星載合成孔徑雷達對海內波檢測與參數估計
發(fā)布時間：2018-10-30 17:22
【摘要】：合成孔徑雷達(Synthetic Aperture Radar,SAR)具有全天時、全天候、觀測范圍廣的特點,在遙感領域具有越來越廣泛的應用。內波(Internal Waves,IW)是指在流體內部發(fā)生的一種波動形式,而海洋內波就是一種典型的內波形式。由于海洋內波具有較大的幅度,傳遞巨大的能量,在海洋開發(fā)、船只航線規(guī)劃乃至國防軍事領域都具有重要的研究意義。并且,隨著分辨率的不斷提高,SAR對海面的觀察能力逐漸提升,包含的信息也更加的豐富。由于電磁波無法深入到海洋中直接對海洋內波進行觀測,所以僅能通過海洋表面的變化,推導海洋海洋內波的相關參數。但是,由于從海洋內波到海洋表面,再到接收機的過程會引入較多的干擾因素,使得在運算量和估計的準確性上都存在較大的可改進性�；谏厦娴脑�,本課題在分析內波的形成機理、SAR對海洋表面的成像模型基礎上,著眼于利用SAR數據,探究在SAR參數對內波圖像的影響、利用圖像處理的方法在SAR圖像中進行在圖像內定位內波的位置、使用期望最大化算法對統(tǒng)計直方圖進行模型擬合以及在SAR圖像中進行內波參數提取的方法。本文的主要研究內容如下:論文首先建立海洋內波產生的模型,并推導其所滿足的動力學方程,求取滿足方程的穩(wěn)態(tài)解。對海洋表面在SAR觀測情況下的三種不同成像模型進行了介紹;同時,針對接收幅度過小的情況,引入調制深度的概念,利用仿真工具和特定的內波模型以及成像模型,通過針對沿內波傳播方向上極值點隨著參數變化的趨勢,分析參數對成像效果帶來的影響,尋找最佳的雷達觀測參數組合。其次,利用簡單圖像處理的方法對SAR圖像進行預處理,分離出海洋區(qū)域;利用基于馬爾科夫隨機場的分割方法對含有內波條紋的圖像進行紋理增強,并借助基于Radon變換的紋理檢測方法,從大場景的SAR圖像中,快速定位內波在圖像中的位置;針對SAR圖像分辨率提升帶來統(tǒng)計直方圖出現多模態(tài)、拖尾嚴重的現象,結合之前學者提出的廣義混合模型,利用期望最大化的方法進行分布模型的參數估計,為SAR圖像的分類、降噪和目標檢測提供輔助作用;從估計結果與實際結果的均方差衡量估計的結果,通過迭代次數衡量方法的計算量。最后,利用已知的先驗模型和獲得的圖像剖面數據,采用曲線擬合的方式,對模型中的參數進行確定;同時針對內波圖像剖面的非線性特點,利用基于經驗模態(tài)分解的方法,分解得到內波分量,并利用該分量對內波參數進行估計;根據干涉的概念以及地面運動速度和干涉相位的關系,利用干涉相位提取海洋表面的速度信息;提出基于仿真迭代的參數估計方法,對內波的參數進行迭代估計。
[Abstract]:Synthetic Aperture Radar (Synthetic Aperture Radar,SAR) has been widely used in the field of remote sensing because of its wide range of observation and all-weather. Internal wave (Internal Waves,IW) is a form of internal wave occurring in fluid, and ocean wave is a typical form of internal wave. Because of the large amplitude of ocean internal wave and the transmission of huge energy, it is of great significance in the field of ocean exploitation, ship route planning and even national defense and military affairs. Moreover, with the improvement of resolution, the ability of SAR to observe sea surface is improved gradually, and the information is more abundant. Because the electromagnetic wave can not penetrate into the ocean directly to observe the ocean internal wave, it can only deduce the relevant parameters of the ocean internal wave by the variation of the ocean surface. However, the process from the ocean internal wave to the ocean surface, and then to the receiver, will introduce more interference factors, which makes it possible to improve the computational complexity and the accuracy of the estimation. Based on the above reasons, based on the analysis of the formation mechanism of internal waves and the imaging model of SAR on the ocean surface, we focus on the use of SAR data to explore the influence of SAR parameters on the internal wave images. The method of image processing is used to locate the position of the internal wave in the SAR image, the model fitting of the statistical histogram and the extraction of the internal wave parameters in the SAR image are carried out by using the expectation maximization algorithm. The main contents of this paper are as follows: firstly, the model of ocean internal wave generation is established, and the satisfied dynamic equation is derived, and the steady-state solution of the satisfied equation is obtained. Three different imaging models of ocean surface under SAR observation are introduced. At the same time, the concept of modulation depth is introduced in the case of too small receiving amplitude. By using the simulation tool, the specific internal wave model and the imaging model, the trend of the extreme point changing with the parameters along the direction of the internal wave propagation is analyzed. The effect of the parameters on the imaging effect is analyzed and the best combination of radar observation parameters is found. Secondly, the SAR image is preprocessed by the simple image processing method, and the ocean area is separated. The segmentation method based on Markov random field is used to enhance the texture of the image with internal wave fringes. With the help of texture detection method based on Radon transform, the position of internal wave in large scene SAR image is quickly located. Aiming at the phenomenon of multi-modal and serious trailing in the statistical histogram caused by the resolution enhancement of SAR image, combined with the generalized mixed model proposed by the previous scholars, the parameter estimation of the distribution model is carried out by using the method of expectation maximization, which is the classification of the SAR image. Noise reduction and target detection provide auxiliary effect; The estimated results are measured from the RMS of the estimated results and the actual results, and the computational complexity of the method is measured by the number of iterations. Finally, using the known prior model and the obtained image profile data, the parameters in the model are determined by the way of curve fitting. At the same time, according to the nonlinear characteristics of the internal wave image profile, the internal wave component is obtained by using the empirical mode decomposition method, and the internal wave parameters are estimated by this component. According to the concept of interference and the relationship between the velocity of ground motion and the phase of interference, the velocity information of ocean surface is extracted by using the phase of interference, and a parameter estimation method based on simulation iteration is proposed to estimate the parameters of internal wave iteratively.
【學位授予單位】：哈爾濱工業(yè)大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TN958
本文編號：2300663
[Abstract]:Synthetic Aperture Radar (Synthetic Aperture Radar,SAR) has been widely used in the field of remote sensing because of its wide range of observation and all-weather. Internal wave (Internal Waves,IW) is a form of internal wave occurring in fluid, and ocean wave is a typical form of internal wave. Because of the large amplitude of ocean internal wave and the transmission of huge energy, it is of great significance in the field of ocean exploitation, ship route planning and even national defense and military affairs. Moreover, with the improvement of resolution, the ability of SAR to observe sea surface is improved gradually, and the information is more abundant. Because the electromagnetic wave can not penetrate into the ocean directly to observe the ocean internal wave, it can only deduce the relevant parameters of the ocean internal wave by the variation of the ocean surface. However, the process from the ocean internal wave to the ocean surface, and then to the receiver, will introduce more interference factors, which makes it possible to improve the computational complexity and the accuracy of the estimation. Based on the above reasons, based on the analysis of the formation mechanism of internal waves and the imaging model of SAR on the ocean surface, we focus on the use of SAR data to explore the influence of SAR parameters on the internal wave images. The method of image processing is used to locate the position of the internal wave in the SAR image, the model fitting of the statistical histogram and the extraction of the internal wave parameters in the SAR image are carried out by using the expectation maximization algorithm. The main contents of this paper are as follows: firstly, the model of ocean internal wave generation is established, and the satisfied dynamic equation is derived, and the steady-state solution of the satisfied equation is obtained. Three different imaging models of ocean surface under SAR observation are introduced. At the same time, the concept of modulation depth is introduced in the case of too small receiving amplitude. By using the simulation tool, the specific internal wave model and the imaging model, the trend of the extreme point changing with the parameters along the direction of the internal wave propagation is analyzed. The effect of the parameters on the imaging effect is analyzed and the best combination of radar observation parameters is found. Secondly, the SAR image is preprocessed by the simple image processing method, and the ocean area is separated. The segmentation method based on Markov random field is used to enhance the texture of the image with internal wave fringes. With the help of texture detection method based on Radon transform, the position of internal wave in large scene SAR image is quickly located. Aiming at the phenomenon of multi-modal and serious trailing in the statistical histogram caused by the resolution enhancement of SAR image, combined with the generalized mixed model proposed by the previous scholars, the parameter estimation of the distribution model is carried out by using the method of expectation maximization, which is the classification of the SAR image. Noise reduction and target detection provide auxiliary effect; The estimated results are measured from the RMS of the estimated results and the actual results, and the computational complexity of the method is measured by the number of iterations. Finally, using the known prior model and the obtained image profile data, the parameters in the model are determined by the way of curve fitting. At the same time, according to the nonlinear characteristics of the internal wave image profile, the internal wave component is obtained by using the empirical mode decomposition method, and the internal wave parameters are estimated by this component. According to the concept of interference and the relationship between the velocity of ground motion and the phase of interference, the velocity information of ocean surface is extracted by using the phase of interference, and a parameter estimation method based on simulation iteration is proposed to estimate the parameters of internal wave iteratively.
【學位授予單位】：哈爾濱工業(yè)大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TN958
【參考文獻】
相關期刊論文 前10條
1 范開國;顧艷鎮(zhèn);陳春濤;許可;陳鵬;;希爾伯特—黃變換在遙感圖像中的實例研究與應用展望[J];海洋技術學報;2015年04期
2 陳捷;陳標;陶榮華;許素芹;;SAR圖像海洋內波參數自動提取方法[J];海洋技術學報;2014年06期
3 楊剛;陳標;張本濤;;EMD在海洋內波研究中的應用[J];艦船電子工程;2014年03期
4 丁燦;張本濤;高國興;陳標;吳寶勤;;合成孔徑雷達圖像中海洋內波的特征檢測[J];海洋測繪;2012年05期
5 汪雄良;王春玲;朱炬波;梁甸農;;基于EEMD的SAR海洋內波參數反演[J];測試技術學報;2012年01期
6 宋海斌;拜陽;董崇志;宋洋;;南海東北部內波特征——經驗模態(tài)分解方法應用初探[J];地球物理學報;2010年02期
7 宋海斌;Luis Pinheiro;王東曉;董崇志;宋洋;拜陽;;海洋中尺度渦與內波的地震圖像[J];地球物理學報;2009年11期
8 陳捷;陳標;許素芹;;一種新的SAR圖像內波檢測與參數估計方法[J];中國圖象圖形學報;2009年01期
9 陳標;侯海平;陶榮華;丁燦;;艦船尾跡SAR圖像特征的譜分析研究[J];艦船科學技術;2008年06期
10 甘錫林;黃韋艮;楊勁松;周長寶;史愛琴;金為民;;基于希爾伯特-黃變換的合成孔徑雷達內波參數提取新方法[J];遙感學報;2007年01期
相關會議論文 前2條
1 高建虎;陳杰;張履謙;;基于壓縮感知和EMD的SAR海洋內波探測方法[A];第九屆全國信息獲取與處理學術會議論文集Ⅱ[C];2011年
2 何宜軍;趙進平;;TOPEX/POSEIDON衛(wèi)星高度計海洋潮波分析[A];第十三屆全國遙感技術學術交流會論文摘要集[C];2001年
相關碩士學位論文 前1條
1 李海艷;利用合成孔徑雷達研究海洋內波[D];中國海洋大學;2004年
，本文編號：2300663
本文鏈接：http://sikaile.net/kejilunwen/xinxigongchenglunwen/2300663.html
