Texture classification with combined rotation and scale invariant wavelet features

Muneeswaran, K.; Ganesan, L.; Arumugam, S.; Soundar, K. Ruba

doi:10.1016/j.patcog.2005.03.021

Cited by 56 publications

(37 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…On image regions extracted from four VisTex and Brodatz image datasets, classification results perform very well and demonstrate to be superior to those reached on the same image datasets by three recent texture classification methods found in literature (Arivazhagan et al, 2006;Muneeswaran et al, 2005;Li et al, 2003 …”

Section: Resultsmentioning

confidence: 76%

“…Ranklets+SVM represents the proposed ranklet-based approach. Ridgelets+Dist (Arivazhagan et al, 2006), Wavelets+Dist (Muneeswaran et al, 2005), and Wavelets+SVM (Li et al, 2003) Li et al (2003) used wavelet transform and SVM (i.e., Wavelets+SVM) on a number of image regions extracted from the same 30 Brodatz images used in Test-4.As evident from the seventh and eighth rows of Tab. 1, they reached 96.34% of accuracy versus the 100.00% achieved by Ranklets+SVM.…”

Section: Resultsmentioning

confidence: 99%

“…First, to propose a system as general-purpose as possible, rather than overly suited for a specific texture classification problem. Second, to be able to compare the proposed system directly with three recent automatic texture classification systems found in literature and having being evaluated on the same images, namely, (1) a texture classification approach based on ridgelets and minimum distance classification (Arivazhagan et al, 2006), (2) a rotation-invariant approach based on wavelets and minimum distance classification (Muneeswaran et al, 2005), and (3) an approach based on wavelets and SVM (Li et al, 2003).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Texture classification using invariant ranklet features

Masotti

Campanini

2008

Pattern Recognition Letters

View full text Add to dashboard Cite

A novel invariant texture classification method is proposed. Invariance to linear/nonlinear monotonic gray-scale transformations is achieved by submitting the image under study to the ranklet transform, an image processing technique relying on the analysis of the relative rank of pixels rather than on their gray-scale value. Some texture features are then extracted from the ranklet images resulting from the application at different resolutions and orientations of the ranklet transform to the image. Invariance to 90 • -rotations is achieved by averaging, for each resolution, correspondent vertical, horizontal, and diagonal texture features. Finally, a texture class membership is assigned to the texture feature vector by using a support vector machine (SVM) classifier. Compared to three recent methods found in literature and having being evaluated on the same Brodatz and Vistex datasets, the proposed method performs better. Also, invariance to linear/non-linear monotonic gray-scale transformations and 90 • -rotations are evidenced by training the SVM classifier on texture feature vectors formed from the original images, then testing it on texture feature vectors formed from contrast-enhanced, gamma-corrected, histogram-equalized and 90 • -rotated images.

show abstract

Section: Resultsmentioning

confidence: 76%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Texture classification using invariant ranklet features

Masotti

Campanini

2008

Pattern Recognition Letters

View full text Add to dashboard Cite

show abstract

“…The proposed method takes some important advantages over classical fractal signature techniques like that based on wavelets 17 or multifractal. 18 One of such advantages is that the technique here presented gathers information from frequency domain inherently, allowing the capturing of details and patterns which escapes from the conventional spatial analysis.…”

Section: Introductionmentioning

confidence: 99%

Fractal descriptors in the Fourier domain applied to color texture analysis

Florindo¹,

Bruno²

2011

Chaos

View full text Add to dashboard Cite

A wavelet transform algorithm for peak detection and application to powder x-ray diffraction data Rev. Sci. Instrum. 82, 015105 (2011) Support vector machine-based feature extractor for L/H transitions in JET Rev. Sci. Instrum. 81, 10E123 (2010) Determining the absorption tolerance of single chromophore photodiodes for machine vision Appl. Phys. Lett. 96, 253303 (2010) Determining the absorption tolerance of single chromophore photodiodes for machine vision APL: Org. Electron. Photonics 3, 123 (2010) The present work proposes the development of a novel method to provide descriptors for colored texture images. The method consists of two steps. First, we apply a linear transform in the color space of the image aiming at highlighting spatial structuring relations among the color of pixels. Second, we apply a multiscale approach to the calculus of fractal dimension based on Fourier transform. From this multiscale operation, we extract the descriptors that are used to discriminate the texture represented in digital images. The accuracy of the method is verified in the classification of two color texture datasets, by comparing the performance of the proposed technique to other classical and state-of-the-art methods for color texture analysis. The results showed an advantage of almost 3% of the proposed technique over the second best approach. Fractal objects constitute a particular category of nonlinear dynamic system. Essentially, fractals are geometric objects which do not obey the classical Euclidian rules. Besides, they are characterized by the self-similarity, that is, we observe some geometrical and/or statistical patterns which repeat themselves along different observation scales. In a true fractal, the self-similarity is observed at infinite observation scales. The inherent composition rules of fractals yield to some interesting characteristics. For instance, fractals present non-proportionality between cause and effect. Still they present infinite complexity (complexity in the sense of observed details) and they show a high dependence level from initial conditions. Such characteristics allow for fractals are also to be considered as a chaotic system. This physical interpretation collaborated for authors, such as Mandelbrot, 1 to suggest the use of fractals to model objects found in the nature. More recently, Manoel et al. 2 developed a method for using fractal theory in the extraction of features from natural objects represented in a digital image. This is an important problem in computational vision. Bruno et al. 3 applied the technique to a shape recognition problem. Here, we propose the development and study of a novel fractal-based method for the analysis of color texture. We verify the efficiency of the proposed approach in the solution of a color texture classification problem. Additional information on Chaos

show abstract

“…However, textures can be classified using a small set of filters, which gives rise to the filter selection problem. The wavelet based texture classifiers are similar to Gabor based methods with the Gabor filters replaced by the Discrete Wavelet Transform (DWT) [4,5,6,7,8,9]. The DWT based methods employ multiscale decomposition to extract high-frequency components from wavelet subbands for texture representation.…”

Section: Introductionmentioning

confidence: 99%

Multiscale texture classification and retrieval based on magnitude and phase features of complex wavelet subbands

Çelik

Tjahjadi

2011

Computers & Electrical Engineering

View full text Add to dashboard Cite

This paper proposes a multiscale texture classifier which uses features extracted from both magnitude and phase responses of subbands at different resolutions of the dual-tree complex wavelet transform decomposition of a texture image. The mean and entropy in the transform domain are used to form a feature vector. The proposed method can achieve a high texture classification rate even for small number of samples used in training stage. This makes it suitable for applications where the number of texture samples used in training is very limited. The superior performance and robustness of the proposed classifier is shown for classifying and retrieving texture images from image databases.

show abstract

Texture classification with combined rotation and scale invariant wavelet features

Cited by 56 publications

References 14 publications

Texture classification using invariant ranklet features

Texture classification using invariant ranklet features

Fractal descriptors in the Fourier domain applied to color texture analysis

Multiscale texture classification and retrieval based on magnitude and phase features of complex wavelet subbands

Contact Info

Product

Resources

About