期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2012
卷号：35
期号：2
页码：141-148
出版社：Journal of Theoretical and Applied
摘要：Now a days clinical diagnosis is emerging with digital images. In this paper we have dealt with computed tomography (CT), positron emission tomography (PET) and Magnetic Resonance Image (MRI). Where CT is rich in denser tissue with less distortion. PET with border of anatomical structure are blurred. MRI image has the complementary property, which provides better information on soft tissues but with more distortion. By combining any of these two complementary images we end up with new image which contains denser tissue with lesser distortion. For this process we use Nonsubsampled Contourlet Transform to decompose the images and Pulse Coupled Neural Network is used to motivate the lower frequency pixels. Thereby the fused image�s spatial property is improved. The exact edges of the fused images are found by applying it to a canny edge detection method to find tumour present in the FUSED image.
