Combining multiple segmentation methods for improving the segmentation accuracy

In this paper, an alternative method based on decision fusion is presented to improve the segmentation accuracy. The proposed method concludes multiple methods instead of a single one. It consists of a set of segmentation methods that are consulted in parallel. The decisions of the various methods are then combined by a fusion module. The individual methods, in this case, are capable of independent and simultaneous operation. Then, we apply and compare the fusion schemes to the area of image segmentation. We seek answers to the questions: Can combining multiple segmentation methods achieve better, final partitioning of an image? If so, how much is this improvement? And which fusion scheme may perform best of all?