Research On Oceanic Internal Wave Detection And Parameter Extraction Technology Based On SAR Image

Oceanic internal wave is an important manifestation of the internal movement of sea water.Remote sensing observation is the main observation method in the field of oceanic internal wave monitoring,which provides a large amount of data support for the analysis and prediction of ocean internal wave in key sea areas.Oceanic internal wave monitoring has important significance and research value for marine environmental safety and battlefield auxiliary decision-making.Nowadays,more and more attention has been paid to the multi-scale security guarantee capability of the marine environment,which is also a research hotspot in the field of marine environmental information application.Therefore,the northern of the South China Sea is selected as the research area in this paper,and based on SAR image data,the detection and parameter extraction techniques of oceanic internal wave are deeply explored.Firstly,this paper introduces the background and significance of the research on oceanic internal wave detection and parameter extraction based on SAR image,and analyzes the research status of related technologies at home and abroad.Combined with the knowledge of the SAR image,the imaging principle and characteristics of oceanic internal wave in SAR image are analyzed,and the SAR image data format used in this paper is briefly introduced.Secondly,the SAR images are preprocessed according to the fact that the SAR image contains a lot of speckle noise.The filtering methods in space domain and frequency domain are selected to process the sample image,and the filtering ability for speckle noise and the edge retention ability of each filtering method are evaluated by using some evaluation indexes.The comparative experiment shows that the Refined-Lee filtering algorithm has the best comprehensive performance.So this paper select it as the noise reduction method for SAR image preprocessing.Histogram equalization is selected to enhance the SAR image after noise reduction,and the visual effect of enhanced image details is more significant.Thirdly,according to the characteristics of oceanic internal waves in SAR images,gray level co-occurrence matrix and fractal dimension are introduced to extract features of a large number of SAR image samples.The differential box dimension method is improved from three aspects: the calculation formula of the number of boxes,the limitation range of the grid scale and the movement mode of the grid.The fractal dimension calculated by this method is more suitable as the basis for the detection of oceanic internal wave than the eigenvalue of gray level co-occurrence matrix.In addition,the oceanic internal wave peak line in the SAR image is extracted from the angle of edge detection.Based on the research of various edge detection operators,this paper proposes a new method applied to the field of SAR image oceanic internal wave detection.An edge feature vector is established considering of the edge property,then the FCM clustering algorithm is used to divide the pixels into edge points and non-edge points.It is proved by contrast experiments that this method can extract the internal wave peaks better than other methods,and it still a good effect in detecting different forms of oceanic internal waves.Finally,this paper selects image measurement method and empirical mode decomposition method to extract the half amplitude width of oceanic internal waves in SAR images.Then the two parameter extraction methods are introduced in detail.The advantages and disadvantages of the two methods are verified by different examples.On the basis of obtaining the half-amplitude width of the internal wave,the historical hydrological data and the relationship between parameters are combined to obtain more parameters related to the oceanic internal wave.The SAR image oceanic internal wave detection and parameter extraction system is constructed,and the software is realized,applying theory to practice.