摘要
由液晶显示器(LCD)的"抽样-保持"特性与人眼视觉系统(HVS)的运动跟踪特性引起的运动模糊现象可以近似地用sinc函数的频域模型来描述。故本文在论述采用sinc^(-1)预补偿模型对LCD运动图像去模糊的基本原理的基础上,导出了采用Volterra模型替代sinc^(-1)时的参数选取方法和选择模型阶数的具体条件,从而为Volterra预补偿系统的硬件实现提供了理论依据。仿真结果显示:基于Volterra模型的图像预补偿系统在性能方面优于原sinc^(-1)系统,在处理小速度图像运动时造成的模糊现象中的效果与原sinc^(-1)系统相近,而在高速度造成的运动模糊的改善效果中,Volterra预补偿系统具有优于原sinc^(-1)系统的特点。
The ＂LCD （ Liquid Crystal Display） Motion Blur＂ problem results from the inherent sample-and-hold nature of LCD and the motion tracking features of the HVS （ Human Visual System） can be described in Frequency Domain with the sinc function. This paper presents the fundamentals of LCD motion deblurring using the sinc-1 model which based on the pre-compensation system at first. In order to implement the deblurring theory in the Hardware Environment, the Volterra model instead of the sirtc^-1 one is considered to establish the pre-compensation system, because of the parameters and order of the former can be determined under certain conditions. Finally, the simulation indicates that the fitting Volterra system is better than the original sirtc^-1 system. When it is used to deal with the low speed motion images, both effects of the motion deblurring are similar. But in the high speed situation, the effect of the Volterra system is better than the sinc^-1 system＇s.
出处
《信号处理》
CSCD
北大核心
2010年第9期1419-1422,共4页
Journal of Signal Processing
基金
国家自然科学基金资助
