A Model-Based Method for Rotation Invariant Texture Classification

Kashyap, R.L.; Khotanzad, A.

doi:10.1109/tpami.1986.4767811

Cited by 304 publications

(142 citation statements)

References 21 publications

Supporting

Mentioning

138

Contrasting

Unclassified

Order By: Relevance

“…A mathematical description of these measures computed for center-symmetric pairs of pixels in a 3x3 neighborhood (see Fig. 1) is presented in equations (1) - (4). µ denotes the local mean and σ 2 the local variance in the equations.…”

Section: Measures Based On Center-symmetric Auto-correlationmentioning

confidence: 99%

“…Then, the rotated images used for testing the texture classifier were generated by rotating each image counter-clockwise around its center with the same bicubic interpolation method. We used the same set of six different rotation angles that was used by Kashyap and Khotanzad (4) : 30, 60, 90, 120, 150, and 200 degrees. In other words this is a true test of rotation-invariant texture classification, for the classifier 'sees' only instances of reference textures, and it is tested with instances of rotated textures it has not 'seen' before.…”

Section: Texture Imagesmentioning

confidence: 99%

“…In other words, almost identical results are obtained at 30 and 120 degrees, just like is the case with 60 and 150 degrees. This suggests that the set of six rotation angles we adopted from Kashyap and Khotanzad (4) is suboptimal, at least when 3x3 operators are used. A more comprehensive test could be obtained by choosing a set of rotation angles that does not contain any two angles differing a multiple of 45 degrees.…”

Section: Single Feature Performancementioning

confidence: 99%

“…Other early approaches proposed for rotationinvariant texture classification include the methods based on polarograms (1) , generalized co-occurrence matrices (2) and texture anisotropy (3) . Kashyap and Khotanzad developed a method based on the circular symmetric autoregressive random field (CSAR) model for rotation-invariant texture classification (4) . Mao and Jain proposed a multivariate rotation-invariant simultaneous autoregressive model (RISAR) that is based on the CSAR model, and extended it to a multiresolution SAR model MR-RISAR) (5) .…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Rotation-invariant texture classification using feature distributions

Pietikäinen

Ojala

2000

Pattern Recognition

399

170

View full text Add to dashboard Cite

Section: Measures Based On Center-symmetric Auto-correlationmentioning

confidence: 99%

Section: Texture Imagesmentioning

confidence: 99%

Section: Single Feature Performancementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Rotation-invariant texture classification using feature distributions

Pietikäinen

Ojala

2000

Pattern Recognition

399

170

View full text Add to dashboard Cite

“…However, real world textures can occur at arbitrary spatial resolutions and rotations and they may be subjected to varying illumination conditions. This has inspired a collection of studies, which generally incorporate invariance with respect to one or at most two of the properties spatial scale, orientation and gray scale, among others [1,2,3,4,5,6,7,8,10,11,13,14].…”

Section: Introductionmentioning

confidence: 99%

A Generalized Local Binary Pattern Operator for Multiresolution Gray Scale and Rotation Invariant Texture Classification

Ojala

Pietikäinen

Mäenpää

2001

Lecture Notes in Computer Science

233

113

View full text Add to dashboard Cite

Abstract. This paper presents generalizations to the gray scale and rotation invariant texture classification method based on local binary patterns that we have recently introduced. We derive a generalized presentation that allows for realizing a gray scale and rotation invariant LBP operator for any quantization of the angular space and for any spatial resolution, and present a method for combining multiple operators for multiresolution analysis. The proposed approach is very robust in terms of gray scale variations, since the operator is by definition invariant against any monotonic transformation of the gray scale. Another advantage is computational simplicity, as the operator can be realized with a few operations in a small neighborhood and a lookup table. Excellent experimental results obtained in a true problem of rotation invariance, where the classifier is trained at one particular rotation angle and tested with samples from other rotation angles, demonstrate that good discrimination can be achieved with the occurrence statistics of simple rotation invariant local binary patterns. These operators characterize the spatial configuration of local image texture and the performance can be further improved by combining them with rotation invariant variance measures that characterize the contrast of local image texture. The joint distributions of these orthogonal measures are shown to be very powerful tools for rotation invariant texture analysis. nonparametric texture analysis distribution histogram Brodatz contrast

show abstract

Image Texture

Manjunath¹,

Chellappa²

1999

Wiley Encyclopedia of Electrical and Electronics Engineering

View full text Add to dashboard Cite

The sections in this article are Texture Classification and Segmentation Gray Level Image Statistics and CO‐Occurrence Matrices Random Field Models Spatial Filtering Rotation Invariant Texture Classification Applications Summary Acknowledgments

show abstract

A Model-Based Method for Rotation Invariant Texture Classification

Cited by 304 publications

References 21 publications

Rotation-invariant texture classification using feature distributions

Rotation-invariant texture classification using feature distributions

A Generalized Local Binary Pattern Operator for Multiresolution Gray Scale and Rotation Invariant Texture Classification

Image Texture

Contact Info

Product

Resources

About