"Analysis and Optimization of Weighted Order Statistics and Stack Filters"
S. Siren, P. Kuosmainen, K. Egiazarian, M. Gabbouj and J. Astola 1.1 Introduction 1.2 Median and Order Statistic Filters 1.3 Stack Filters 1.4 Image Processing Applications 1.5 Summary
"Image Enhancement and Analysis with Weighted Medians"G. Arce and J. L. Paredes
2.1 Introduction 2.2 Weighted Median Smoothers and Filters 2.3 Image Denoising 2.4 Image Zooming 2.5 Image Sharpening 2.6 Optimal Frequency Selection WM Filtering 2.7 Edge Detection 2.8 Conclusion
"Spatial-Rank Order Selection Filters"R. Hardie and K. Barner
3.1 Introduction 3.2 Selection Filters and Spatial-Rank Ordering 3.3 Spatial-Rank Order Selection Filters 3.4 Optimization 3.5 Applications 3.6 Future Directions
"Signal-Dependent Rank-Ordered Mean (SD-ROM) Filter" E. Abreu 4.1 Introduction 4.2 Impulse Noise Model 4.3 Definitions 4.4 The SD-ROM Filter 4.5 Generalized SD-ROM Method 4.6 Experimental Results 4.7 Restoration of Images Corrupted by Streaks 4.8 Conclusing Remarks
"Nonlinear Mean Filters and Their Applications in Image Filtering and Edge Detection" M. Pappas and I. Pitas 5.1 Introduction 5.2 Nonlinear Mean Filters 5.3 Signal-Dependent Filtering by Nonlinear Means 5.4 Edge Detectors Based on Nonlinear Means 5.5 Grayscale Morphology
This state-of-the-art book deals with the most important aspects of non-linear imaging challenges. The need for engineering and mathematical methods is essential for defining non-linear effects involved in such areas as computer vision, optical imaging, computer pattern recognition, and industrial automation challenges.
@bul:* Presents the latest developments in a variety of filter design techniques and algorithms
- Contains essential information for development of Human Vision Systems (HVS)
- Provides foundations for digital imaging and image capture technology
Electrical and computer engineers.
- No. of pages:
- © Academic Press 2001
- 5th September 2000
- Academic Press
- eBook ISBN:
- Hardcover ISBN:
@qu:"The book considers the following filter families, with varying emphasis, according to popularity and impact in image processing tasks:
- honomorphic filters, relying on a generalized superposition principle
- nonlinear mean filters, using nonlinear definitions of means
- morphological filters, based on geometrical rather than analytical properties
- order statistics filters, based on ordering properties of the input samples
- polynomial filters, using polynomial expressions in the input and output samples
- fuzzy filters, applying fuzzy reasoning to model the uncertainty typical of some image processing issues
- nonlinear operations modeled in terms of nonlinear partial differential equations."
University of California, Santa Barbara, U.S.A.
University of Trieste, Italy