Fully abundance-constrained endmember finding for hyperspectral images

Fully constrained least squares (FCLS) method has been widely used in linear spectral mixture analysis (LSMA). This paper presents two versions of FCLS, to be called SeQuential FCLS (SQ-FCLS) and SuCcessive FCLS (SC-FCLS) to find endmembers. In order to address random issues in initial conditions, two versions of FCLS, Iterative FCLS (I-FLCS) and Random FCLS (RFCLS) are also developed.