Mode of access: World Wide Web.

System requirements: Adobe Acrobat Reader.

Part of: Synthesis digital library of engineering and computer science.

Series from website.

Includes bibliographical references (p. 129-146).

1. Detection of oriented features in images -- Oriented features in images -- Directionally sensitive filters -- Steerable filters -- Gabor filters -- Line operator -- Frequency-domain analysis -- Performance analysis of the oriented feature detectors -- Performance in terms of detection and angular accuracy -- Analysis of performance with multiscale features -- Computational time -- Discussion -- Robustness to noise and scale -- Angular accuracy -- Computational speed -- Application of Gabor filters to the detection of curvilinear structures in mammograms -- Remarks --

2. Analysis of oriented patterns using phase portraits -- Phase portraits -- Analysis of orientation fields using phase portraits -- Local analysis of orientation fields -- Analysis of large orientation fields -- Illustrative examples -- Synthetic test images -- Analysis of a mammographic ROI -- Remarks --

3. Optimization techniques -- Introduction -- The optimization problem -- Optimization procedures -- Linear least-squares -- Iterative linear least-squares -- Nonlinear least-squares -- Simulated annealing -- Particle swarm optimization -- Comparative analysis of the optimization procedures -- Remarks --

4. Detection of sites of architectural distortion in mammograms -- The first method -- Filtering and downsampling the orientation field -- Post-processing of the node map and detection in the first method -- Results of the first method -- The second method -- Removal of the DC component in the Gabor filters -- Selection of curvilinear structures -- Filtering and downsampling the orientation field -- Estimating the phase portrait maps -- Detection of architectural distortion -- Results of the second method -- The third method -- Adoption of a symmetric matrix A -- Vote casting and detection -- Results of the third method -- Remarks --

A. Computer-aided diagnosis of breast cancer -- A.1. Screening for breast cancer -- A.2. Mammographic signs of breast cancer -- A.3. Enhancement of mammograms -- A.4. Segmentation of mammograms and analysis of breast density -- A.5. Detection and classification of microcalcifications -- A.6. Detection and classification of masses -- A.7. Analysis of curvilinear structures -- A.8. Analysis of bilateral asymmetry -- A.9. Detection of architectural distortion -- A.10. Analysis of prior mammograms -- A.11. Full-field digital mammography -- A.12. Indexed atlases, data mining, and content-based retrieval -- A.13. Computer-aided diagnosis of breast cancer -- A.14. Remarks --

B. Event detection in medical images -- B.1. Sensitivity and specificity -- B.2. Receiver operating characteristic analysis -- Bibliography -- Authors' biographies.

Abstract freely available; full-text restricted to subscribers or individual document purchasers.

Compendex

INSPEC

Google scholar

Google book search

The presence of oriented features in images often conveys important information about the scene or the objects contained; the analysis of oriented patterns is an important task in the general framework of image understanding. As in many other applications of computer vision, the general framework for the understanding of oriented features in images can be divided into low- and high-level analysis. In the context of the study of oriented features, low-level analysis includes the detection of oriented features in images; a measure of the local magnitude and orientation of oriented features over the entire region of analysis in the image is called the orientation field. High-level analysis relates to the discovery of patterns in the orientation field, usually by associating the structure perceived in the orientation field with a geometrical model. This book presents an analysis of several important methods for the detection of oriented features in images, and a discussion of the phase portrait method for high-level analysis of orientation fields. In order to illustrate the concepts developed throughout the book, an application is presented of the phase portrait method to computer-aided detection of architectural distortion in mammograms.

Also available in print.

Title from PDF t.p. (viewed on December 9, 2010).