This is an indepedent project that uses a sequence of undersampled CT images with subpixel movement to reconstruct a higher resolution image. This process removes the aliasing effects in the undersampled images. You can refer my report_dealiasing.pdf for more detail.
If you want to verify my result and run my code,
First you need to clone this git repository.
Then you have to download data here.
Put the data into the same working directory with the code.
In Matlab Console, Type in realdata_recon_main.
Remember to include both code and data in the working directory.
Hardie, R. C., K. J. Barnard, and E. E. Armstrong. “Joint MAP Registration and High-Resolution Image Estimation Using a Sequence of Undersampled Images.” IEEE Transactions on Image Processing 6.12 (1997): 1621–1633. Web. 13 May 2016.
- The CT image of stripes used in this study is in the courtesy of the Johns Hopkins Advanced Imaging Algorithm and Instrumentation Laboratory (AIAI), Baltimore, Maryland.
- im.m and readImg.m are the utility function contributed by the member of the Johns Hopkins I-star Lab, Baltimore, Maryland.
- Efficient_subpixel_registration package : Copyright (c) 2016, Manuel Guizar Sicairos, James R. Fienup, University of Rochester All rights reserved.