Improving Research On The Algorithm Of Retrieving Ocean Wind Field From X-band Radar Images

Ocean surface wind is one of the important environmental parameters of oceandynamics, which has great influence on navigation, tourism, national defense,development of marine resources. So, it’s of significant value to monitor it. X-bandradar monitoring system for ocean dynamic environment is a new type of marineenvironment monitoring system, which has the advantages of small volume,convenient moving, good resolution and so on. It has a wide range of applications,one of which is to monitor ocean surface wind field. As ocean wind crossing the seasurface, it causes small roughness of sea surface, which has great impact on echosignal of the X-band radar. So, it contains the information of sea wind in the X-bandradar images. A great number of researches on retrieval of ocean surface wind fieldfrom X-band radar images have been carried out, one of which is the optical flowalgorithm. This method does not need correction of radar phase and it’s convenient tooperate. Besides, it’s suitable for business applications. As a result, the optical flowalgorithm is of great research value. However, as the inversion results of current flowalgorithm are still not desirable, it’s difficult to achieve business application.Therefore, it is of great significance to try to make several modifications of the opticalflow algorithm in this paper.Firstly, the principle of optical flow algorithm and the realization process ofretrieval algorithm of sea surface wind field are introduced, and each step of theoperation process is described in detail.Then, based on the existing achievements on the retrieval algorithm of seasurface wind field, the optical flow algorithm is modified from improving the seacurrent calculation and synthesizing different methods of solving optical flowequation. Specifically:1), Considering both the fundamental and high-order dispersion relation for the filtering of radar signal, the image spectrum of none wavesignal is removed through the new band-pass filter;2), The Bayesian method iscreatively introduced into the calculation of sea current, which fully consider theexperience information and data information. By using of Bayesian method, thestability and accuracy of sea current field is improved;3), According to the limitationof the Lucas-Kanade’s local method for solving the flow equations, Horn–Sehunek’sglobal smoothness constraint technique is introduced to solve the optical flowequation, and the integrated model of local and global method is put forward toimprove the model’s applicability and stability.Finally, combined with the anemometer and wave information, the data of the Xband radar image of Binhai and Longkou is used to verify the modified optical flowalgorithm. In contrast to the original optical flow algorithm, the results of modifiedalgorithm are more concentrated, stable and accurate. Therefore, it has certainresearch value to improve the original optical flow algorithm in this paper.