Near Set Evaluation And Recognition (NEAR) System - Rough Fuzzy Image Analysis: Foundations and Methodologies - page 128

7

Near Set Evaluation And Recognition
(NEAR) System

7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1

7.2 Near sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2

Perceptual Tolerance relation • Nearness Measure

7.3 Perceptual Image Processing 7-7

. . . . . . . . . . . . . . . . . . . . . . . .

7.4 Probe functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-8

Average greyscale value • Normalized RGB • Shannon's
entropy • Pal's entropy • Edge based probe functions

7.5 Equivalence class frame 7-12

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7.6 Tolerance class frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-14

Christopher Henry
Computational Intelligence Laboratory,
Electrical & Computer Engineering, Rm.
E2-390 EITC Bldg., 75A Chancellor's Circle,
University of Manitoba, Winnipeg R3T 5V6
Manitoba Canada

7.7 Segmentation evaluation frame . . . . . . . . . . . . . . . . . . . . . . 7-15

7.8 Near image frame 7-17

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7.9 Feature display frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-18

7.10 Conclusion 7-18

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Bibliography 7-19

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7.1 Introduction

Near sets introduced in (Peters, 2007b,c; Henry and Peters, 2009d), elaborated in (Pe-
ters and Wasilewski, 2009; Peters, 2009c, 2010; Peters and Wasilewski, 2010) and their
applications (Peters, 2009b,c; Peters and Ramanna, 2009; Henry and Peters, 2007, 2008;
Hassanien, Abraham, Peters, Schaefer, and Henry, 2009; Peters, Shahfar, Ramanna, and
Szturm, 2007a; Peters and Ramanna, 2007; Gupta and Patnaik, 2008; Henry and Peters,
2009c; Fashandi, Peters, and Ramanna, 2009; Meghdadi, Peters, and Ramanna, 2009; Pe-
ters and Puzio, 2009) grew out of a generalization of the approach to the classification of
objects proposed by Z. Pawlak during the early 1980s (see, e.g., (Pawlak, 1981, 1982), elab-
orated in (Pawlak and Skowron, 2007a,b,c)), E. Orlowska's suggestion that approximation
spaces are the formal counterpart of perception or observation (Orlowska, 1982), and a
study of the nearness of objects (Peters, Skowron, and Stepaniuk, 2006, 2007b). This chap-
ter introduces the NEAR system (available for downloading for free at (Peters, 2009a)),
an application implemented to demonstrate and visualize concepts from near set theory
reported in (Henry and Peters, 2007; Peters, 2007a,c; Henry and Peters, 2008; Peters, 2008;
Peters and Ramanna, 2009; Peters, 2009b; Henry and Peters, 2009c; Peters and Wasilewski,
2009; Peters, 2009c; Hassanien et al., 2009; Henry and Peters, 2009b).
The NEAR system implements a Multiple Document Interface (MDI) (see, e.g., Fig. 7.1)
where each separate processing task is performed in its own child frame. The objects (in the

7-1