Fusion of heterogeneous bands and kernels in hyperspectral image processing

Abstract

Hyperspectral imaging is a powerful technology that is plagued by large dimensionality. Herein, we explore a way to combat that hindrance via non-contiguous and contiguous (simpler to realize sensor) band grouping for dimensionality reduction. Our approach is different in the respect that it is flexible and it follows a well-studied process of visual clustering in high-dimensional spaces. Specifically, we extend the improved visual assessment of cluster tendency and clustering in ordered dissimilarity data unsupervised clustering algorithms for supervised hyperspectral learning. In addition, we propose a way to extract diverse features via the use of different proximity metrics (ways to measure the similarity between bands) and kernel functions. The discovered features are fused with l ∞ -norm multiple kernel learning. Experiments are conducted on two benchmark datasets and our results are compared to related work. These datasets indicate that contiguous or not is application specific, but heterogeneous features and kernels usually lead to performance gain.Hyperspectral imaging is a demonstrated tool for numerous earth and space-borne applications involving target detection, 1-3 invasive species monitoring, 4, 5 and precision agriculture. 6, 7 However, the field suffers from the "curse of dimensionality" (spatial, spectral and temporal). Of interest is new theory for dimensionality reduction or identification of fewer spectral bands for purposes like multispectral vs hyperspectral imaging, typically relative to some task, which ultimately aids efficient computation, storage, transmission, classification and lower system cost. While numerous noteworthy methods have been explored, effectiveness and efficiency of search, fusion and classification remain unsolved.