光子学报, 2013, 42 (3): 354, 网络出版: 2013-03-05
基于小波包和Niblack法的枪号图像二值化算法
Gun Code Binary Image Algorithm Based on Wavelet Packet and Niblack Method
摘要
针对枪号图像字符易磨损、对比度太大或太小的特点, 提出了小波包滤波器浮动阈值和Niblack法相结合的二值化算法.首先采用小波包滤波器对图像滤波, 把得到的低通图像作为浮动阈值, 并对图像进行二值化, 得到第一幅待融合图像; 然后采用Niblack法对图像进行二值化, 得到第二幅待融合图像; 最后把两幅图像按照灰度最小的方法进行融合, 得到最终的二值化图像.实验结果表明该综合算法在二值化效果、适用范围等方面均有良好的表现, 具有一定的推广价值.
Abstract
Aiming at the characteristics of the gun code image that characters are easy to wear and contrast gradient is too large or too small, a binarization method based on wavelet packet filter floating threshold and Niblack method is proposed. Firstly, the image is filtered using wavelet packet filter and binarized with the low pass image after filtering. Then, the image is binarized with the Niblack method to obtain the second fusing image. Finally, two fusing images are merged based on the minimum principle of image gray values to get the final binarization image. Experimental results show that the method has a good performance in binarization effect and adaptation range and can imply in many situation.
申森, 李艾华, 姚良, 马治明. 基于小波包和Niblack法的枪号图像二值化算法[J]. 光子学报, 2013, 42(3): 354. SHEN Sen, LI Ai-hua, YAO Liang, MA Zhi-ming. Gun Code Binary Image Algorithm Based on Wavelet Packet and Niblack Method[J]. ACTA PHOTONICA SINICA, 2013, 42(3): 354.