摘要
提出了基于小波分解的医学显微图像融合。通过先将图像进行小波分解,并对小波分解后的不同频率域分别进行了讨论,即分别讨论了选择高频系数和低频系数的原则。选择低频系数时,设计了两种选择方案,即区域能量法和边缘选择法。选择高频系数时,设计了两种选择方案,即区域能量法和目标清晰与模糊判据法。最后选择了在低频系数时采用边缘选择法,在高频系数时采用目标清晰与模糊判据法。从仿真结果的比较来看,在将多个目标聚焦图像融合后,图像能很好地保留多幅原来图像清晰部分的信息,融合后,提高了整幅图像的信噪比,从而获得了整幅高清晰图像。
Fusion medical microscopic image using wavelet decomposition is developed.At first,a wavelet decomposition of each source image is performed,and then different choice of high frequency coefficient and low frequency coefficient are discussed.The method of region energy and method of extracting the edge information in low frequency coefficient and method of region energy and method of decision of in-focus objects and out-focus objects in high frequency coefficient are introduced.The method of extracting the edge information in low frequency coefficient and method of decision of in-focus objects and out-focus objects in high frequency coefficient are adopted.Experiments show that the image edge and stripe detail are more obvious after image fusion.
出处
《光学技术》
CAS
CSCD
北大核心
2006年第z1期471-474,共4页
Optical Technique
关键词
显微图像融合
小波变换
融合算法
microscopic image fusion
wavelet transform
fusion algorithm
作者简介
符强(1976-),男,海南省人,桂林电子科技大学计算机系硕士研究生,主要从事图像处理方面的研究.E-mail:gjd@gliet.edu.cn
