Matching suitable feature construction for SAR images based on evolutionary synthesis strategy

Abstract In the paper, a set of algorithms to construct synthetic aperture radar (SAR) matching suitable features are firstly proposed based on the evolutionary synthesis strategy. During the process, on the one hand, the indexes of primary matching suitable features (PMSFs) are designed based on the characteristics of image texture, SAR imaging and SAR matching algorithm, which is a process involving expertise; on the other hand, by designing a synthesized operation expression tree based on PMSFs, a much more flexible expression form of synthesized features is built, which greatly expands the construction space. Then, the genetic algorithm-based optimized searching process is employed to search the synthesized matching suitable feature (SMSF) with the highest efficiency, largely improving the optimized searching efficiency. In addition, the experimental results of the airborne synthetic aperture radar ortho-images of C-band and P-band show that the SMSFs gained via the algorithms can reflect the matching suitability of SAR images accurately and the matching probabilities of selected matching suitable areas of ortho-images could reach 99 ± 0.5%.

[1]  Rafael García,et al.  Match Selection in Batch Mosaicing Using Mutual Information , 2009, IbPRIA.

[2]  Krzysztof Krawiec,et al.  Visual Learning by Evolutionary and Coevolutionary Feature Synthesis , 2007, IEEE Transactions on Evolutionary Computation.

[3]  Jason M. Daida,et al.  Genetic Programming for Automatic Target Classification and Recognition , 1998, Evolutionary Programming.

[4]  Bir Bhanu,et al.  Evolutionary feature synthesis for object recognition , 2005, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[5]  Bir Bhanu,et al.  Genetic algorithm based feature selection for target detection in SAR images , 2003, Image Vis. Comput..

[6]  Bin Li,et al.  Selection of matching area in SAR scene-matching-aided navigation based on manifold learning , 2011, International Symposium on Multispectral Image Processing and Pattern Recognition.

[7]  Lincheng Shen,et al.  Constriction of Mutual Information Based Matching-Suitable Features for SAR Image Aided Navigation , 2009, 2009 International Conference on Environmental Science and Information Application Technology.

[8]  Lincheng Shen,et al.  A Point-feature Method for Multi-source SAR Image Matching Suitability Analysis , 2012 .

[9]  Damian M. Lyons Selection and recognition of landmarks using terrain spatiograms , 2010, 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[10]  Huan Liu,et al.  Feature Selection for Classification , 1997, Intell. Data Anal..

[11]  Bir Bhanu,et al.  Image retrieval with feature selection and relevance feedback , 2010, 2010 IEEE International Conference on Image Processing.

[12]  Jean-Marc Le Caillec,et al.  Intensity-Based Block Matching Algorithm for Mosaicing Sonar Images , 2011, IEEE Journal of Oceanic Engineering.

[13]  Bir Bhanu,et al.  Tracking pedestrians with bacterial foraging optimization swarms , 2011, 2011 IEEE Congress of Evolutionary Computation (CEC).

[14]  Wei Jiao,et al.  Matching area intelligent selection method in geomagnetic navigation , 2010, 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems.

[15]  Gang Li,et al.  Automatic building extraction based on improved watershed segmentation, mutual information match and snake model , 2012, Int. J. Comput. Appl. Technol..

[16]  Liang Pan,et al.  A Novel Data Fusion Method Based on Signal's Reliability , 2013 .

[17]  Hongdu Aviation Automatic suitable-matching area selection method based onmulti-feature fusion , 2011 .

[18]  Bir Bhanu,et al.  Evolutionary feature synthesis for facial expression recognition , 2006, Pattern Recognit. Lett..

[19]  Majid Ahmadi,et al.  Illumination invariant feature extraction and mutual-information-based local matching for face recognition under illumination variation and occlusion , 2011, Pattern Recognit..