Since the spatial distribution of surface materials is usually regular and locally continuous,it is reasonable to classify the hyperspectral images (HSI) from superpixel viewpoint,which can be considered as a process of segmenting the spatial image into many regions.In this paper,a superpixel-level Gabor feature fusion approach (abbreviated as SPGF) has been proposed for hyperspectral image classification.Firstly,a set of predefined two-dimensional (2D) Gabor filters are applied to hyperspectral images to extract sufficient features.Meanwhile,a classic superpixel segmentation method,called simple linear iterative clustering (SLIC),is adopted to divide the original hyperspectral image into disjoint superpixels.Secondly,the Support Vector Machine classifier (SVM) is applied on each extracted 2D Gabor feature cube,and the majority voting strategy is adopted to combine the classification results.Finally,the superpixel map obtained by SLIC is used to regularize the classification map.Extensive experiments on two real hyperspectral data sets have demonstrated higher performance of the proposed SPGF approach over several state-of-the-art methods in the literature.