摘要
医学图像降噪必须做到既降低图像噪声又保留图像细节。通过对二维离散小波变换滤波去噪的研究以及实验表明。采用硬阈值法时,在去噪过程中如果阈值选取太小,降噪后的图像仍然有噪声,如果阈值太大,重要图像特性被滤掉,会引起偏差。因此对于不同尺度的小波系数应该选取不同的阈值进行医学图像处理。
Medical image denoising must do both to reduce image noise and retain image details.Research based on the two-dimensional discrete wavelet transform denoising filter and experiment.The hard threshold method in denoising process,if the threshold is too small,the denoised image is still noise,if the threshold is too large,an important characteristic of image is filtered out,will cause the deviation.The wavelet coefficients of different scales should select different thresholds for medical image processing.
作者
王静
WANG Jing(Zhengyang County Hospital of Traditional Chinese Medicine,Zhumadian 463000,Henan,China)
关键词
离散小波
变换滤波
去噪
Discrete wavelet
Transform filter
Denoising
