Research On Retrieving Sea Surface Elevation Based On X-band Navigation Radar Images

As well known,it is important to understand the waves and their associated characteristics for military and civilian.With the rapid development of marine detection technology,using X-band navigation radar on the ocean to conduct a large area measurement and real-time monitoring is becoming a research hotspot of ocean remote sensing in recent years.It also has a greatly important guiding significance for the security of the port and the sea operations by using the collected radar image to retrieve various sea information,such as the significant wave height,sea surface elevation.Currently,the use of X-band navigation radar image to retrieve sea surface elevation is one of the hot issues of international research,but the theoretical research and engineering applications of domestic in this area are still blank.Based on the background and significance of the subject,the research results of retrieving sea surface elevation based on X-band navigation radar at home and abroad are summarized in this paper.The imaging mechanism of X-band radar image and the basic principle of retrieving the significant wave height and sea surface elevation are introduced.And then,both the key problems during the retrieving,such as the target interference detection,the design of bandpass filter,extracting significant wave height and the retrieving sea surface elevation are deeply researched.In addition,the effectiveness of research methods is proved in the experiments.The main contents are as follows:By investigating the sparse characteristics of the target in the navigation radar echo signal in this paper,the intrinsic relation between the electromagnetic scattering characteristic of the target of the radar image and the compressed sensing theory is revealed.Thus,the research on target detection is carried out based on the compressed sensing theory.Currently,the MP and OMP detection algorithms based on compressive sensing require knowing the sparsity of radar signal in advance so as to design the iteration times.In this paper,the existing detection algorithms are improved.And then,a SAMP target detection method which can adaptively finish the detection task without the prior information is proposed.The real navigation radar images are used to validate the effectiveness of the algorithm.The experimental results show that the designed sparse adaptive target detection method can accomplish the signal detection task without a priori information.The existing broadband filter ignores the influence of water depth on the upper and lower boundaries of the band-pass filter,and the bandwidth of the broadband filter will be greatlyamplified on the moving platform.Thus,the inversion results contain a large amount of noise and the inversion accuracy is reduced.Against the defects that the traditional broadband filter can not work effectively in moving condition,a new broadband pass filter which has the characteristic of following the wave energy distribution is designed based on the dispersion relation.The novel filter not only inherits the advantage of the broadband filter,which does not need to estimate accurately the surface current,but also takes the moving speed of the radar platform into the calculation equation of band-pass,which makes the bandwidth in motion state the same as the static state.Thus,the novel filter effectively solves the problem that the broadband of the traditional filter will increase with the increasing of the speed of movement.Compared to the results of commercial WaMoS and WAVEX systems,the results show that the new band-pass filter can accurately extract the ocean wave signal from the image spectrum,and is suitable for the filtering of the sea clutter image in the dynamic environment.The method of estimating the significant wave height based on the shadow statistics of radar images has the advantage of being less susceptible to environmental influences and without external reference.It is found that the existing method not only uses the strategy of fixed block,but also neglects the influence of the water depth on the inversion formula of the wave height,when using the geometric shadow method to retrieve the significant wave height from the navigation radar image.Thus,it leads to generating large error for shadow ratio calculation under different sea conditions and missing the effective wave information.And the inversion effect is not ideal in nearshore area.To these problems,the traditional geometric shadow method of retrieving the significant wave height is improved by using the stratigies of the adaptive block and the shallow water condition correction in this paper.Therefore,the improved algorithm not only can divide accurately the shadow image but also is suitable to the nearshore shallow water area.The effectiveness of the algorithm is verified by using the measured shore-based data.The estimation accuracy of the significant wave height is improved after using the stratigies of the adaptive block and the shallow water condition correction.Currently,the inverse 3DFFT method of retrieving the sea surface elevation from the navigation radar image has the advantage of fast speed,but the resolution of the retrieved sea surface elevation is limited by the time and spatial resolution of the original radar image.In addition,the wave number and frequency of the single wave component extracted do not strictly satisfy the dispersion relation and the problem of spectrum leakage exists after the radar image sequence is transformed.These problem will seriously affect the inversionaccuracy of the sea surface elevation.To solve these problems,a new method to extract wave components,whose wave number and frequency strictly satisfy the dispersion relation,and retrieving sea surface elevation by superposing the extracted wave components is proposed by studying the sea clutter model in this paper.This method not only avoids the spectral leakage problem in the spatial domain but also allows the free choice of time and spatial resolution,which has important value in practice.The simulated radar images are used to verify the effectiveness of the algorithm.The experimental results show that the accuracy of the extracted wave components and the retrieving sea surface elevation are better than that of the existing method.