Front Inner Page - Volume 2 No.4 August 2015

  • » Back to Index

  • Title

    :

    An Efficient Daubechies Complex Wavelet Based Multi-Resolution Approach For Multimodal Medical Image Fusion

    Authors

    :

    Ussain Basha.Goresahebgari1, Nagaraja Kumar.Neravati2

    Keywords

    :

    Multi Spectral images (MS), Non-Subsampled Contourlet Transform (NSCT), Stationary Wavelet Transform (SWT), Lifting Wavelet Transform (LWT), Structure Similarity (SSIM), Multi Wavelet Transform (MWT). 

    Issue Date

    :

    August – 2015

    Abstract

    :

    Multimodality approach for medical images provides multiple advantages for the detection, diagnosis and management of the diseases. Image fusion techniques are used to combine the high resolution images with the color information of the low resolution Multi Spectral (MS) images to produce a high resolution MS image. In this paper, a novel Daubechies complex wavelet transform based multi-resolution approach is proposed to fuse the brain images obtained from Open Access Series of Imaging Studies (OASIS) and Simulated Brain Database(SBD) datasets. Initially, the input images from the datasets are provided as an input to the Non-Subsampled Contourlet Transform (NSCT). It converts the images into high frequency and low frequency images. The high and low frequency sub bands are provided as an input to the Stationary Wavelet Transform (SWT) and Lifting Wavelet Transform (LWT). The SWT is used to prevent the lack of translation and invariance in the Discrete Wavelet Transform (DWT). The LWT splits the input images into odd and even set of samples and the lifting steps enhances the details of the images. The Discrete Complex Wavelet Transform (DWCT) captures both the frequency and location information of the images. In the Multi Wavelet Transform (MWT), the smoothness of the contour of the images is obtained in various elongated shapes and in various directions. The curvelet transform is used to highlight the edges of the multidirectional images. The Daubechies complex wavelet transform produces an approximate fused image. With reference to the approximate result, inverse NCST is used to generate the reconstructed image. The result obtained from the Inverse NCST is considered as the actual fused image. The performance of the proposed Daubechies complex wavelet transform is measured using the metrics, such as, entropy, Structure Similarity (SSIM), mutual information and time delay.  

    Page(s)

    :

    1-9

    ISSN

    :

    2347- 4734

    Source

    :

    Vol. 2, No.4, August 2015

    Download

    :


  • » Back index