Key Technology Research On SAR-to-Optical Translation Based On Deep Learning

To accurately describe dynamic vegetation changes,high temporal and spectral resolution data are urgently required.Optical images provide rich spectral information;however,nobody can ensure the temporal resolution of it.Conversely,synthetic-aperture radar(SAR)is not limited by lighting conditions,climate,or other environmental factors;thus,it can produce images continuously and in all weather conditions,generating time series with high temporal resolution.However,an important limitation of SAR images is that the spectral information is insufficient to recognize certain vegetation changes.Determining the relationship between optical and SAR images can allow us to use SAR data as the input,in the absence of optical data,to generate images similar to the optical images.The generated and existing optical images can then form a complete dataset containing rich spectral information and high temporal resolution,which can be used for accurate and comprehensive analysis of vegetation coverage and changes.The process of generating optical images with SAR images as input can be called SAR-to-Optical image translation,which also has a significate meanings.However,SAR-to-Optical image translation is difficult to accomplish using a simple physical model.Deep learning can effectively simulate complicated relationships by performing im-ageto-image translation tasks.More specifically,conditional Generative Adversarial Networks(cGANs)can be employed to efficiently translate SAR images to optical images,and have been proved to be suitable in the SAR-to-Optical translation process.Therefore,this study employs cGANs to transform SAR images into optical images,then explores the impact of edge information on the image generation process.Additionally,the effects of three different polarization modes are compared:co-polarization(VV),cross-polarization(VH),and dual-polarization(VV&VH).The major findings are as follows.(1)The addition of edge information improves the structural similarity between the generated optical remote sensing image and the original optical remote sensing image in brightness,contrast and structure,makes the boundaries between surface objects clearer in the generated image,and provides the cGANs with more effective information,resulting in better image quality when VH and VV&VH polarization modes are used as the input.(2)Overall,the accuracy of NIR and SWIR bands in the generated image is higher than that of visible bands(for Landsat8 images,bands 5-7 are more accurate than bands 2-4).(3)The optimal polarization mode with edge information added in the input is VV&VH,whereas the optimal polarization mode without edge information is VV.(4)Moreover,different surface object types have different optimal input features.For example,VV&VH polarization with edge information is the optimal input for vegetation,VV polarization with edge information is the optimal input for water bodies,and VV polarization without edge information is the optimal input for building land.The conclusions of this study could serve as an important reference for selecting cGANs input features,and as a potential reference for the applications of cGANs to the SAR-to-Optical translation of other multi-source remote sensing data.