As a powerful classifier, sparse representation-based classification (SRC) has successfully been applied in various visual recognition problems. However, due to the highly correlated bands and insufficient training samples of hyperspectral image (HSI) data, it still remains a challenging problem to effectively apply SRC in HSI. Considering therich information of spatial structure of materials in HSI, that means the adjacent pixelsbelong to the same class with a high probability, in this paper, we propose an efficientsuperpixel-based sparse representation framework for HSI classification. Each superpixel can be regarded as a small region consisting of a number of pixels with similar spectral characteristics. The proposed framework utilizes superpixel to exploit spatial information which can greatly improve classification accuracy. Specifically, SRC is firstly usedto classify the HSI data. Meanwhile, an efficient segmentation algorithm is applied todivide the HSI into many disjoint superpixels. Then, each superpixel is used to fuse the SRC classification results in superpixel level. Experimental results on two real-world HSIdata sets have shown that the proposed superpixel-based SRC (SP-SRC) framework hasa significant improvement over the pixel-based SRC method.