摘要
根据显微图像的特点,从水平、垂直、+对角线、-对角线四个方向将二维图像转为一维数组,对不同的分解尺度上的高频系数采用不同的阈值处理,对阈值处理后的高频系数进行增强处理,进行一维小波逆变换,一维数组转化为二维图像,最后将四个二维图像进行非均权值计算得到去噪后的图像。在原子力显微镜得到的显微图像中进行仿真实验,与改进软阈值去噪算法和Keesook.J.Han提出的去噪算法相比较。结果证明多方向一维小波变换的显微图像去噪算法具有更好的去噪性能,边缘细节保持明显优于其他算法。
According to the characteristic of microscopic image,the two-dimensional image data is transformed into onedimensional array data from horizontal , vertical , ＋diagonal line,the diagonal line and four directions,by using different threshold value on the different wavelet decomposition high frequency coefficient on the different wavelet decomposition,enhancing the processed coefficient and changing one-dimensional array data into two-dimensional image data for obtaining de-noise image after inversing wavelet transform and calculating on four two-dimensional pictures by unequal coefficient. Simulation experiment is done with microscopic image which gets from atomic force microscope.The results indicate the proposed algorithm is effective both in reserving the edge and in removing noise compared with Keesook.J. Han algorithm and improving soft-thresholded denoised algorithm.
出处
《计算机工程与应用》
CSCD
北大核心
2007年第12期31-33,共3页
Computer Engineering and Applications
基金
国家高技术研究发展计划(863)(the National High-Tech Research and Development Plan of China under Grant No.2005AA26013)
测绘遥感信息工程国家重点实验室基金项目(No.WKL(03)0101)。
关键词
小波变换
显微图像
去噪
边缘细节保持
wavelet transform
microscopic image
de-noised
edge detail maintenance
