Polarimetric synthetic aperture radar image classification by a hybrid method

Khan, Kamran Ullah; Yang, Jian

doi:10.1016/s1007-0214(07)70015-9

Cited by 10 publications

(6 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Wishart maximum likelihood (WML) method has often been used for PolSAR classification [ 3 ]. However, it does not take explicitly into consideration the phase information contained within polarimetric data, which plays a direct role in the characterization of a broad range of scattering processes.…”

Section: Introductionmentioning

confidence: 99%

Crop Classification by Forward Neural Network with Adaptive Chaotic Particle Swarm Optimization

Zhang

2011

Sensors

104

View full text Add to dashboard Cite

This paper proposes a hybrid crop classifier for polarimetric synthetic aperture radar (SAR) images. The feature sets consisted of span image, the H/A/α decomposition, and the gray-level co-occurrence matrix (GLCM) based texture features. Then, the features were reduced by principle component analysis (PCA). Finally, a two-hidden-layer forward neural network (NN) was constructed and trained by adaptive chaotic particle swarm optimization (ACPSO). K-fold cross validation was employed to enhance generation. The experimental results on Flevoland sites demonstrate the superiority of ACPSO to back-propagation (BP), adaptive BP (ABP), momentum BP (MBP), Particle Swarm Optimization (PSO), and Resilient back-propagation (RPROP) methods. Moreover, the computation time for each pixel is only 1.08 × 10−7 s.

show abstract

Section: Introductionmentioning

confidence: 99%

Crop Classification by Forward Neural Network with Adaptive Chaotic Particle Swarm Optimization

Zhang

2011

Sensors

104

View full text Add to dashboard Cite

show abstract

“…While, Rodionova [4] demonstrated that textural features computed in every scattering categories of Freeman and Durden decomposition make better object discrimination of PolSAR images. In [5], good classification results have been achieved using neural network with a feature set including undecimated wavelet, transform-based features and texture features along with nonlinear features and a partial set from the elements of the coherence matrix. Liang, [6] also, investigated different texture features using neural network classifier and studied their performances.…”

Section: Introductionmentioning

confidence: 99%

Polarimetric SAR images segmentation incorporating texture features

Kourgli

Oukil

Hirche

et al. 2011

2011 17th International Conference on Digital Signal Processing (DSP)

View full text Add to dashboard Cite

Many polarimetric classification algorithms have been proposed in the literature. It has been shown that the use of spatial information provides better sensitivity for class separation such as forests. In this paper, we wish to address this issue, testing and comparing two polarimetric SAR (Synthetic Aperture Radar) segmentation approaches incorporating contextual information. The first approach is contextual fuzzy clustering based on the use of bias correction defined by a texture feature, while the second one is a Markovian segmentation based on non parametric textural modeling. These approaches have been tested on Oberpfaffenhofen area in Munich and the PolSAR images are acquired in the P band. In both cases, texture considering allowed to improve greatly the rates of good identification.

show abstract

“…In this paper, the NN was adopted since the performance of NN classifiers is independent of the type of distribution while depends only on the training data and the discrimination power of the features [7,8]. Classification accuracy depends mainly on the quality of features, which should be robust with maximum discrimination power and must encompass most of the information available in the data [9].…”

Section: Introductionmentioning

confidence: 99%

A New Classifier for Polarimetric Sar Images

Zhang¹,

Wu²,

Wei³

2009

PIER

View full text Add to dashboard Cite

Abstract-This paper proposes a hybrid classifier for polarimetric SAR images. The feature sets consist of span image, the H/A/α decomposition, and the gray-level co-occurrence matrix (GLCM) based texture features. Then, the features are reduced by principle component analysis (PCA). A 3-layer neural network (NN) is constructed, trained by resilient back-propagation (RPROP) method to fasten the training and early stop (ES) method to prevent the overfitting. The results of San Francisco and Flevoland sites compared to Wishart Maximum Likelihood and wavelet-based method demonstrate the validness of our method in terms of confusion matrix and overall accuracy. In addition, NNs with and without PCA are compared. Results show the NN with PCA is more accurate and faster.

show abstract

Polarimetric synthetic aperture radar image classification by a hybrid method

Cited by 10 publications

References 25 publications

Crop Classification by Forward Neural Network with Adaptive Chaotic Particle Swarm Optimization

Crop Classification by Forward Neural Network with Adaptive Chaotic Particle Swarm Optimization

Polarimetric SAR images segmentation incorporating texture features

A New Classifier for Polarimetric Sar Images

Contact Info

Product

Resources

About