Advanced fractal approach for unsupervised classification of SAR images

Pant, Triloki; Singh, Dharmendra; Srivastava, Tanuja

doi:10.1016/j.asr.2010.01.008

Cited by 26 publications

(26 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, combining other texture measures or using multifractal dimensions were also proposed to overcome the limitation of using one single fractal dimension. Pant, Singh, and Srivastava (2010) used both fractal dimension and spatial autocorrelation statistics (Moran's I) and the combined texture image yielded better classification results. Improved classification accuracies were also achieved by adopting multi-fractal textural descriptors, since multi-fractal measures consider the whole spectrum of dimensions and thus are able to provide a higher level of textural information (Parrinello and Vaughan 2006;Wawrzaszek, Krupinski, and Aleksandrowicz.…”

Section: Pre-classificationmentioning

confidence: 99%

A survey of methods incorporating spatial information in image classification and spectral unmixing

Wang

Shi

Diao

et al. 2016

International Journal of Remote Sensing

View full text Add to dashboard Cite

Over the past decade, the incorporation of spatial information has drawn increasing attention in multispectral and hyperspectral data analysis. In particular, the property of spatial autocorrelation among pixels has shown great potential for improving understanding of remotely sensed imagery. In this paper, we provide a comprehensive review of the state-of-the-art techniques in incorporating spatial information in image classification and spectral unmixing. For image classification, spatial information is accounted for in the stages of pre-classification, sample selection, classifiers, post-classification, and accuracy assessment. With regards to spectral unmixing, spatial information is discussed in the context of endmember extraction, selection of endmember combinations, and abundance estimation. Finally, a perspective on future research directions for advancing spatial-spectral methods is offered.

show abstract

Section: Pre-classificationmentioning

confidence: 99%

A survey of methods incorporating spatial information in image classification and spectral unmixing

Wang

Shi

Diao

et al. 2016

International Journal of Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The criteria for the selection of window size are based on the resolution, classification specificity and the nature of the classes. However, windows with the size of 5×5 to 11×11 are commonly used for texture extraction from moderate resolution images [Pant et al, 2010]. Figure 2 shows the used membership function for the input and output parameters.…”

Section: Fractal Feature Vectormentioning

confidence: 99%

Contextual PolSAR image classification using fractal dimension and support vector machines

Aghababaee

Amini

Tzeng

2013

European Journal of Remote Sensing

View full text Add to dashboard Cite

In this paper, a new classification scheme of polarimetric synthetic aperture radar (PolSAR) images using fractal dimension as contextual information is proposed. Support vector machines (SVM) due to their ability to handle the nonlinear classifier problem are applied to a new fractal feature vector, which is constructed from Pauli decomposed vector and fractal dimensions. Fractal dimension is computed based on the concepts of fractional Brownian motion (fBm) and wavelet multi-resolution analysis using a self-adaptive window approach and fuzzy logic. The experimental results on AIRSAR images prove effectiveness of the proposed vector in comparison to the Pauli decomposed vector.

show abstract

“…The method is an extension of the variation method, which takes care of the variation of pixels in local neighbourhood and hence the distribution of pixels. The detailed mathematical description of the method can be found in Berizzi et al (2006) and Pant et al (2010). It is a method recently developed for fractal dimension estimation and easy to implement, and so it is used for the present study.…”

Section: Estimation Of Fractal Dimension By Two-dimensional Variationmentioning

confidence: 99%

The potential application of fractal approach for surface roughness retrieval: A study for simulated surfaces

Pant

Singh

Srivastava

2010

Geomatics, Natural Hazards and Risk

Self Cite

View full text Add to dashboard Cite

Natural surfaces can be modelled with fractals because fractals properly account for scale invariance and self similarity of these surfaces. The well-known measure, i.e. fractal dimension (D), is the property used to describe the roughness of fractal surfaces. Retrieving the Earth's surface roughness with satellite images, particularly synthetic aperture radar (SAR) images, is an interesting and challenging task. Consequently, many researchers are using electromagnetic models, various inversion techniques, semi-empirical models to retrieve the roughness parameters, i.e. RMS surface height (s) and autocorrelation length (l). Most of the models require some a priori information or some given values to solve the equations and retrieve l and s. Uncertainty still exists to retrieve these parameters with minimum or no a priori information. Therefore, in this paper, the fractal dimension approach has been applied to correlate l and s with fractal properties for development of a surface parameter retrieval algorithm. For this purpose, 1500 synthetic surfaces for known l and s have been generated, and their fractal dimension has been computed. D has also been computed after introducing Gaussian and speckle noise to the generated surfaces. The analysis among D, l and s shows the potentiality of relationship among these parameters and is helpful in developing a relationship among them by which l and s can be retrieved. The values of l and s are retrieved with the help of a look-up table for the synthetic surfaces which can be extended for retrieval of roughness parameters from the SAR images.

show abstract

Advanced fractal approach for unsupervised classification of SAR images

Cited by 26 publications

References 21 publications

A survey of methods incorporating spatial information in image classification and spectral unmixing

A survey of methods incorporating spatial information in image classification and spectral unmixing

Contextual PolSAR image classification using fractal dimension and support vector machines

The potential application of fractal approach for surface roughness retrieval: A study for simulated surfaces

Contact Info

Product

Resources

About