An SMP soft classification algorithm for remote sensing

• A symmetric multiprocessing (SMP) version of the CIGSCR algorithm is introduced.
• CIGSCR stands for continuous iterative guided spectral class rejection.
• The CIGSCR algorithm is a soft classification algorithm used on remote sensing images.
• The SMP implementation of CIGSCR demonstrates superlinear speedup for reasonably large images.

This work introduces a symmetric multiprocessing (SMP) version of the continuous iterative guided spectral class rejection (CIGSCR) algorithm, a semiautomated classification algorithm for remote sensing (multispectral) images. The algorithm uses soft data clusters to produce a soft classification containing inherently more information than a comparable hard classification at an increased computational cost. Previous work suggests that similar algorithms achieve good parallel scalability, motivating the parallel algorithm development work here. Experimental results of applying parallel CIGSCR to an image with approximately 108 pixels and six bands demonstrate superlinear speedup. A soft two class classification is generated in just over 4 min using 32 processors.