Scale and rotation invariant texture classification

Rotation-invariant texture classification using feature distributions

“…A method for classification of rotated and scaled textures using Gaussian Markov random field models was introduced by Cohen et al (6) . Approaches based on Gabor filtering have been proposed by, among others, Leung and Peterson (7) , Porat and Zeevi (8) , and Haley and Manjunath (9) . A steerable oriented pyramid was used to extract rotation invariant features by Greenspan et al (10) and a covariance-based representation to transform neighborhood about each pixel into a set of invariant descriptors was proposed by Madiraju and Liu (11) .…”

Brief review of invariant texture analysis methods

“…A set of Garbor ÿlters and mental transforms are combined to obtain invariant texture features by Leung et al [102] and Porter et al [63]. Invariant moment features [102] are also studied in comparison with the Gabor features that are found to have better performance on invariant texture classiÿcation.…”

“…Invariant moment features [102] are also studied in comparison with the Gabor features that are found to have better performance on invariant texture classiÿcation.…”

“…[102], 18 test sets consisting of 256 feature vectors are obtained from 6 different orientations (0, 30, 60, 90, 120, 150) and 3 different scaling factors (1, 0.67, 0.5) to get an overall error rate of 10%, when log=polar Gabor ÿlters and mental transforms are applied.…”

Normalized Texture Motifs and Their Application to Statistical Object Modeling

“…The first is to either extract rotationally invariant features [1][2] [3], or make them invariant through post-processing [4][5] [6]. Extracting isotropic features dismisses the issue of rotation invariance from the outset.…”