This paper focuses on the state-of-the-art technology which is useful for medical diagnosis and proper treatment planning. Using this scheme, different data formats such as MRI (magnetic resonance image), CT (computed tomography), PET (positron emission tomography), and SPECT (specialized positron emission tomography) of the same patient can be registered. These medical images provide complementary information which is conflicting occasionally due to nonalignment problem. However, the registered image provides more information for medical personals. In the registration process, images are aligned with each other and the size of the object is made equal. So in this process, the nonaligned image is transformed with respect to the reference image. Here, we have registered the biomedical images by maximizing the mutual information. Genetic algorithm (GA) is used to optimize rotation, scaling and translation parameters. Results presented reveal the suitability of the proposed method for biomedical image registration.
[1]
H. Kim,et al.
A Method for Image Registration by Maximization of Mutual Information
,
2006,
2006 SICE-ICASE International Joint Conference.
[2]
R. Suganya,et al.
MUTUAL INFORMATION AND GENETIC ALGORITHM BASED REGISTRATION OF MRI BRAIN IMAGES
,
2011
.
[3]
Barry R. Masters,et al.
Digital Image Processing, Third Edition
,
2009
.
[4]
David E. Goldberg,et al.
Genetic Algorithms in Search Optimization and Machine Learning
,
1988
.
[5]
Nemir Ahmed Al-Azzawi,et al.
MRI Monomodal Feature-Based Registration Based on the Efficiency of Multiresolution Representation and Mutual Information
,
2012
.
[6]
Rafael C. González,et al.
Digital image processing using MATLAB
,
2006
.