Improved kN Nearest Neighbor Estimation Algorithm for SAR Image Clutter Statistical Modeling

In order to solve the clutter background problem of Synthetic Aperture Radar (SAR) image in non-uniform clutter environment, a non-parametric statistical modeling algorithm for SAR image estimation based on kN nearest neighbor estimation is proposed. The average region volume is defined, and the nearest neighbors are accurately set according to the frequency of occurrence of each observed sample value. At the same time, the average regional volume centered on the observation point is set in order to prevent the influence of extreme values. The problem that the original algorithm is incorrectly estimated, time consuming or even impossible to estimate due to the fixed number of neighbors is solved. The algorithm does not need to interfere with the prior knowledge of the target. Based on the experimental results of real SAR image data, the accuracy, robustness and simplicity of the proposed method for background clutter modeling of SAR images are verified.

[1]  Peter F. McGuire,et al.  Target detection in synthetic aperture radar imagery: a state-of-the-art survey , 2013, 1804.04719.

[2]  Li Yan.  Ship detection in polarimetric SAR images based on the conditional entropy and Parzen windows , 2012 .

[3]  Matthew P. Wand,et al.  Kernel Smoothing , 1995 .

[4]  Amin Kargarian,et al.  Parzen Window Density Estimator-Based Probabilistic Power Flow With Correlated Uncertainties , 2016, IEEE Transactions on Sustainable Energy.

[5]  Gui Gao,et al.  A Parzen-Window-Kernel-Based CFAR Algorithm for Ship Detection in SAR Images , 2011, IEEE Geoscience and Remote Sensing Letters.

[6]  Xie Juan-ying,et al.  K-nearest neighbors optimized clustering algorithm by fast search and finding the density peaks of a dataset , 2016 .

[7]  Kan Guangyuan.  Study on Application and Comparison of Data-driven Model and Semi-data-driven Model for Rainfall-runoff Simulation , 2017 .

[8]  Xing Mengdao.  An Algorithm of Ship Target Detection Based on the Adaptive Background Window Function , 2013 .

[9]  Mourad Barkat,et al.  Performance analysis of dual automatic censoring and detection in heterogeneous Weibull clutter: A comparison through extensive simulations , 2013, Signal Process..

[10]  Vahid Tabataba Vakili,et al.  Introducing excision switching-CFAR in K distributed sea clutter , 2009, Signal Process..

[11]  Samrat L. Sabat,et al.  Fast Sequency-Ordered Complex Hadamard Transform-Based Parzen Window Entropy Detection for Spectrum Sensing in Cognitive Radio Networks , 2016, IEEE Communications Letters.

[13]  Xi Zhang,et al.  Fast SAR Sea Surface Distribution Modeling by Adaptive Composite Cubic Bézier Curve , 2016, IEEE Geoscience and Remote Sensing Letters.