Skip to main content
. Author manuscript; available in PMC: 2010 Feb 23.
Published in final edited form as: IEEE Trans Med Imaging. 2009 Feb 10;28(7):991–999. doi: 10.1109/TMI.2008.2008956

Fig. 6.

Fig. 6

Comparison 1: Results comparing ability of CSVM and CHO to generalize to test images of the same type, but different noise realizations, as the training images. In this situation, both the CSVM and CHO are very successful in predicting human-observer performance.