Near Set Evaluation And Recognition (NEAR) System - Rough Fuzzy Image Analysis: Foundations and Methodologies - page 141

FIGURE 7.13: Sample run of the equivalence class frame using a window size of 5×5 and
B={φ ,φ

H Shannon }. NormG

of the mouse click determine the equivalence class that is displayed. The results may be
saved by clicking on the save button.

7.6 Tolerance class frame

This frame calculates tolerance classes using the Tolerance relation of Defn. 3, i.e., given
an image X, it will calculate X / ∼

= B where the objects are subimages of X. This approach
is similar to the one given in Section 7.3 with the exception that Defn. 1 is replaced with
Defn. 3. A sample calculation using this frame is given in Fig. 7.14 and was obtained by
the following steps:

1. Click Load Image button.
2. Select number of features (maximum allowed is four).
3. Select features (see Section 7.4 for a list of probe functions).
4. Select window size. The value is taken as the square root of the area for a square
subimage, e.g., a value of 5 creates a subimage of 25 pixels.
5. Select , a value in the interval [0, 1].
6. Click Run.

The result is given in Fig. 7.14 where the left side is the original image, and the right side
is used to display the tolerance classes. Since the tolerance relation does not partition an
image, the tolerance classes are displayed upon request. For instance, by clicking on either
of the two images, all the tolerance classes are displayed that are within of the subimage
containing the coordinates of the mouse click. Further, the subimage containing the mouse
click is coloured black.