Methodology for hyperspectral image classification using novel neural network
- Oak Ridge National Lab., TN (United States)
- Jet Propulsion Laboratory, Pasadena, CA (United States)
A novel feed forward neural network is used to classify hyperspectral data from the AVIRIS sector. The network applies an alternating direction singular value decomposition technique to achieve rapid training times (few seconds per class). Very few samples (10-12) are required for training. 100% accurate classification is obtained using test data sets. The methodology combines this rapid training neural network together with data reduction and maximal feature separation techniques such as principal component analysis and simultaneous diagonalization of covariance matrices, for rapid and accurate classification of large hyperspectral images. The results are compared to those of standard statistical classifiers. 21 refs., 3 figs., 5 tabs.
- Research Organization:
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
- Sponsoring Organization:
- National Aeronautics and Space Administration, Washington, DC (United States)
- OSTI ID:
- 488737
- Report Number(s):
- CONF-970465-12; ON: DE97005147; CNN: NASA STTR Phase I Contract No. NAS5-3295
- Resource Relation:
- Conference: SPIE international conference, Orlando, FL (United States), 21-25 Apr 1997; Other Information: PBD: 1997
- Country of Publication:
- United States
- Language:
- English
Similar Records
Application of neural networks for sea ice classification in polarimetric SAR images
Application of convolutional neural networks for stellar spectral classification