Research On The Wake Search Method Of Surface Ships Based On Dot-Curve System

Since 2020,with the new crown pneumonia epidemic sweeping the world,the world structure has accelerated and the international military situation has become more complex and turbulent.Recently,Japan and the United States have made waves on the Diaoyu Islands issue and challenged China’s bottom line.Therefore,my country has an urgent need for real-time control of territorial sea dynamics.Remote sensing images provide a better monitoring method.Even for "stealth" ships,the existence of ship wakes is inevitable.And it can reverse the information such as the size and heading of the ship.Therefore,the search and capture of the ship’s wake is particularly important.Therefore,this article mainly focuses on the search method of ship wakes on the sea surface under remote sensing images.The main research contents are as follows:First,the research scenario of this article is clearly defined,and the Dot-Curve description system is introduced innovatively.Combined with the Dot-Curve description system,ship wakes and various sea backgrounds were modeled and analyzed.The characteristic law of the ship wakes and various background was characterized and elaborated,and a good characteristic law distinguishing the wake of the ship from the background was obtained,and verified under multi-wavelength,which proved the completeness of the characteristic law of the obtained wake.After that,a ship wake search algorithm based on the Dot-Curve system is proposed.The spindle correction strategy based on residual analysis is innovatively introduced to improve the positioning accuracy of the algorithm.At the same time,combined with the relevant difference of the ship wake and background in the Dot-Curve system,all the obtained suspected targets were characteristic screened,and then matched with the bow of the ship to realize the accurate search and capture of the wake of the ship,greatly reduced the false alarm rate.After that,multiple sets of samples containing wakes are selected to test the algorithm,which proves the effectiveness and good robustness of the algorithm.Finally,the algorithm is analyzed and verified by experiments under multi-band,which proves the universality and completeness of the algorithm again.Finally,the search performance analysis of the algorithm under short-wave infrared and long-wave infrared is carried out.By comparing the calculated detection rate and false alarm rate indicators with other literature methods,the superiority of the algorithm is proved.At the same time,it also verified the good applicability of the algorithm for the medium wave band not included in the Landsat-8 remote sensing satellite.Then,it analyzed the causes of missed detection and misjudgment in each band.Finally,the detection performance of the target search algorithm is summarized and analyzed under the influence of multi-band,different seasons and glare,and suggestions for the selection of bands and spectrum bands corresponding to various situations in practical engineering applications are given.