一种从SAR影像到光学影像的翻译方法

An Approach to Translate SAR Image into Optical Image

光学传感器在夜晚和云雨天气难以成像,合成孔径雷达(synthetic aperture radar,SAR)虽然能够全天时、全天候工作,但其成像难以理解,对此提出利用SAR影像翻译为光学影像的新思路来弥补二者的缺陷。给出了遥感影像翻译定义,提出一套包含图像理解、目标转换等环节的影像翻译技术流程。通过支持向量机分类、种子填充和基于样本的纹理合成算法等手段实现SAR影像典型目标向光学影像的转换与表达。最后,利用该方法实现了ENVISAT-ASAR转换为Landsat TM,ALOS PALSAR转换为GeoEye的两类影像翻译,并利用SAR影像翻译结果修补光学影像空缺。影像翻译和补缺实验证明了SAR影像翻译为光学影像的可行性和有效性。

Synthetic aperture radar （SAR） sensors can work at all times and under all weather conditions, but SAR images are less intuitive and more difficult to understand. To complement advantages of optical and SAR sensors, a technique of image translation is put forward. Firstly, the definition of remote sensing image translation is presented. Then, some specific solutions for image understanding, and object transformation which are considered as key steps for image translation are proposed. Fea- ture extraction, Support Vector Machine classification, and sample-based texture synthesis algorithms are adopted to translate typical classes of SAR data into optical images. Finally, two kinds of SAR im- ages with different resolution are translated into Landsat TM and GeoEye images respectively, and the translated result could be applied to fill the blanks in those incomplete optical images. Overall, the re- search indicates that the proposed techniques of image translation from SAR to optical data are rational and effective.