### Table 4: Image-space temporal coherence versus object-space temporal coherence.

2005

"... In PAGE 15: ... The distinction between these two types of coherence is key to differentiate contributions. Table4 shows the different types of coherence used in the existing... ..."

### Table 1. Results of applying distortion and coherence metrics to datasets and refractive images from Figure 5.

2004

"... In PAGE 16: ... 6.3 Quantitative Analysis of Results The data in Table1 , which is obtained from the metrics discussed in Section 5, sug- gests that regularized anisotropic nonlinear diffusion performs better than all other methods. Direct comparisons are dif cult, since values for both md and mc do not exactly correspond.... ..."

Cited by 1

### Table 2 Comparison of cokriging image and HPF image using ERGAS (ER) statistic, Q statistic, spatial quality (SP) statistic and coherence (CH) statistic HPF image Cokriging image

2006

"... In PAGE 7: ... We used the correlation coefficient (CC), the mean error (ME), the mean squared error (MSE), (results are shown in Table 1) and the ERGAS statistic (Wald, 2000), the Q statistic (Wang amp; Bovik, 2002), the spatial quality (SQ) statistic (Zhou et al., 1998) and the coherence (CH) statistic (results are shown in Table2 ). The best case is to have a CC as close to one as possible, a ME as close to zero as possible, a MSE as small as possible, an ERGAS as small as possible, a Q index as close to one as possible, a SQ as close to one as possible and a CH as close to one as possible.... In PAGE 8: ... This is the coherence statistic in Table 2. Table2 shows the high coherence of the HPF image and the perfect coherence of the cokriging image. 4.... ..."

### Table 5.2: Comparing the mean squared error of reconstructions using different coherence methods for the matrix dataset with 5 input images.

2002

### Table 5.2: Comparing the mean squared error of reconstructions using di erent coherence methods for the matrix dataset with 5 input images.

### Table 3.2 Comparison of performance of 12-component and 24-component representation for strong coherent weighting (200 training, 1520 test images).

2007

### TABLE 1 Supported features of direct volume rendering methods for unstructured grids. ISC and OSC denote image-space and object-space coherencies, respectively. I-O and O-I denote image-then-object and object-then-image traversal orders, respectively.

2003

Cited by 3

### Table 3: Con dence value associated with the true class for every image in the database using edge direction coherence vector-based 5-NN classi er; 89% of the 2; 716 images were correctly classi ed with a con dence of 75% and above.

1998

Cited by 29

### Table 4.1: RMS bandwidths and coherence distances for a few standard images, as weil as for two illustrative test images constructed from 2D signals at a very low fkequency and at the Nyquist sampling frequency.

in Abstract

### Table 1: Results of applying distortion and coherence metrics to datasets and refractive images from Figure 5. Methods consid- ered include Gaussian convolution, inhomogeneous linear diffusion (IHLD), anisotropic nonlinear diffusion (ANLD), and regularized anisotropic nonlinear diffusion (R-ANLD).

"... In PAGE 8: ... 6.3 Quantitative Analysis of Results The data in Table1 , which is obtained from the metrics discussed in Section 5, suggests that regularized anisotropic nonlinear diffu- sion performs better than all other methods. Direct comparisons are difficult, since values for both md and mc do not exactly correspond.... ..."