空间遥感相机星上图像复原系统设计

A Real-time Digital Image Restoration System Design of Space Remote Sensing Camera

为了能够在轨、实时对空间遥感图像做增强处理,设计了一个基于现场可编程门阵列(FPGA),应用于空间遥感相机的星上实时图像复原系统,实现在轨实时处理遥感图像,增强图像清晰度并有效抑制噪声。图像复原系统采用模块化设计,调制传递函数(MTF)在轨测量计算模块实现了边缘扩展函数计算、线扩展函数计算、归一化MTF计算,以及调制传递函数补偿(MTFC)参数计算与存储等功能;MTFC图像滤波抑噪算法模块实现了二维图像实时滤波以及抑制噪声功能。使用ISE软件的系统生成器模块设计了图像处理算法,简化了系统设计和验证流程。复原系统运行结果表明,图像经过实时复原处理后,图像灰度梯度、点锐度、边缘能量、方差都有大幅提高;复原后图像信噪比与原始图像相比降低很小,对于MTF的中高频域增强。图像复原系统采用模块化设计,实时复原效果好,可灵活应用于多种场合。

In order to process space remote sensing images on orbit in real time, a real-time digital image restoration system for space camera was designed based on field programmable gate array （FPGA） in the paper, which can enhance image definition and suppress the noise effectively. Modular design was adopted in the system. The modulation transfer function （MTF） measurement and calculation module on-orbit has the function of calculating many characteristics, such as the edge spread function （ESF）, line spread function （LSF）, ESF difference operation, normalization MTF and MTFC parameters. The modulation transfer function （MTFC） image filtering and noise suppression module has the function of filtering and effectively suppressing the noise. The system generator software in ISE was used to design the image processing algorithms, thus simplifying the design structure of system and the process redesign. The real-time processing results of the restoration system showed that the gray gradient, dot sharpness, edge energy and variance all had a huge increase after restoration process, the signal to noise ratio （SNR） of the recovered image was very close to the SNR of original image. Especially the SNR was dramatically enhanced at medium-high frequency of image. The designed image restoration system can be widely used in various fields with better restoration effect.