灰度文本图像自适应二值化滤波算法设计及应用

On the adaptive binarization filtering algorithm for gray character images and its application

将自适应滤波与局部阈值二值化有机结合起来,提出自适应二值化滤波算法,旨在将淹没在噪声和失真中的灰度文本图像的二值图像信息通过滤波的方式恢复出来。采用改进的大津法计算局部阈值,对滤波结果进行二值化。依据二值化的结果估计滤波的误差,并调整滤波器参数。实验结果表明该算法克服了光照不均和噪声对灰度文本图像二值化的影响,得到了较理想的二值图像。

In order to restore the binary image information from the gray character image with noise and distortion through the filtering method, an adaptive binarization filtering algorithm is put forward based on the combination of adaptive filtering with local threshold binarization. The local threshold is selected through an improved Da-Jin algorithm and the filtering result is converted to binary data. Then the filtering error is estimated and filter parameters are adjusted based on the binary data. The experiment results show that with the algorithm, the influence of uneven light and noise on the binarization of gray character images has been overcome and an excellent binary image is obtained.