access icon free Spatial-spectral classification of hyperspectral images: a deep learning framework with Markov Random fields based modelling

For the spatial-spectral classification of hyperspectral images (HSIs), a deep learning framework is proposed in this study, which consists of convolutional neural networks (CNNs) and Markov random fields (MRFs). Firstly, a CNN model to learn the deep spectral feature from the HSI is built and the class posterior probability distribution is estimated. The CNN with a dropout layer can relieve the overfitting in classification. The CNN is utilised as a pixel-classifier, so it only works in the spectral domain. Then, the spatial information will be encoded by MRF-based multilevel logistic prior for regularising the classification. To derive the correlation of both spectral and spatial features for improving algorithm performance, the marginal probability distribution in HSI is learned using MRF-based loopy belief propagation. In comparison with several state-of-the-art approaches for data classification on three publicly available HSI datasets, experimental results have demonstrated the superior performance of the proposed methodology.

Inspec keywords: image classification; statistical distributions; neural nets; geophysical image processing; probability; Markov processes; hyperspectral imaging; regression analysis; belief networks; remote sensing; learning (artificial intelligence)

Other keywords: MRF-based loopy belief propagation; CNN model; marginal probability distribution; deep spectral feature; class posterior probability distribution; deep learning framework; spatial information; HSI; spatial-spectral classification; convolutional neural networks; spectral features; hyperspectral images; data classification; spatial features; Markov random fields

Subjects: Markov processes; Probability theory, stochastic processes, and statistics; Computer vision and image processing techniques; Geography and cartography computing; Markov processes; Other topics in statistics; Geophysical techniques and equipment; Image recognition; Other topics in statistics; Instrumentation and techniques for geophysical, hydrospheric and lower atmosphere research; Neural computing techniques; Knowledge engineering techniques

http://iet.metastore.ingenta.com/content/journals/10.1049/iet-ipr.2018.5727
Loading

Related content

content/journals/10.1049/iet-ipr.2018.5727
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading