Multiple classi®er systems (MCSs) based on the combination of outputs of a set of dierent classi®ers have been proposed in the ®eld of pattern recognition as a method of developing high performance classi®cation systems
Multiple classi®er systems (MCSs) based on thecombination of outputs of a set of dierent classi®ershave been proposed in the ®eld of patternrecognition as a method of developing high performanceclassi®cation systems