一种新的去除视觉文本图像椒盐噪声的算法  

New algorithm for removal of salt-pepper noises of visual text images

作　　者：田大增[1] 郝永[2] 哈明虎[2] 

机构地区：[1]河北大学物理科学与技术学院,河北保定071002 [2]河北大学数学与计算机学院,河北保定071002

出　　处：《计算机工程与应用》2007年第14期81-83,共3页Computer Engineering and Applications

基　　金：国家自然科学基金(No.60573069);河北省科学技术研究发展计划项目(No.03213501);河北大学重点资助课题(No.2003Z12);教育部科学技术研究重点项目(No.206012);河北省教育厅重点资助项目(No.2005001D)~~

摘　　要：针对视觉文本图像,在获取、处理、传输中产生的椒盐噪声,提出一种基于文字笔画特性消除椒盐噪声的新算法。该算法智能判断被处理像素及其邻域内的像素点是否为噪声,从而有效地把文字目标和椒盐噪声加以区分。实验结果显示,相比传统的去噪和平滑算法,该方法能准确地抑制噪声,同时避免了对文字笔画的影响,减少了对后继识别的干扰。For the salt-pepper noise which deduces in the gaining,processing,transmission of the visual text images,a new algorithm for removal of salt-pepper noises is proposed,which based on the characters stroke characteristic.This algorithm can estimate whether a pixel and it＇s neighboring pixels are salt-pepper noises or not,and differentiate effectively characters from salt- pepper noise.The experimental results show clearly that this algorithm can restrain salt-pepper noises well and truly,at the same time,comparing with those traditional denoise algorithms,it avoids the influence for characters stroke,and decreases the disturbance for subsequent recognition.

关 键 词：椒盐噪声 视觉文本图像 文字笔画 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]