Novel approach for rotation invariant texture recognition

Authors:
Naeem Qaiser;Mutawarra Hussain;Nadeem Qaiser;Muhammad Iqbal
Affiliations:
DCIS, PIEAS, Pakistan;DCIS, PIEAS, Pakistan;DEE, PIEAS, Pakistan;DCE, CASE, Pakistan
Venue:
CI '07 Proceedings of the Third IASTED International Conference on Computational Intelligence
Year:
2007

Citing 10
Cited 0

A model-based method for rotation invariant texture classification

IEEE Transactions on Pattern Analysis and Machine Intelligence
Digital image processing (2nd ed.)

Digital image processing (2nd ed.)
Texture classification and segmentation using multiresolution simultaneous autoregressive models

Pattern Recognition
Pattern recognition with moment invariants: a comparative study and new results

Pattern Recognition
Texture analysis

Handbook of pattern recognition & computer vision
Rotation-invariant texture classification

Pattern Recognition Letters
Rotation invariant texture classification using even symmetric Gabor filters

Pattern Recognition Letters
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)
Nonlinear operator for oriented texture

IEEE Transactions on Image Processing
Comparison of texture features based on Gabor filters

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In machine vision, rotation invariant feature extraction is one of the most challenging texture analysis tasks, because pattern orientation itself contributes substantially to extracted features. As a consequence, the prime objective of such techniques has always been to extract features that maintain reasonable discrimination while achieving invariance. This paper addresses the issue by proposing a novel moment invariant based feature set for efficient rotation invariant texture segmentation. In deriving proposed feature set, a moment mask based technique has been employed innovatively and in the process only seven moment images are computed. A Fisher discriminant analysis based criterion has been devised to evaluate the rotation invariance/discrimination capabilities of proposed feature set in a systematic and quantitative way. The effectiveness of the solution has been verified through segmentation as well as supervised classification of benchmark textures taken from Brodatz album. The results show significant improvement when compared with an existing technique.