Non-Subsampled Contourlet Transform based Multimodal Medical Image Fusion and its Performance Evaluation
Shaik Afroz Begum1, K. Suresh Reddy2, M. N. Giri Prasad3

1Shaik Afroz Begum, Research Scholar, Department of ECE, JNTUA, (Andhra Pradesh), India.
2Dr. K. Suresh Reddy, Professor & Head, Department of ECE, G Pulla Reddy Engineering College, (Andhra Pradesh), India.
3Dr. M. N. Giri Prasad, Professor, Department of ECE, JNTUA, (Andhra Pradesh), India.
Manuscript received on 23 March 2019 | Revised Manuscript received on 04 April 2019 | Manuscript Published on 18 April 2019 | PP: 355-359 | Volume-7 Issue-6S March 2019 | Retrieval Number: F02700376S19/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: Integration of medical images from different sensors can create data that can’t be obtained by observing the images separately. Here, in this framework, the fusion of the images is implemented using non-subsampled Contourlet transform (NSCT) and this tool offers some special features like multi-resolution, invariant shifting, and multi-directional band decomposition tool. First, the input images of medical field decomposed into complementary low frequency and frequency of high range sub-bands by NSCT is implemented. Then, by considering the significance of these complementary sub images, a new selection method is implemented in different ways. This scheme use local energy rule to select the low-frequency band and the weighted sum of modified Laplacian (WSML) rule to select high-frequency directional bands. In the final step, the merged image is recovered by inverse NSCT tool implementation on merged bands. This effective novel fusion scheme is compared with existing traditional image merging rules in the transform domain. The results can reveal the efficiency of the novel fusion structure through visual and quantitative measures.
Keywords: Nonsubsampled Contourlet Transform, Computed Tomography (CT) Image, Magnetic Resonance Imaging (MRI), Local Energy, Modified Laplacian.
Scope of the Article: Image Security