: Claudia I. Gonzalez, Patricia Melin, Juan R. Castro, Oscar Castillo
: Edge Detection Methods Based on Generalized Type-2 Fuzzy Logic
: Springer-Verlag
: 9783319539942
: 1
: CHF 47.50
:
: Allgemeines, Lexika
: English
: 94
: Wasserzeichen/DRM
: PC/MAC/eReader/Tablet
: PDF
In this book four new methods are proposed. In the first method the generalized type-2 fuzzy logic is combined with the morphological gra-dient technique. The second method combines the general type-2 fuzzy systems (GT2 FSs) and the Sobel operator; in the third approach the me-thodology based on Sobel operator and GT2 FSs is improved to be applied on color images. In the fourth approach, we proposed a novel edge detec-tion method where, a digital image is converted a generalized type-2 fuzzy image. In this book it is also included a comparative study of type-1, inter-val type-2 and generalized type-2 fuzzy systems as tools to enhance edge detection in digital images when used in conjunction with the morphologi-cal gradient and the Sobel operator. The proposed generalized type-2 fuzzy edge detection methods were tested with benchmark images and synthetic images, in a grayscale and color format.

< iv>Another contribution in this book is that the generalized type-2 fuzzy edge detector method is applied in the preprocessing phase of a face rec-ognition system; where the recognition system is based on a monolithic neural network. The aim of this part of the book is to show the advantage of using a generalized type-2 fuzzy edge detector in pattern recognition applications.

< div>The main goal of using generalized type-2 fuzzy logic in edge detec-tion applications is to provide them with the ability to handle uncertainty in processing real world images; otherwise, to demonstrate that a GT2 FS has a better performance than the edge detection methods based on type-1 and type-2 fuzzy logic systems.
Preface6
Contents9
1 Introduction11
2 Generalized Type-2 Fuzzy Logic13
2.1 Definition of Generalized Type-2 Fuzzy Sets13
2.2 ?-Planes Representation14
2.3 Generalized Type-2 Fuzzy Systems Based on ?-Planes15
2.3.1 Fuzzifier Process15
2.3.2 Fuzzy Rules16
2.3.3 Inference Engine17
2.3.4 Type Reducer18
2.3.5 Defuzzification Process19
3 Edge Detection Methods and Filters Used on Digital Image Processing20
3.1 Edge Detection Methods20
3.1.1 Morphological Gradient Approach20
3.1.2 Sobel Operator22
3.1.3 Sobel Operator Applied on Color Images23
3.2 Filters24
3.2.1 Low-Pass Filters24
3.2.2 High-Pass Filters25
4 Metrics for Edge Detection Methods26
4.1 Figure of Merit of Pratt (FOM)26
4.2 Quality Measurement Using the MSE, PSNR and SSIM Indices27
5 Edge Detection Methods Based on Generalized Type-2 Fuzzy Logic Systems29
5.1 Edge Detection Method Based on GT2 FSs and the Morphological Gradient29
5.2 Edge Detection Method Based on GT2 FSs and the Sobel Operator32
5.3 Generalized Type-2 Fuzzy Edge Detection Method Applied on Color Images33
5.4 Edge Detection Method Using GT2 Fuzzy Images36
5.4.1 Fuzzy Synthetic Images36
5.4.1.1 Fuzzy Images Using Type-1 Membership Functions38
5.4.1.2 Fuzzy Images Using Interval Type-2 Membership Functions39
5.4.1.3 Fuzzy Images Using Generalized Type-2 Membership Functions39
5.4.2 The Fuzzy Euclidean Distance40
5.4.3 Edge Detection Method Applied on Fuzzy Images42
6 Generalized Type-2 Fuzzy Edge Detection Applied on a Face Recognition System44
6.1 Generalized Type-2 Fuzzy Edge Detection Method Using the Sobel Operator and Filters44
6.2 Face Recognition System Based on a Monolithic Neural Network46
7 Experimentation and Results Discussion49
7.1 Generalized Type-2 Fuzzy Systems Combined with the Morphological Gradient51
7.2 Generalized Type-2 Fuzzy Systems Combined with the Sobel Operator55
7.3 Generalized Type-2 Fuzzy Edge Detection Method Using Color Images62
7.3.1 Simulation Results Using Synthetic Color Images63
7.3.2 Simulation Results Using Real Color Images68
7.4 Edge Detection Method Applied on GT2 Fuzzy Images75
7.5 Simulation Results Achieved by the Face Recognition System77
8 Conclusions82
Appendix A: A.1 Function to Calculate Morphological Gradient Edge Detection84
A.2. Function to Define the FIS for the MG + GT2 FSs85
A.3. Function to Obtain the Sobel Operator87
A.4. Function to Define the FIS for the Sobel + GT2 FSs87
A.5. Function to Obtain Edge Detection Method on GT2 Fuzzy Images89
References90
Index94