Texture Classification Using Rotation- and Scale-Invariant Gabor Texture Features

Riaz, Farhan; Hassan, Ali; Rehman, Saad; Qamar, Usman

doi:10.1109/lsp.2013.2259622

Cited by 89 publications

(28 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Gabor transform is closely related to human visual, and has been widely used in face recognition [46], texture classification [47] and information mining [48]. As the 3D extension of the traditional two-dimensional Gabor (2D-Gabor), 3D-Gabor is a powerful tool for feature extraction of the HSI data.…”

Section: D-gabormentioning

confidence: 99%

3D-Gabor Inspired Multiview Active Learning for Spectral-Spatial Hyperspectral Image Classification

Hu¹,

He²,

Li³

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

Active learning (AL) has been shown to be very effective in hyperspectral image (HSI) classification. It significantly improves the performance by selecting a small quantity of the most informative training samples to reduce the complexity of classification. Multiview AL (MVAL) can make the comprehensive analysis of both object characterization and sampling selection in AL by using various features of multiple views. However, the original MVAL cannot effectively exploit the spectral-spatial information by respecting the three-dimensional (3D) nature of the HSI and the query selection strategy in the MVAL is only based on the disagreement of multiple views. In this paper, we propose a 3D-Gabor inspired MVAL method for spectral-spatial HSI classification, which consists of two main steps. First, in the view generation step, we adopt a 3D-Gabor filter to generate multiple cubes with limited bands and utilize the feature assessment strategies to select cubes for constructing views. Second, in the sampling selection step, a novel method is proposed by using both internal and external uncertainty estimation (IEUE) of views. Specifically, we use the distributions of posterior probability to learn the "internal uncertainty" of each independent view, and adopt the inconsistencies between views to estimate the "external uncertainty". Classification accuracies of the proposed method for the four benchmark HSI datasets can be as high as 99.57%, 99.93%, 99.02%, 98.82%, respectively, demonstrating the improved performance as compared with other state-of-the-art methods.

show abstract

Section: D-gabormentioning

confidence: 99%

3D-Gabor Inspired Multiview Active Learning for Spectral-Spatial Hyperspectral Image Classification

Hu¹,

He²,

Li³

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…For the application at hand, rotation and scale dependence are undesirable for the characterization of textures in the VHR image. Instead, we apply a recent technique based on filtering with Gabor functions, followed by the 2D Discrete Fourier Transform of the mean and standard deviation values of the filter response [8]. Gabor functions are the product of a plane wave function and a 2D Gaussian.…”

Section: Valorization Of Vhr Imagerymentioning

confidence: 99%

“…This paper uses textural features extracted from a very high resolution (VHR) orthophoto for improving domain adaptation of two HS images pertaining to the same geographical area but recorded in different flight lines. These features, proposed in [8], are derived from filtering the VHR image with a Gabor filter bank and making them scale and rotation invariant using the shift invariance property of the Discrete Fourier Transform.…”

Section: Introductionmentioning

confidence: 99%

Spectral adaptation of hyperspectral flight lines using VHR contextual information

Jacobs

Thoonen

Tuia

et al. 2014

2014 IEEE Geoscience and Remote Sensing Symposium

View full text Add to dashboard Cite

Due to technological constraints, hyperspectral earth observation imagery are often a mosaic of overlapping flight lines collected in different passes over the area of interest. This causes variations in aqcuisition conditions such that the reflected spectrum can vary significantly between these flight lines. Partly, this problem is solved by atmospherical correction, but residual spectral differences often remain. A probabilistic domain adaptation framework based on graph matching using Hidden Markov Random Fields was recently proposed for transforming hyperspectral data from one image to better correspond to the other. This paper investigates the use of scale and angle invariant textural features for improving the performance of the used Hidden Markov Random Field matching framework in the case of hyperspectral flight lines. These textural features are derived from the filtering of VHR optical imagery with a bank of Gabor filters with varying orientation, scale and frequency and subsequently rendering them invariant to scale and frequency by applying the 2D DFT on the filter responses in the scale and frequency space.

show abstract

“…Lee reformulated 2D Gabor wavelets based on physiological evidence and the wavelet frame criterion [52]. 2D Gabor filters have been extensively applied to many image processing and computer vision applications [53,54]. We used the form of 2D Gabor functions derived by Lee [52],…”

Section: Gabor Magnitude-based Detectormentioning

confidence: 99%

Fast marching over the 2D Gabor magnitude domain for tongue body segmentation

Cui

Zhang

et al. 2013

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

Tongue body segmentation is a prerequisite to tongue image analysis and has recently received considerable attention. The existing tongue body segmentation methods usually involve two key steps: edge detection and active contour model (ACM)-based segmentation. However, conventional edge detectors cannot faithfully detect the contour of the tongue body, and the initialization of ACM suffers from the edge discontinuity problem. To address these issues, we proposed a novel tongue body segmentation method, GaborFM, which initializes ACM by performing fast marching over the two-dimensional (2D) Gabor magnitude domain of the tongue images. For the enhancement of the contour of the tongue body, we used the 2D Gabor magnitude-based detector. To cope with the edge discontinuity problem, the fast marching method was utilized to connect the discontinuous contour segments, resulting in a closed and continuous tongue body contour for subsequent ACM-based segmentation. Qualitative and quantitative results showed that GaborFM is superior to the other methods for tongue body segmentation.

show abstract

Texture Classification Using Rotation- and Scale-Invariant Gabor Texture Features

Cited by 89 publications

References 13 publications

3D-Gabor Inspired Multiview Active Learning for Spectral-Spatial Hyperspectral Image Classification

3D-Gabor Inspired Multiview Active Learning for Spectral-Spatial Hyperspectral Image Classification

Spectral adaptation of hyperspectral flight lines using VHR contextual information

Fast marching over the 2D Gabor magnitude domain for tongue body segmentation

Contact Info

Product

Resources

About