Date of Award
Doctor of Philosophy
Recently, superpixel-based approaches have been proposed for hyperspectral unmixing. The basic assumption of this approach is that the superpixel over-segmentation segments the image into small homogeneous areas. A low-dimensional (LD) image representation is obtained by using the average of the superpixels, which are then used in other image processing tasks up the processing chain. Due to superpixel-segmentation algorithm limitations, the region inside a superpixel may not be homogeneous. Therefore, the average may not be an adequate representation for the superpixel, leading to inaccuracies in the low dimensional representation resulting in errors in the image processing tasks or analysis. Here we present an improved superpixel-based dimensionality reduction approach that accounts for superpixel homogeneity. Homogeneous superpixels are represented by their mean but heterogeneous superpixels are represented by multiple representative signatures selected using the SVDSS column subset selection algorithm. The representative signatures for the homogeneous and heterogeneous superpixels provide an improved low-dimensional representation for the hyperspectral image that better captures the image structure. We present experiments applying the proposed and the conventional superpixel dimensionality reduction approaches to unmixing using the constrained non-negative matrix factorization (cNMF). Experimental results comparing the proposed approach with other superpixel-based unmixing approaches are presented. Experimental results show that unmixing results are improved by the enhanced superpixel-based low dimensional representation.
Received from ProQuest
Yi, Jiarui, "Superpixel-Based Unsupervised Hyperspectral Unmixing" (2018). Open Access Theses & Dissertations. 1562.
Available for download on Monday, August 24, 2020