Validation of Information Products of Airborne Hyperspectral Imagery Processing

In regards to the problem of airborne hyperspectral imagery processing, most attention is paid to the construction of an automated classifier which implements computational procedures of recognizing forest vegetation of different species composition and age. The validation of the information products obtained as a result of processing the data of hyperspectral remote sensing is based on finding a maximum of a posteriori probability for selected classes of objects using the contemporary views of Markov random fields allowing a priori probabilities for each element of hyperspectral apparatus resolution to be calculated. The necessity of taking into account the noise component of the apparatus to enhance the accuracy of solving the applied problems under study is shown. Examples of comparing the data of hyperspectral imagery processing with the data of ground-based forest inventory for a selected territory are demonstrated.

[1]  V. V. Kozoderov,et al.  Recognition of natural and man-made objects in airborne hyperspectral images , 2014, Izvestiya, Atmospheric and Oceanic Physics.

[2]  Stan Z. Li,et al.  Markov Random Field Modeling in Computer Vision , 1995, Computer Science Workbench.

[4]  Jon Atli Benediktsson,et al.  Decision Fusion for the Classification of Urban Remote Sensing Images , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[5]  V. V. Kozoderov,et al.  System for processing of airborne images of forest ecosystems using high spectral and spatial resolution data , 2014, Izvestiya, Atmospheric and Oceanic Physics.

[6]  E. Dmitriev.  Classification of the forest cover of Tver oblast using hyperspectral airborne imagery , 2014, Izvestiya, Atmospheric and Oceanic Physics.

[7]  R. Lucas,et al.  The delineation of tree crowns in Australian mixed species forests using hyperspectral Compact Airborne Spectrographic Imager (CASI) data , 2006 .

[8]  Eric R. Ziegel,et al.  The Elements of Statistical Learning , 2003, Technometrics.

[9]  V. V. Kozoderov,et al.  Optimization of spectral bands for hyperspectral remote sensing of forest vegetation , 2013, Remote Sensing.

[10]  Azriel Rosenfeld,et al.  Compact Object Recognition Using Energy-Function-Based Optimization , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[11]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[12]  Lorenzo Bruzzone,et al.  On the role of spectral resolution and classifier complexity in the analysis of hyperspectral images of forest areas , 2007, SPIE Remote Sensing.

[13]  G. Asner,et al.  Spectral unmixing of vegetation, soil and dry carbon cover in arid regions: Comparing multispectral and hyperspectral observations , 2002 .

[14]  David A. Landgrebe,et al.  Covariance Matrix Estimation and Classification With Limited Training Data , 1996, IEEE Trans. Pattern Anal. Mach. Intell..