CCD扫描噪声建模及自适应滤波算法  

Modeling for CCD Scan Noise and Adaptive Filtering Algorithm

作　　者：王娜[1] 曾平[1] 孔月萍[1] 

机构地区：[1]西安电子科技大学计算机学院,西安710071

出　　处：《计算机工程》2008年第20期244-246,共3页Computer Engineering

基　　金：国家部委预研基金资助项目

摘　　要：为去除图像因扫描引入的电荷耦合器件(CCD)噪声,提出一种基于变窗口的自适应滤波算法。分析CCD主要噪声:散粒,固有和输出噪声的噪声特性,建立CCD扫描噪声模型,根据噪声和边缘处理方式的不同设计了局部和全局自适应滤波器。实验结果表明,该算法具有较好的信噪分离能力,在充分抑制噪声的同时,很好地保持了边缘,提高对比度,视觉效果良好。In order to remove CCD noise for scanner in the image, an adaptive filtering algorithm based on window-altered is proposed. This paper analyzes CCD noisy characteristic, which mainly includes： photon noise, fixed-pattern noise and amplifier noise. A model for CCD scanning noise is developed. According to different processing methods for noise and edge, local and overall adaptive filters are designed. Experimental results show that the proposed algorithm can separate the CCD noise and signal better than the traditional algorithm. The algorithm has the efficient edge preserving ability and high contrast gradient and excellent visual effect, while CCD noise is suppressed fully.

关 键 词：CCD扫描噪声 噪声模型 噪声和边缘 自适应滤波 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]