Wavelet-Based Classification of Hyperspectral Images Using Extended Morphological Profiles on Graphics Processing Units

摘要: The availability of graphics processing units (GPUs) provides a low-cost solution to real-time processing, which may benefit many remote sensing applications. In this paper, spectral–spatial classification scheme for hyperspectral images is specifically adapted computing on GPUs. It based wavelets, extended morphological profiles (EMPs), and support vector machine (SVM). Additionally, preprocessing stage used remove noise in the original image. local computation techniques proposed makes them particularly suitable parallel by blocks threads GPU. Moreover, block-asynchronous updating process applied EMP speedup reconstruction. results over different show that execution can be speeded up $8.2\times$ compared an efficient OpenMP implementation, achieving image while maintaining high accuracy values scheme.