAR影像相比较。传统的二尺度模型低估了雷达后向散射信号,而引入破碎波贡献的RIM模型结果则在模拟的幅度和相位上都表现较好,尤其是在较大入射角的情况下。(4)探讨和分析了网格分辨率对模拟SAR影像的影响。采用理想数值模型排除了在实际情景条件下非海底地形因素的影响,模拟和分析不同尺度水下沙波情况下,不同网格分辨率对海表面粗糙度的影响。结果表明,对于小尺度的水下沙波(波长小于100米),采用10米的高分辨率网格模拟得到的相对雷达后向散射系数可以达到0.43 dB以上,这说明在适宜水下地形雷达成像的条件下,高分辨率的SAR影像具备探测小尺度水下地形的能力。在相同的水下地形条件下,网格分辨率的提升可以模拟得到更大的相对海表面粗糙度。相比于高分辨网格结果,使用低分辨率网格计算得到的雷达图像较为模糊,并且无法充分描述出水下地形的准确几何形状及其所在的空间位置。当网格大小达到一定阈值后,模拟得到的雷达后向散射系数会趋于稳定,这表明在水下地形雷达成像研究中,水下沙波的波长需要与网格间距达到一定的比例关系以充分描述水下地形所引起的海表面粗糙度变化。对于更复杂的实际环境,模型中网格大小的选择还需要兼顾海底地形的复杂程度和实测SAR影像的空间分辨率。
[Abstract]:The depth and change of the coastal zone are of great significance to navigation safety, fishery culture, military and other offshore and offshore operations. With the increasing global climate change and human activities, the hydrological and ecological environment of the coastal zone is facing a threat. The trend of coastal evolution is becoming more and more uncertain. Therefore, scientists need accurate water depth data to verify relevant scientific theories, while coastal managers need to formulate corresponding strategies based on water depth. Compared with the traditional means of ship borne sounding, remote sensing technology has the advantages of large range sampling and short time imaging. SAR (synthetic aperture radar, Synthetic Aperture Radar) not only takes account of the advantages of remote sensing technology, but also has the ability of all-weather and all-weather observation. It is a potential effective tool for monitoring coastal landform evolution in the future. An important application of radar sensors is that we can observe the change of underwater topography. The reason is that the change of water depth can generate the convergence and divergence area of the sea surface flow field, and change the roughness of the sea surface. This complex phenomenon is formed by the interaction between the underwater topography and the flow field and the sea surface waves, and can be explained by the combination of the flow field model and the radar imaging model (underwater terrain radar imaging model). Based on the measured data, the 2 dimensional radar imaging theory of shallow underwater terrain is studied based on the remote sensing technology and numerical simulation. Based on the integration of the flow field numerical model and the RIM (Radar Imaging Model) model, the underwater terrain radar imaging model of the Changjiang Estuary is established. The characteristics of bright and dark stripes caused by complex terrain and hydrodynamic process in the SAR image of the Yangtze River estuary are analyzed and explained. In order to further verify the performance of the underwater terrain radar imaging model, the influence of water depth, wind field and radar parameters on the simulated sea surface roughness is quantitatively analyzed by means of control variables. In addition, an ideal model is established to explore the relationship between grid resolution and analog SAR images. The main achievements are as follows: (1) a 2 dimensional underwater radar imaging model for the Yangtze River estuary is built, which can simulate and radar at any time with complicated underwater topography. The bright and dark stripe appearing in the SAR image of the Yangtze River Estuary can not be explained by the traditional 1 dimensional radar imaging model. The 1 dimensional flow field model or radar imaging model will limit the imaging model to describe the complex underwater topography. Applying the 2 dimensional flow field model to the Changjiang Estuary, we can simulate the complex hydrodynamic environment in the estuary area and the variation of the flow field caused by the change of the water depth, which proves the reliability of the model. The sea surface is obtained by the simulation of flow convergence and divergence and SAR images of 2 dimensional irregular light and dark stripes coincides with the position of the input, through the wind and radar parameters, 2 dimensional radar imaging model simulated SAR images and measured ERS-2 and Sentinel-1 SAR image is more consistent, which indicates that this model has a more reliable terrain radar imaging ability under water, it also laid the foundation for the follow-up of different simulation parameters sensitivity analysis. (2) the main factors and conditions of underwater terrain radar imaging are simulated and analyzed, such as water depth, wind speed, wind direction and radar vision. Based on the established radar imaging model, the influence of water depth, wind speed, wind direction and radar direction on radar backscatter signal is simulated and compared by controlling variables. The results show that the surface roughness of the sea is less affected by the change of the total water depth, and is greatly influenced by the change of the slope of the underwater terrain. According to the results of comprehensive analysis, the downwind (or adverse wind) flow field, relatively low wind speed and the radar direction condition perpendicular to the wind direction are more suitable for shallow underwater terrain radar imaging. (3) the contribution of broken waves is considered in the underwater terrain radar imaging model. Input the same flow field data, and use different radar imaging models to simulate and compare the radar backscatter signals along the cross section of the underwater terrain slope, and compare them with the measured SAR images. The traditional two scale model underestimated the backscatter signal of radar, while the RIM model introduced by breaking wave contributed better in the amplitude and phase of simulation, especially in the case of larger incidence angle. (4) the influence of grid resolution on simulated SAR images is discussed and analyzed. Excluding the influence in the actual conditions of non seabed terrain factors using the ideal numerical model, simulation and analysis of sand water under different scales, different grid resolution of sea surface roughness. The results show that for small scale underwater Sandwaves (wavelength less than 100 meters), relative to the radar with high resolution grid of 10 meters of the simulated backscattering coefficient can reach more than 0.43 dB, indicating that in radar imaging for underwater terrain condition, SAR shadow image with high resolution capability of terrain detection scale under water. In the same underwater terrain, the enhancement of grid resolution can simulate a larger relative sea surface roughness. Compared with the result of high-resolution grid, the radar image obtained from low resolution grid computation is fuzzy, and it can not fully describe the exact geometry and the location of the underwater terrain. When the grid size reaches a certain threshold, the simulated radar backscattering coefficient will be stable, this shows that in the study of underwater terrain radar imaging, and wavelength grid spacing of sand under water reaches a certain proportion in order to fully describe the underwater sea surface topography caused by the roughness change. about
【学位授予单位】：华东师范大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：P715.5

【相似文献】