Mathematical Problems in Engineering

Research Article

Improved Unsupervised Color Segmentation Using a Modified Color Model and a Bagging Procedure in -Means++ Algorithm

Segmentation performance results for image “b” using GMM.


GMM	Cluster 1	Cluster 2	Cluster 3	Cluster 4	Metrics

					Best BAcc.: 0.99833 Mean BAcc: 0.98109 Mean sen.: 0.96164 Mean spe.: 0.98576 Avg. time: 4.6552

n					Best BAcc.: 0.99846 Mean BAcc: 0.95547 Mean sen.: 0.61632 Mean spe.: 0.90544 Avg. time: 6.5616

					Best BAcc.: 0.99733 Mean BAcc: 0.98261 Mean sen.: 0.88530 Mean spe.: 0.98876 Avg. time: 3.3779

					Best BAcc.: 0.99838 Mean BAcc: 0.97665 Mean sen.: 0.84241 Mean spe.: 0.95545 Avg. time: 3.10470

YCrCb					Best BAcc.: 0.99833 Mean BAcc: 0.98288 Mean sen.: 0.91770 Mean spe.: 0.98878 Avg. time: 3.8567

CrCb					Best BAcc.: 0.99829 Mean BAcc: 0.96058 Mean sen.: 0.69106 Mean spe.: 0.91341 Avg. time: 4.0068

					Best BAcc.: 0.99818 Mean BAcc: 0.91195 Mean sen.: 0.84421 Mean spe.: 0.96268 Avg. time: 3.47339

					Best BAcc.: 0.97405 Mean BAcc: 0.91007 Mean sen.: 0.83561 Mean spe.: 0.96637 Avg. time: 3.71440

Orig					Best BAcc.: 0.92188 Mean BAcc: 0.83645 Mean sen.: 0.75261 Mean spe.: 0.86185 Avg. time: 2.5643