Bayesian Gravitation-Based Classification for Hyperspectral Images

Integration of spectral and spatial information is extremely important for the classification high-resolution hyperspectral images (HSIs). Gravitation describes interaction among celestial bodies which can be applied to measure similarity between data image classification. However, gravitation hard combine with rarely been in HSI This paper proposes a Bayesian based Classification (BGC) integrate local neighbors training samples. In BGC method, each testing pixel first assumed as massive object unit volume particular density, where density taken mass BGC. Specifically, formulated an exponential function distribution its prior surrounding samples on theorem. Then, joint model developed measure, weigh contribution different region. Four benchmark datasets, i.e. Indian Pines, Pavia University, Salinas, Grss_dfc_2014, are tested verify method. The experimental results compared that several well-known methods, including support vector machines, sparse representation, other eight state-of-the-art methods. shows apparent superiority HSIs also flexibility limited