Abstract
Abstract. Temporal analysis of Landsat 8 images opens up new opportunities in the unmixing procedure. Although spectral analysis of time series Landsat imagery has its own advantage, it has rarely been studied. Nevertheless, using the temporal information can provide improved unmixing performance when compared to independent image analyses. Moreover, different land cover types may demonstrate different temporal patterns, which can aid the discrimination of different natures. Therefore, this letter presents time series K-P-Means, a new solution to the problem of unmixing time series Landsat imagery. The proposed approach is to obtain the “purified” pixels in order to achieve optimal unmixing performance. The vertex component analysis (VCA) is used to extract endmembers for endmember initialization. First, nonnegative least square (NNLS) is used to estimate abundance maps by using the endmember. Then, the estimated endmember is the mean value of “purified” pixels, which is the residual of the mixed pixel after excluding the contribution of all nondominant endmembers. Assembling two main steps (abundance estimation and endmember update) into the iterative optimization framework generates the complete algorithm. Experiments using both simulated and real Landsat 8 images show that the proposed “joint unmixing” approach provides more accurate endmember and abundance estimation results compared with “separate unmixing” approach.
Highlights
Hyperspectral remote sensing technology consists of acquiring a set of images capturing a spatial scene at a few hundreds of contiguous spectral bands
Given an a priori known number of endmembers K, the time series K-P-Means model is based on K-P-Means algorithm, which characterizes a class by the dominant endmember, whose fractional abundance is the biggest (Xu et al, 2014)
We evaluate the performance of time series K-P-Means using the Xi and Ar, hereafter called “joint unmixing”
Summary
Hyperspectral remote sensing technology consists of acquiring a set of images capturing a spatial scene at a few hundreds of contiguous spectral bands. The significant spectral information they convey is somewhat compromised by their lower spatial resolution This limitation, combined with the complexity of the ground targets and environmental conditions, lead to the observed pixel spectra composed of several pure materials– referred to as endmembers. Several attributes of Landsat 8 are wide scope of coverage, higher spatio-temporal resolution and cost-free status These data are well available at regular time intervals. Somers et al (Liu et al, 2012) proposed a method to minimize the effect of endmember variability in the image by grouping multiple spectral signatures (or bundles) to describe a particular endmember class in the scene. Accounting for the temporal information, we will refer to as time series K-P-Means algorithm, which alternates iteratively between two main steps (abundance estimation and endmember update) until convergence to yield final endmember estimates (Xu et al, 2014).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
