摘要
简要分析了有损JPEG压缩算法和几种经典的边缘提取算法;经对原始的和有损JPEG压缩的MRI图像的编程实验,分析了图像边缘的视觉效果和相异率,说明MRI图像经有损JPEG压缩后对图像的边缘和噪声都有所平滑改变和丢失;由于有损JPEG具有低通滤波的性质,所分析的几种经典的边缘提取算法具有高通滤波的性质,使MRI图像的边缘和噪声在JPEG压缩前后的差异随门限值的增大而变小。
The paper analyzed the JPEG compressing method with losing information and several classical methods of edge distilling in brief. The original MRI image and the MRI image that compressed by JPEG were programmed, the optical impression and ratio of dissimilitude of these images are showed, the result indicated that the edges and the noises in the MRI image would be smoothed and lost after compressing by the JPEG method. The JPEG method has the characteristic of lowpass and methods of edge distilling we adopted in this paper have the characteristic of anti-lowpass, which make the discrepancy of the edges and noises between former and later MRI images diminished with the augmentation of the threshold value.
出处
《信息技术》
2004年第12期14-18,共5页
Information Technology
基金
国家自然科学基金资助项目(40272016)
关键词
MRI图像
JPEG
边缘提取
相异率
门限值
MRI image
JPEG
edge distilling
ratio of dissimilitude
threshold value