Los Angeles, CA
March 31, 2009 to April 2, 2009
In recent years, processed medical image becomes more and more important to diagnosis. Anamorphic image may result in wrong judgment on state of an illness, even leads to fateful danger. Meanwhile, more efficient data transportation and storage are required with the development of high-resolution photography. Therefore, thousands of researchers work on the field of medical image processing. In this paper, contourlet transform, which may provide with tight bracing and mutli-scale analysis, is introduced. And a new medical image processing algorithm based on contourlet transform and correlation theory is presented. This algorithm possesses some excellent performances such as multi-scale analysis, time-frequency-localization and multi-directions. Especially, this algorithm has excellent performance when describing anisotropic 2-D data. So high-quality medical image can be reconstructed even though a relative few coefficients are employed.To verify the algorithm, some medical images were processed. The experimental results show that this algorithm has better performance in compression and denoising than that of wavelet transform.
Contourlet Transform, Correlation Theory, Medical Image, Image Processing
Jun Wang, Yan Kang, "Study on Medical Image Processing Algorithm Based on Contourlet Transform and Correlation Theory", CSIE, 2009, 2009 WRI World Congress on Computer Science and Information Engineering, CSIE, 2009 WRI World Congress on Computer Science and Information Engineering, CSIE 2009, pp. 233-238, doi:10.1109/CSIE.2009.1010