Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Algorithm of invariant image description using modified Gabor transform

Not Accessible

Your library or personal account may give you access

Abstract

We propose an algorithm of image coding which is invariant to scale transformation due to a method of scale weighting function (SWF) described below. So it could be a base for an algorithm of pattern recognition. We get an image code at the output—a set of 60 numbers which are independent on the image size. Let's describe the algorithm briefly. An input image I(x,y) is processed in some channels which have identical organization and different scales forming logarithmic series with the step 1/2 octave. Processing of an image in K-th channel includes two steps: (a, b). Then using SWF method the image code is built in two steps: (c, d).

© 1994 IEEE

PDF Article
More Like This
Invariant Pattern Recognition Using Quantum-Limited Images

Thomas A. Isberg and G. Michael Morris
ThC3 Machine Vision (MV) 1987

Recovering the Affine Transformation of Images by Using Moments and Invariant Axes

Morris Lee
MA1 Image Understanding and Machine Vision (IUMV) 1989

Fast recognition of Young's fringe images via the indirect radon transform

Vladimir A. Shapiro and Vassil I. Kavardjikov
CThI49 The European Conference on Lasers and Electro-Optics (CLEO/Europe) 1994

Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.