Multichannel Texture Analysis Using Localized Spatial Filters
IEEE Transactions on Pattern Analysis and Machine Intelligence
Image Representation Using 2D Gabor Wavelets
IEEE Transactions on Pattern Analysis and Machine Intelligence
On steerability of Gabor-type filters for feature detection
Pattern Recognition Letters
Simplified Gabor wavelets for human face recognition
Pattern Recognition
Using skew Gabor filter in source signal separation and local spectral orientation analysis
Image and Vision Computing
IEEE Transactions on Signal Processing
On-road vehicle detection using evolutionary Gabor filter optimization
IEEE Transactions on Intelligent Transportation Systems
Analysis of Brute-Force Break-Ins of a Palmprint Authentication System
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
A fast learning algorithm for Gabor transformation
IEEE Transactions on Image Processing
Quadratic Gabor filters for object detection
IEEE Transactions on Image Processing
| Hi-index | 0.01 |
An elementary function that is now commonly referred to Gabor function, Gabor filter and Gabor wavelet was derived from uncertainty relation for information by Gabor to overcome the representation limit of Fourier analysis. Analyzing a signal by a Gabor filter in terms of convolution or spatial filtering, two pieces of information--phase and magnitude--can be obtained. In the paper, Gabor filter is considered as a Gabor atom detector. This analysis demonstrates that when the k -value defined as k = *** g ni *** 2 / *** g nr *** 2, where g nr and g ni are respectively the real and imaginary parts of a Gabor filter g n , is close to one, the target phase can be estimated by Gabor phase and the target magnitude can be estimated by Gabor magnitude. However, when the k -value decreases, the quality of this approximation also decreases. The corresponding error bounds are derived.