摘要: The hyperspectral images are conventionally classified using spectral information. Spectral information does not include the neighborhood relationship of pixels. In this paper we classfy by state-of-the-art superpixel methods and compare their performances. Since superpixels take into account pixels, benefit from cooperation spatial superpixel-based classification method outperforms pixel-wise methods.
sci-hub.se 参考文章(6)
Andrea Vedaldi, Stefano Soatto, Quick Shift and Kernel Methods for Mode Seeking european conference on computer vision. pp. 705- 718 ,(2008) , 10.1007/978-3-540-88693-8_52
Rainer Stiefelhagen, Mika Fischer, Alexander Schick, Measuring and evaluating the compactness of superpixels international conference on pattern recognition. pp. 930- 934 ,(2012)
Zhi He, Yue Shen, Miao Zhang, Qiang Wang, Yan Wang, Renlong Yu, Spectral-spatial hyperspectral image classification via SVM and superpixel segmentation instrumentation and measurement technology conference. pp. 422- 427 ,(2014) , 10.1109/I2MTC.2014.6860780
Shanshan Li, Xiuping Jia, Bing Zhang, Superpixel-based Markov random field for classification of hyperspectral images international geoscience and remote sensing symposium. pp. 3491- 3494 ,(2013) , 10.1109/IGARSS.2013.6723581
Jianbo Shi, J. Malik, Normalized cuts and image segmentation IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. 22, pp. 888- 905 ,(2000) , 10.1109/34.868688
Ming-Yu Liu, Oncel Tuzel, Srikumar Ramalingam, Rama Chellappa, Entropy rate superpixel segmentation computer vision and pattern recognition. pp. 2097- 2104 ,(2011) , 10.1109/CVPR.2011.5995323