. 2012 Mar 29;7(3):e33697. doi: 10.1371/journal.pone.0033697

Table 1. Statistical evaluation of filter performance using synthetic volumes contaminated with different levels of Gaussian and impulse noise shown in Figure 1.^a .

NoiseGaussian/Impulse	3D BLE (%)	3D recursive (%)	3D Canny (%)
0/0	29.66 (1.35)	30.03 (1.39)	26.91 (1.11)
20/20	46.24 (3.20)	118.46 (21.58)	38.52 (2.28)
40/40	52.52 (4.20)	120.74 (22.42)	42.05 (2.72)
60/60	63.36 (6.17)	123.00 (23.27)	43.73 (2.94)
80/80	67.86 (7.08)	124.96 (24.01)	45.62 (3.20)

RMSE scores between the input volumes and the three filter outputs are shown in bold. Smaller scores represent higher levels of correlation with the input volume. Values in brackets are the percentage voxel variation between input volumes and the three filtered outputs.

Table 1. Statistical evaluation of filter performance using synthetic volumes contaminated with different levels of Gaussian and impulse noise shown in Figure 1.a .

Table 1. Statistical evaluation of filter performance using synthetic volumes contaminated with different levels of Gaussian and impulse noise shown in Figure 1.^a .