基于LS-SVM的LCD运动图像模糊的改善研究  

Study of LCDs Motion Blur Based on Least Squares Support Vector Machine

作　　者：徐永键[1] 李丹[1] 谭洪舟[1] 

机构地区：[1]中山大学信息科学与技术学院,广州510275

出　　处：《信号处理》2009年第8期1209-1214,共6页Journal of Signal Processing

基　　金：国家自然科学基金(60874060)资助

摘　　要：运动模糊是影响LCD图像质量的主要问题。LCD运动模糊是由LCD慢响应、"抽样-保持"特性与HVS的低通滤波特性所引起。本文提出了一种基于最小二乘支持向量机(LS-SVM)的运动去模糊方法。该算法从系统辨识角度,通过对样本数据的训练,构造预补偿系统的预测PSF模型。最后,使用该预测PSF模型实现运动图像的去模糊过程。仿真结果显示,该算法具有比较理想的效果,可以应用于实际的LCD运动图像去模糊。The primary＂ image quality issues of LCD is the motion blur. LCD motion blur is caused jointly by the slow LCD＇ s temporal response and the hold-type LCD＇ s rendering method together with the motion pursuing function of human visual system （HVS）. In this paper we develop a method for motion blur reduction using the Least Square Support Vector Machine（LS-SVM） algorithm. From the point of system identification, this algorithm constructs the predictive PSF model of the pre-compensation system through training the sample data. And then, using the predictive PSF model to realize the motion de-blurring. Simulation results show that the algorithm has a good effect, and can be used for actual LCD motion de-blurring.

关 键 词：运动模糊 LS-SVM 系统辨识 预补偿系统 PSF模型 

分 类 号：TN941[电子电信—信号与信息处理]