Image Processing Reference
In-Depth Information
(9.6a)
original
image
(9.6b)
original
image
(9.6c) original image
(Water)
(Leaf1)
(Leaf2)
(9.6d) covering,p=60
(9.6e) covering, p=60
(9.6f) covering, p=60
(9.6g) original
image (Leaves)
(9.6h) covering,
p=60
FIGURE 9.6: GUI example
TABLE 9.4 Image comparison results using TCD measure
Images 1-Barbara2-Lena3-Leaves4-Walnut5-Leaf16-leaf27-lake8-water
1 1.00 0.92 0.71 0.89 0.86 0.81 0.87 0.97
2 0.92 1.00 0.78 0.96 0.94 0.89 0.94 0.90
3 0.71 0.78 1.00 0.82 0.82 0.83 0.83 0.69
4 0.89 0.96 0.82 1.00 0.95 0.89 0.98 0.87
5 0.86 0.94 0.82 0.95 1.00 0.94 0.95 0.84
6 0.81 0.89 0.83 0.89 0.94 1.00 0.89 0.80
7 0.87 0.94 0.83 0.98 0.95 0.89 1.00 0.85
8 0.97 0.90 0.69 0.87 0.84 0.80 0.85 1.00
distribution nearness measure (TCD) is based on the difference between the statistical dis-
tribution of the size (cardinality) of the tolerance classes in each image. The size of each
tolerance class is defined as the number of perceptual objects (subimages) in that tolerance
class. TCD is fundamentally different from tNM proposed in (Henry, 2009) and TOD pro-
posed in (Meghdadi et al., 2009). While TCD calculates and compares distribution of the
size of tolerance classes, TOD only measures the overlap between tolerance classes and the
size of tolerance classes have no direct impact on TOD. By contrast, tNM compares the
sizes of tolerance classes in the coverings of pairs of images.
The results of comparing different images with these three nearness measures are shown
in this chapter. The results show that all three measures work well. However, there are
Search WWH ::




Custom Search