TITLE:
Multimodal Medical Image Fusion in Non-Subsampled Contourlet Transform Domain
AUTHORS:
Periyavattam Shanmugam Gomathi, Bhuvanesh Kalaavathi
KEYWORDS:
Image Fusion, Non-Subsampled Contourlet Transform (NSCT), Medical Imaging, Fusion Rules
JOURNAL NAME:
Circuits and Systems,
Vol.7 No.8,
June
15,
2016
ABSTRACT: Multimodal medical image
fusion is a powerful tool for diagnosing diseases in medical field. The main
objective is to capture the relevant information from input images into a
single output image, which plays an important role in clinical applications. In
this paper, an image fusion technique for the fusion of multimodal medical
images is proposed based on Non-Subsampled Contourlet Transform. The proposed
technique uses the Non-Subsampled Contourlet Transform (NSCT) to decompose the
images into lowpass and highpass subbands. The lowpass and highpass subbands
are fused by using mean based and variance based fusion rules. The
reconstructed image is obtained by taking Inverse Non-Subsampled Contourlet
Transform (INSCT) on fused subbands. The experimental results on six pairs of
medical images are compared in terms of entropy, mean, standard deviation, QAB/F as performance parameters. It reveals
that the proposed image fusion technique outperforms the existing image fusion
techniques in terms of quantitative and qualitative outcomes of the images. The
percentage improvement in entropy is 0% - 40%, mean is 3% - 42%, standard deviation
is 1% - 42%, QAB/Fis 0.4% - 48% in proposed method comparing to
conventional methods for six pairs of medical images.