The Research On Improving The Wind Field Inversion Algorithm From X-band Radar Images

Sea surface wind stress is the main power of the surface movement. For Marine fishery production, Marine exploration and development as well as Marine military action, accurate and real-time monitoring sea surface wind field has important significance. Contrast with traditional means and other remote sensing method, using marine radar to monitor wind field have the advantage, which can obtain real-time, high resolution wind field information in a wide range. Many scholars at home and abroad Study how to use the navigation radar method for obtaining wind field.This paper summarizes the research results of the scholars both at home and abroad. At the same time, this paper summarizes the experience and the existing algorithms. Provide a reference for further research and provide some help for beginners who want to enter this field.Through the review, we can found that the existing research has achieved a certain effect. But overall progress is slow, available technology is less. So there is a big development space. Especially the domestic research only stay on verify the validity of the algorithm abroad. There has been no research by exploring new ways to improve the inversion accuracy. In this paper, we through the following several aspects to improve the x-band navigation radar inversion algorithm:1) Through further suppress the noise to improve the inversion accuracy of the wind gradient method;2) To explore a new method of the inversion of the wind, which method uses spectrum;3) To explore fitting method for inversing wind speed;4) Fusion the optical flow algorithm and the fitting method results.Then the paper proposes improved method or the new algorithm respectively from the four research directions. Finally, validate the improved results through a lot of effective data, which obtain from Fujian and Taiwan strait and Binhai experiment. The results showed that, by suppressing noise can improve the stability and accuracy of gradient method inversion; Second, the spectrum algorithm has certain operability, but still needs to be further improved; Third, in view of the data quality control in advance, the mean absolute error of fitting method to the inversion of wind speed is lower than 1.5 meters per second; Four, by the fusion of optical flow algorithm and fitting method can improve the accuracy of fitting method.Finally, it should be pointed out that, the author participated in study of the 724 research institute of the China Shipbuilding Industry Corporation. To prove the feasibility of the algorithm, in order to apply the wind field inversion algorithm in the new marine radar, which have the very high research value.