一种基于数学形态学的二值图像去噪算法

A Noise Reduction Algorithm for Binary Image Based on Mathematical Morphology

为处理强噪声二值图像中的字符,文章提出了一种基于数学形态学的去噪算法。该算法利用形态学的腐蚀操作提取出字符主干,再通过膨胀操作提取出字符较细的部分,从而有效地去除图像噪声。实验结果表明,本文算法具有较强的字符提取能力和去噪能力,并能有效地保护字符较细的部分。

A noise reduction algorithm based on mathematical morphology is proposed for processing characters in binary image containing heavy noise.Through erosion operation of the mathematical morphology,the algorithm extracts the wide parts of the character,then extracts the thin parts of the character through dilation operation.Thus,the noise is efficiently reduced.The experimental result indicates that the algorithm has very strong ability to extract the characters and reduce noise,and can efficiently protect the thin parts of the characters.