Development and Application of Image Analysis Techniques for Identification and Classification of Microscopic Particles

Ph.D. Thesis

Volodymyr Kindratenko

[pdf version] [fractal analysis software]


Table of  Contents

preface

part 1: IMAGE PROCESSING TECHNIQUES

1.1. Basics of image formation

1.1.1. Image formation in SEM

1.1.2. Image formation in TEM

1.2. Image processing software

1.2.1. KS400

1.2.2. Iage processing/analysis software developed

1.2.3. Image processing with MathCAD and MatLab

1.3. Image storage and manipulation

1.3.1. Windows bitmap file format

1.3.2. Sun raster file format

1.4. Image enhancement

1.4.1. Gray level histogram modifications

1.4.2. Smoothing of noisy images

1.4.3. Sharpening

1.5. Image segmentation

1.5.1. Global thresholding using a correlation criterion

1.5.2. Local binarization using discrete convolution

1.5.3. Segmentation based on watershed transform

1.6. Processing of binary images

1.6.1. Image enhancement

Binary morphology

Shrink and swell filters

1.6.2. Contour following techniques

'Turtle' procedure

Crack following

Border following

1.6.3. Perimeter estimation by different yardsticks

1.6.4. Contour filling and object labeling

1.6.5. Watershed segmentation of touching objects

References

part 2: SHAPE ANALYSIS

2.1. Functional approach

2.1.1. Contour functions

Cross-section functions for symmetric figures

Radius-vector functions

Support functions

Width function

Contour parametric and contour complex functions

Tangent-angle function

The intrinsic equation of the contour

Concluding remarks on contour functions

2.1.2. Application of contour functions to shape analysis

Invariant contour function parameters

Line moments and invariants

Approximation of contour functions by other simple functions

Fourier analysis of contour functions

Some other possible series expansions of contour functions

Multiscale shape analysis using continuous wavelet transform

Shape curvature scale space representation

2.2. Set theory approach

2.2.1. Simple geometrical shape parameters

2.2.2. Fractals in shape analysis

Definition of fractal dimension

References

part 3: APPLICATIONS

3.1. Classification of individual fly ash and soil dust aerosol particles

3.1.1. Introduction

3.1.2. Types of shapes of aerosol particles

3.1.3. Fractal description of particle shapes: a brief overview

3.1.4. Experimental

3.1.5. Results and discussion

3.1.6. Conclusions

3.2. Differentiation between individual algae cells and their agglomerates

3.2.1. Introduction

3.2.2. Complex Fourier shape description

3.2.3. Classification algorithms

3.2.4. Experimental

3.2.5. Results and discussion

3.2.6. Conclusions

3.3. Classification of tabular grain silver halide microcrystals according to their shape

3.3.1. Introduction

3.3.2. Shape representation of the microcrystals

3.3.3. Reconstruction of the shape of overlapping microcrystals

Extraction of hexagonal and truncated triangular microcrystals

Extraction of triangular microcrystals

3.3.4. Classification of microcrystals via their shape descriptors

Nearest neighbor classification algorithms

Labeled samples and prototypes

3.3.5. Experimental

3.3.6. Results and discussion

3.3.7. Conclusions

3.4. On fractal dimension calculation

3.4.1. 'Hand and dividers' method: theory

3.4.2. 'Hand and dividers' method: practice

3.4.3. Problems associated with the 'hand and dividers' method

3.4.4. Analysis of the Richardson plot

3.5. Study of quasi-fractal many-particle systems and percolation networks

3.5.1. Introduction

3.5.2. Experimental

Samples and sample preparation and image acquisition

Image processing and analysis

3.5.3. Results and discussions

Ag colloids

Ag filament networks

3.5.4. Conclusions

References

SUMMARY

SAMENVATTING

APPENDIX

A.1. Publications in refereed scientific journals and conference proceedings

A.2. Conference contributions

A.3. Research reports


Document is created by Volodymyr Kindratenko
Last modified: 08/22/99