Temporal updating scheme for probabilistic neural network with applicationto satellite cloud classification - Further results

A novel temporal updating approach for probabilistic neural network (PNN) c lassifiers was developed [1] to account for temporal changes of spectral an d temperature features of clouds in the visible and infrared (IR) GOES 8 (G eostationary Operational Environmental Satellite) imagery data. In this bri ef paper, a new method referred to as moving singular value decomposition ( MSVD) is introduced to improve the classification rate of the boundary bloc ks or blocks containing cloud types with nonuniform texture. The MSVD metho d is then incorporated into the temporal updating scheme and its effectiven ess is demonstrated on several sequences of GOES 8 cloud imagery data. Thes e results indicate that the incorporation of the new MSVD improves the over all performance of the temporal updating process by almost 10%.