摘要
针对文本图像在获取、处理和传输过程中产生的椒盐噪声,提出一种去除文本图像椒盐噪声的方法。该方法首先标记噪声点,然后采用Otsu阈值分割法将信号点分为文字与背景两类,最后对邻域内信号点进行扩散处理去除噪声。实验结果表明,相比传统的去噪算法,该方法能够减少笔画断裂和伪笔画的产生。
For the salt-pepper noise which generates in the getting,processing,transmission of the text images,a method of salt-pepper denoising for text images is presented.It first marks the noisy points.Then it uses the Otsu threshold method to divide the pixels into two classes,some are literal points and others are background points.At the end,it uses the diffusion method to processing noisy points.Comparing with those traditional de-noising algorithms,primary experiment suggests that this method can reduce characters stroke and reduce phony characters.
出处
《长春理工大学学报（自然科学版）》
2010年第2期129-132,共4页
Journal of Changchun University of Science and Technology(Natural Science Edition)
关键词
椒盐噪声
文本图像
中值滤波
均值滤波
salt and pepper noise
text image
median filtering
mean filtering
