Skip to main content
. 2019 May 13;5:e192. doi: 10.7717/peerj-cs.192

Table 5. Average classification accuracy on the Extended Yale B Database over K for a fixed rate of 0.4 missing pixels.

Best result for each classifier is marked bold. Feature dimension = 128.

K OMP GOMP CMP Cauchy Fair Huber Tukey Welsch
1 0.51 0.54 0.57 0.61 0.62 0.56 0.54 0.54
2 0.54 0.52 0.89 0.87 0.86 0.88 0.88 0.88
3 0.57 0.48 0.95 0.91 0.90 0.93 0.94 0.94
4 0.56 0.45 0.95 0.92 0.90 0.94 0.96 0.95
5 0.55 0.45 0.95 0.92 0.89 0.94 0.96 0.96
6 0.54 0.45 0.95 0.91 0.89 0.94 0.97 0.96
7 0.53 0.45 0.94 0.91 0.89 0.94 0.97 0.96
8 0.52 0.45 0.94 0.91 0.89 0.94 0.96 0.96
9 0.51 0.45 0.94 0.91 0.89 0.94 0.96 0.95
10 0.50 0.45 0.94 0.91 0.89 0.93 0.96 0.95