实时小波滤波器在空间相机像移速度滤波中的设计与实施

The design and implementation of real-time wavelet filter on filtering for image movement velocity of space camera

通常情况下,通过多参量计算得到的空间相机像移速度中含有随机误差,必须要对其进行滤波处理,否则将会影响像移补偿的控制精度,导致图像质量下降。通过对像移速度误差统计特性的分析,提出了一种基于小波变换的处理对策。为实时处理并应用,设计了基于滑动窗分解和重构的处理结构,引入了Donaho阈值降噪算法,并设计了基于SOC的硬件滤波芯片。通过滤波处理后信噪比提高了9.78dB,取得了理想的滤波效果,适合在实际工程中采用。

Normally,the image movement compensation velocity calculated and obtained by multiple parameter contains random error,so it must be processed by some filtering algorithms.Without filtering,the control precision of image movement compensation will be decreased,whice will cause bad image quality.By analyzing the statistical characteristics of error of image movement velocity,a method based on wavelet transform is proposed.For application for real-time processing and application,a structure based on moving window about decomposition and reconstruction is designed and an algorithm called threshold noise reduction by Donoho is proposed.Furthermore,the hardwae filtering chip based on SOC is designed.The SNR which can be calculated by filtering is improved by 9.78dB,which proves that the filtering works well and is fit for application of actual engineering.