In this paper, we propose a novel method for morphological hierarchical segmentation. First, we review two state of the art methods: the quasi-flat-zones and the non parametric hierarchical watershed (waterfall algorithm). Then, we propose a novel method which is referred to as the nested homogeneous connections. All three methods have been applied and evaluated on the dataset of one hundred color images from the "Berkeley Segmentation Dataset and Benchmark (BSDB)". The segmentation results have been evaluated using NCD and MSE techniques. The method that we proposed offers better segmentation results than the two above mentioned. This algorithm can be used to simplify the image without the major loss of the image quality.