Improving remote sensing image classification by exploiting adaptive features and hierarchical hybrid decision trees

ABSTRACT In this letter, a hierarchical hybrid decision tree (HDT) framework based on the complementary properties and processing chains is proposed. Unlike the previous HDT framework by multiple features and multiple classifiers (MFMC), the proposed method exploits the complementary score (CS) and the feature selection tools to optimize the processing chains for each node. Consequently, adaptive features are employed to identify each class. The HDT structure is then employed to predict all the pixels after the identification rules for all the nodes are determined. The proposed method and the traditional HDT are tested on the common hyperspectral data. The experimental results demonstrate the effectiveness of our method. It is shown that the overall performance of HDT is improved by optimizing the identification of each node.

[1]  Joydeep Ghosh,et al.  Investigation of the random forest framework for classification of hyperspectral data , 2005, IEEE Transactions on Geoscience and Remote Sensing.

[2]  RokachLior.  Taxonomy for characterizing ensemble methods in classification tasks , 2009 .

[3]  Hongnian Yu,et al.  Maximum relevancy maximum complementary feature selection for multi-sensor activity recognition , 2015, Expert Syst. Appl..

[4]  J. Tenenbaum,et al.  A global geometric framework for nonlinear dimensionality reduction. , 2000, Science.

[5]  Farid Melgani,et al.  Nearest Neighbor Classification of Remote Sensing Images With the Maximal Margin Principle , 2008, IEEE Transactions on Geoscience and Remote Sensing.

[6]  Ming Li,et al.  2D-LDA: A statistical linear discriminant analysis for image matrix , 2005, Pattern Recognit. Lett..

[7]  Ping Zhang,et al.  A novel feature extraction method and hybrid tree classification for handwritten numeral recognition , 2002, Pattern Recognit. Lett..

[8]  Kongqiao Wang,et al.  A Robust Iris Localization Algorithm via Radial Symmetry for Nonideal Capturing Condition , 2009, Int. J. Pattern Recognit. Artif. Intell..

[9]  C. Brodley,et al.  Decision tree classification of land cover from remotely sensed data , 1997 .

[10]  Leen-Kiat Soh,et al.  Texture analysis of SAR sea ice imagery using gray level co-occurrence matrices , 1999, IEEE Trans. Geosci. Remote. Sens..

[11]  Jerry M. Mendel,et al.  Maximum-likelihood classification for digital amplitude-phase modulations , 2000, IEEE Trans. Commun..

[12]  Duncan Fyfe Gillies,et al.  Overfitting in linear feature extraction for classification of high-dimensional image data , 2016, Pattern Recognit..

[13]  D. Reich,et al.  Principal components analysis corrects for stratification in genome-wide association studies , 2006, Nature Genetics.

[14]  Lior Rokach,et al.  Taxonomy for characterizing ensemble methods in classification tasks: A review and annotated bibliography , 2009, Comput. Stat. Data Anal..

[15]  David A. Landgrebe,et al.  A survey of decision tree classifier methodology , 1991, IEEE Trans. Syst. Man Cybern..

[16]  Wei Wei,et al.  Latent subclass learning-based unsupervised ensemble feature extraction method for hyperspectral image classification , 2015 .

[17]  Paolo Gamba,et al.  Combining Hyperspectral Data Processing Chains for Robust Mapping Using Hierarchical Trees and Class Memberships , 2011, IEEE Geoscience and Remote Sensing Letters.

[18]  Xuelong Li,et al.  General Tensor Discriminant Analysis and Gabor Features for Gait Recognition , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[19]  Fuhui Long,et al.  Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy , 2003, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[20]  Paolo Gamba,et al.  Hierarchical Hybrid Decision Tree Fusion of Multiple Hyperspectral Data Processing Chains , 2011, IEEE Transactions on Geoscience and Remote Sensing.

[21]  Kenji Omasa,et al.  Estimation of vegetation parameter for modeling soil erosion using linear Spectral Mixture Analysis of Landsat ETM data , 2007 .