Moments and Moment Invariants in Pattern Recognition / Edition 1 available in Hardcover
- Pub. Date:
Moments as projections of an image’s intensity onto a properpolynomial basis can be applied to many different aspects of imageprocessing. These include invariant pattern recognition, imagenormalization, image registration, focus/ defocus measurement, andwatermarking. This book presents a survey of both recent andtraditional image analysis and pattern recognition methods, basedon image moments, and offers new concepts of invariants to linearfiltering and implicit invariants. In addition to the theory,attention is paid to efficient algorithms for moment computation ina discrete domain, and to computational aspects of orthogonalmoments. The authors also illustrate the theory through practicalexamples, demonstrating moment invariants in real applicationsacross computer vision, remote sensing and medical imaging.
- Presents a systematic review of the basic definitions andproperties of moments covering geometric moments and complexmoments.
- Considers invariants to traditional transforms –translation, rotation, scaling, and affine transform - from a newpoint of view, which offers new possibilities of designing optimalsets of invariants.
- Reviews and extends a recent field of invariants with respectto convolution/blurring.
- Introduces implicit moment invariants as a tool for recognizingelastically deformed objects.
- Compares various classes of orthogonal moments (Legendre,Zernike, Fourier-Mellin, Chebyshev, among others) and demonstratestheir application to image reconstruction from moments.
- Offers comprehensive advice on the construction of variousinvariants illustrated with practical examples.
- Includes an accompanying website providing efficient numericalalgorithms for moment computation and for constructing invariantsof various kinds, with about 250 slides suitable for a graduateuniversity course.
Moments and Moment Invariants in Pattern Recognition isideal for researchers and engineers involved in pattern recognitionin medical imaging, remote sensing, robotics and computer vision.Post graduate students in image processing and pattern recognitionwill also find the book of interest.
|Product dimensions:||6.80(w) x 9.90(h) x 0.90(d)|
About the Author
Professor Jan Flusser, PhD, Dsc, is a director of theInstitute of Information Theory and Automation of the ASCR, Prague,Czech Republic, and a full professor of Computer Science at theCzech Technical University, Prague, and at the Charles University ,Prague. Jan Flusser’s research areas are moments and momentinvariants, image regristration, image fusion, multichannel blinddeconvolution and super-resolution imaging. He has authoredand coauthored more than 150 research publications in these areas,including tutorials (ICIP’05, ICIP’07,EUSIPCO’07, CVPR’08, FUSION’08, SPPRA’09,SCIA’09) and invited/keynote talks (ICCS’06,COMPSTAT’06, WIO’06, DICTA’07, CGIM’08) atmajor international conferences. He gives undergraduate andgraduate courses on digital image processing, pattern recognition,and moment invariants and wavelets. Personal webpagehttp://www.utia.cas.cz/people/flusser.
Tomáš Suk, PhD, is a research fellow of thesame Institute. His research interests include invariant features,moment and point-based invariants, color spaces and geometrictransformations. He has authored and coauthored more than 50research publications in these areas, some of which have elicited aconsiderable citation response. Tomás Suk coauthored tutorialson moment invariants held at international conference ICIP’07and SPPR’09. Personal webpage http://zoi.utia.cas.cz/suk.
Barbara Zitová, PhD, is Head of the Department ofImage Processing at the same Institute. Her research interest ismainly in image regi8stration, invariants, wavelets, and imageprocessing applications in cultural heritage. She has authored andcoauthored more that 30 research publications in these areas,including tutorials at international conferences (ICIP’05,ICIP’07, EUSIPCO’07, FUSION’08 andCVPR’08). Her paper “Image Registration Methods: ASurvey,” Image and Vision Computing, vol. 21, pp.977-1000, 2003, has become a major reference work in imageregistration . She teaches a specialized graduate course on momentinvariants and wavelets at the Czech Technical University. Personalwebpage http://zoi.utia.cas.cz/zitova.
Table of Contents
1 Introduction to moments.
1.2 What are invariants?
1.3 What are moments?
1.4 Outline of the book.
2 Moment invariants to translation, rotation andscaling.
2.2 Rotation invariants from complex moments.
2.4 Combined invariants to TRS and contrast changes.
2.5 Rotation invariants for recognition of symmetricobjects.
2.6 Rotation invariants via image normalization.
2.7 Invariants to nonuniform scaling.
2.8 TRS invariants in3D.
3 Affine moment invariants.
3.2 AMIs derived from the Fundamental theorem.
3.3 AMIs generated by graphs.
3.4 AMIs via image normalization.
3.5 Derivation of the AMIs from the Cayley–Aronholdequation.
3.6 Numerical experiments.
3.7 Affine invariants of color images.
3.8 Generalization to three dimensions.
4 Implicit invariants to elastic transformations.
4.2 General moments under a polynomial transform.
4.3 Explicit and implicit invariants.
4.4 Implicit invariants as a minimization task.
4.5 Numerical experiments.
5 Invariants to convolution.
5.2 Blur invariants for centrosymmetric PSFs.
5.3 Blur invariants for N-fold symmetric PSFs.
5.4 Combined invariants.
6 Orthogonal moments.
6.2 Moments orthogonal on a rectangle.
6.3 Moments orthogonal on a disk.
6.4 Object recognition by ZMs.
6.5 Image reconstruction from moments.
6.6 Three-dimensional OG moments.
7 Algorithms for moment computation.
7.2 Moments in a discrete domain.
7.3 Geometric moments of binary images.
7.4 Geometric moments of graylevel images.
7.5 Efficient methods for calculating OG moments.
7.6 Generalization to n dimensions.
8.2 Object representation and recognition.
8.3 Image registration.
8.4 Robot navigation.
8.5 Image retrieval.
8.7 Medical imaging.
8.8 Forensic applications.
8.9 Miscellaneous applications.
What People are Saying About This
"This text is a little gem in the vast amount of literature on pattern recognition...In conclusion, this is an excellent text on pattern recognition that I highly recommend to practitioners and students in signal and image processing." (Computing Reviews, October 2010)