Skip to main content
Figure 3 | EURASIP Journal on Advances in Signal Processing

Figure 3

From: An augmented Lagrangian multi-scale dictionary learning algorithm

Figure 3

Comparison results of K-SVD, MOD, and ALM-DL with respect to the reconstruction of the original basis on synthetic signals. For each of the tested algorithms and for each noise level, 50 trials were conducted and their results were sorted. The graph labels represent the mean number of successfully learned basis elements (out of 50) over the ordered tests in groups of ten experiments.

Back to article page