Fusion of multi‐modal lumbar spine images using Kekre's hybrid wavelet transform
| Autor: | Dhirendra Mishra, Bhakti Palkar |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Image fusion
business.industry Computer science Wavelet transform Image registration 020206 networking & telecommunications Pattern recognition 02 engineering and technology Image (mathematics) Transformation (function) Wavelet Hadamard transform Signal Processing 0202 electrical engineering electronic engineering information engineering Discrete cosine transform 020201 artificial intelligence & image processing Computer Vision and Pattern Recognition Artificial intelligence Electrical and Electronic Engineering business Software |
| Zdroj: | IET Image Processing. 13:2271-2280 |
| ISSN: | 1751-9667 1751-9659 |
| DOI: | 10.1049/iet-ipr.2018.5609 |
| Popis: | This paper analyses the effects of image fusion on CT and MR images of lumbar spine. Medical image fusion generates a new image which has important features from each source image, thereby making it more informative and hence useful for radiologists and surgeons to make appropriate diagnostic decision. Both monomodal and multimodal medical images can be fused. Pre- and post-operative monomodal images can be fused to see the results of treatment or operation. Multimodal medical images can be fused for treatment planning. Since CT and MR images are completely different, they need to be strictly registered or aligned with each other before fusing. Kekre’s wavelet transformation approach has been followed to generate wavelets using seven different orthogonal transforms. Outcomes of all the wavelet transforms are compared using quantitative and qualitative methods. Quantitative assessment indicates that Hadamard and DCT wavelet transforms give best results. Qualitative assessment indicates that all the fused results look similar and more informative than input images. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Plný text