Rotation Invariant Texture Classification Using Gabor Wavelets

Authors:
Qingbo Yin;Jong Nam Kim;Kwang-Seok Moon
Affiliations:
Division of Electronic Computer and Telecommunication Engineering, Pukyong National University, 599-1 Daeyeon-dong Nam-gu, Busan, 608-737, Korea and College of Computer Science and Technology, Har ...;Division of Electronic Computer and Telecommunication Engineering, Pukyong National University, 599-1 Daeyeon-dong Nam-gu, Busan, 608-737, Korea;Division of Electronic Computer and Telecommunication Engineering, Pukyong National University, 599-1 Daeyeon-dong Nam-gu, Busan, 608-737, Korea
Venue:
ICCS '07 Proceedings of the 7th international conference on Computational Science, Part III: ICCS 2007
Year:
2007

Citing 5
Cited 0

Texture Features for Browsing and Retrieval of Image Data

IEEE Transactions on Pattern Analysis and Machine Intelligence
Rotation Invariant Texture Features and Their Use in Automatic Script Identification

IEEE Transactions on Pattern Analysis and Machine Intelligence
Rotation invariant texture classification using even symmetric Gabor filters

Pattern Recognition Letters
Texture classification using Gabor wavelets based rotation invariant features

Pattern Recognition Letters
Rotation-invariant texture classification using a complete space-frequency model

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a new method for rotation invariant texture classification based on Gabor wavelets. The Gabor representation has been shown to be optimal in the sense of minimizing the joint two-dimensional uncertainty in space and frequency, and the Gabor wavelet can be used to decompose an image into multiple scales and multiple orientations. Two group features, i.e. the global feature vector and local feature matrix, can be constructed by the mean and variance of the Gabor filtered image. The global feature vector is rotation invariant, and the local feature matrix can be adjusted by a circular shift operation to rotation invariant so that all images have the same dominant direction. By the two group features, a discriminant can be found to classify the rotated images. In the primary experiments, comparatively high correct classification rates were obtained using a large sample sets with 1998 rotated images of 111 Brodazt texture classes.