Pseudolabel Guided Kernel Learning for Hyperspectral Image Classification

In this paper, we propose a new framework for hyperspectral image classification, namely pseudolabel guided kernellearning (PLKL). The proposed framework is capable of fully utilizing unlabeled samples, making it very effective to handle the taskwith extremely limited training samples. Specifically, with multipleinitial kernels and labeled samples, we first employ support vectormachine (SVM) classifiers to predict pseudolabels independentlyfor each unlabeled sample, and consistency voting is applied tothe resulting pseudolabels to select and add a few unlabeled samples to the training set. Then, we refine the kernels to improvetheir discriminability with the augmented training set and a typical kernel learning method. Such phases are repeated until stable.Furthermore, we enhance the PLKL in terms of both the computation and memory efficiencies by using a bagging-like strategy,improving its practicality for large scale datasets. In addition, theproposed framework is quite flexible and general. That is, otheradvanced kernel-based methods can be incorporated to continuously improve the performance. Experimental results show thatthe proposed frameworks achieve much higher classification accuracy, compared with state-of-the-art methods. Especially, the classification accuracy improves more than 5% with very few trainingsamples.