Optimization of mutual information for multiresolution image registration

In order to reduce the noise sensitivity of the SAR (synthetic aperture radar) image registration, a image registration algorithm which basing on the ratio mutual information (RatioMI) is proposed in this paper. Firstly, the ratio images of the reference image and the floating image are gotten by using the ratio operator, and then take the two ratio images as a similar characteristic quantity to construct the similarity measure function which was used in the optimization process of the image registration experiment. The experimental results of the SAR image registration show that the new registration algorithm which based on the RatioMI is effectively in avoiding the local maxima point problems causing by speckle noise.

Download Full-text

Self-similarity weighted mutual information: A new nonrigid image registration metric

Medical Image Analysis ◽

10.1016/j.media.2013.12.003 ◽

2014 ◽

Vol 18 (2) ◽

pp. 343-358 ◽

Cited By ~ 60

Author(s):

Hassan Rivaz ◽

Zahra Karimaghaloo ◽

D. Louis Collins

Keyword(s):

Image Registration ◽

Mutual Information ◽

Self Similarity ◽

Nonrigid Image Registration

Download Full-text

Analysis and Research of Subclinical Diabetic Peripheral Vascular Lesions Based on Mutual Information-based Medical Image Registration Algorithm and Colour Ultrasound Imaging (Preprint)

10.2196/preprints.21025 ◽

2020 ◽

Author(s):

Nailong Jia ◽

Long Fan ◽

Chuanzi Li ◽

Zhongshi Nie ◽

Suihuang Wang ◽

...

Keyword(s):

Image Registration ◽

Mutual Information ◽

Blood Vessels ◽

Optimization Algorithm ◽

Peripheral Blood ◽

Medical Image ◽

Hybrid Optimization ◽

Medical Image Registration ◽

Hybrid Optimization Algorithm ◽

Subclinical Diabetes

BACKGROUND Background: At present, the incidence of diabetes is on the rise. When doctors diagnose and treat patients' diseases, they often need to image patients to provide complementary information on patient anatomy and functional metabolism. OBJECTIVE Objective: The aim was to understand the morphological features of peripheral blood vessels of diabetes more accurately and explore its Risk factors for the occurrence of lesions for early diagnosis and early prevention. METHODS Methods: The paper selected subclinical diabetes patients admitted to our hospital from October 2013 to October 2018 as a research object. After performing colour Doppler ultrasonography on peripheral blood vessels, images of ultrasound images were taken. Then the paper proposes a multi-mode medical image registration method based on hybrid optimization algorithm for the multi-extreme problem of mutual information function. Mutual information is used as the similarity measure. The hybrid optimization algorithm is used to search for the best registration exchange parameters. The quasi-colour super images are exchanged for registration purposes. RESULTS Results: The experimental results show that the hybrid optimization algorithm can accurately analyse the colour ultrasound image of the peripheral blood vessels of subclinical diabetes, avoiding falling into the local optimal value, and the accuracy of the registration result reaches the sub-pixel level. CONCLUSIONS Conclusion: With the rapid development of imaging technology, the increasing image resolution, and the increasing amount of image data, parallel performance is high. The quasi-method has a very important significance for multi-modal medical image registration. The parameters in this algorithm can be further optimized. CLINICALTRIAL

Download Full-text

A Coarse-to-Refined Approach of Medical Image Registration Based on Combining Mutual Information and Shape Information

2005 International Conference on Neural Networks and Brain ◽

10.1109/icnnb.2005.1614748 ◽

2006 ◽

Cited By ~ 1

Author(s):

Weiqing Chen ◽

Zongying Ou ◽

Weiwei Song

Keyword(s):

Image Registration ◽

Mutual Information ◽

Medical Image ◽

Medical Image Registration ◽

Shape Information

Download Full-text