Skip to main content
. Author manuscript; available in PMC: 2013 Mar 18.
Published in final edited form as: IEEE Trans Image Process. 2011 Jun 20;21(1):130–144. doi: 10.1109/TIP.2011.2160072

Fig. 1.

Fig. 1

(From left to right) The original horses image, the noisy horses image with the noise standard deviation of 25, the denoised image, and the inferred dictionary (from top left) with its elements ordered in the probability to be used. The low-probability dictionary elements are never used to represent {xi}i=1,N and are drawn from the prior, showing the ability of the model to learn the number of dictionary elements needed for the data.