Abstract
Three dimensional (3D) Magnetic Resonance Imaging (MRI) reconstructions depend heavily on the imaging speed. Magnetic Resonance (MR) images consist of large volume of redundant and sparse data. Therefore, the need to reduce this data without degrading the image information. In Fourier Domain, sparse nature of MR images enables image reconstruction with fewer Fourier coefficients. Fourier Transform (FT) maps the image into the frequency domain using fixed and same size window throughout the analysis. In our paper, a method to perform compressive sensing for MR image is presented. Anisotropic filtering using Active Contour Modelling is performed to smoothen the image in order to preserve edge information. MR image is converted into Fourier Domain using Discrete Fourier Transform (DFT). l1 and l2 reconstruction algorithms are used to reconstruct the images using minimum coefficients that have maximum information.
Original language | English |
---|---|
Title of host publication | Recent Trends in Image Processing and Pattern Recognition - 2nd International Conference, RTIP2R 2018, Revised Selected Papers |
Editors | K.C. Santosh, Ravindra S. Hegadi |
Publisher | Springer Verlag |
Pages | 294-302 |
Number of pages | 9 |
ISBN (Print) | 9789811391835 |
DOIs | |
Publication status | Published - 01-01-2019 |
Event | 2nd International Conference on Recent Trends in Image Processing and Pattern Recognition, RTIP2R 2018 - Solapur, India Duration: 21-12-2018 → 22-12-2018 |
Publication series
Name | Communications in Computer and Information Science |
---|---|
Volume | 1036 |
ISSN (Print) | 1865-0929 |
Conference
Conference | 2nd International Conference on Recent Trends in Image Processing and Pattern Recognition, RTIP2R 2018 |
---|---|
Country | India |
City | Solapur |
Period | 21-12-18 → 22-12-18 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Computer Science(all)
- Mathematics(all)
Cite this
}
Compressive Sensing for Three-Dimensional Brain Magnetic Resonance Imaging. / D’souza, Selrina; Anitha, H.; Kotegar, Karunakar.
Recent Trends in Image Processing and Pattern Recognition - 2nd International Conference, RTIP2R 2018, Revised Selected Papers. ed. / K.C. Santosh; Ravindra S. Hegadi. Springer Verlag, 2019. p. 294-302 (Communications in Computer and Information Science; Vol. 1036).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
TY - GEN
T1 - Compressive Sensing for Three-Dimensional Brain Magnetic Resonance Imaging
AU - D’souza, Selrina
AU - Anitha, H.
AU - Kotegar, Karunakar
PY - 2019/1/1
Y1 - 2019/1/1
N2 - Three dimensional (3D) Magnetic Resonance Imaging (MRI) reconstructions depend heavily on the imaging speed. Magnetic Resonance (MR) images consist of large volume of redundant and sparse data. Therefore, the need to reduce this data without degrading the image information. In Fourier Domain, sparse nature of MR images enables image reconstruction with fewer Fourier coefficients. Fourier Transform (FT) maps the image into the frequency domain using fixed and same size window throughout the analysis. In our paper, a method to perform compressive sensing for MR image is presented. Anisotropic filtering using Active Contour Modelling is performed to smoothen the image in order to preserve edge information. MR image is converted into Fourier Domain using Discrete Fourier Transform (DFT). l1 and l2 reconstruction algorithms are used to reconstruct the images using minimum coefficients that have maximum information.
AB - Three dimensional (3D) Magnetic Resonance Imaging (MRI) reconstructions depend heavily on the imaging speed. Magnetic Resonance (MR) images consist of large volume of redundant and sparse data. Therefore, the need to reduce this data without degrading the image information. In Fourier Domain, sparse nature of MR images enables image reconstruction with fewer Fourier coefficients. Fourier Transform (FT) maps the image into the frequency domain using fixed and same size window throughout the analysis. In our paper, a method to perform compressive sensing for MR image is presented. Anisotropic filtering using Active Contour Modelling is performed to smoothen the image in order to preserve edge information. MR image is converted into Fourier Domain using Discrete Fourier Transform (DFT). l1 and l2 reconstruction algorithms are used to reconstruct the images using minimum coefficients that have maximum information.
UR - http://www.scopus.com/inward/record.url?scp=85069745466&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85069745466&partnerID=8YFLogxK
U2 - 10.1007/978-981-13-9184-2_26
DO - 10.1007/978-981-13-9184-2_26
M3 - Conference contribution
AN - SCOPUS:85069745466
SN - 9789811391835
T3 - Communications in Computer and Information Science
SP - 294
EP - 302
BT - Recent Trends in Image Processing and Pattern Recognition - 2nd International Conference, RTIP2R 2018, Revised Selected Papers
A2 - Santosh, K.C.
A2 - Hegadi, Ravindra S.
PB - Springer Verlag
ER -