An objective global evaluation algorithm of segmentation results is proposed

which is based on computing the deviation of the segmentation results from reference segmentation.By taking into account image understanding view

the discrepancy between two results is weighted based on spatial contextual information

thereby obtaining the different weightiness.Moreover

relative distance

foreground and background searching distance are introduced into the proposed metric

and the value of this metric is invariant to the segmentation images with varying scale.In order to overcome the irrational evaluation caused by the excessive over-segmentation and under-segmentation

a penalized term is proposed.Experiment results prove that our evaluation accords with the view of the image understanding and it is more global and robust.