模糊连接算法在虚拟人图像分割上的应用

The Application of Fuzzy Connectedness Approach on the Segmentation Of Visuial Human Image

为实现对数字虚拟人彩色切片的分割,提出了一种基于竞争机制的相对模糊连接分割方法。首先计算所有像素与分割目标的模糊连接度,在分割结果出现过度分割时,则利用相对分割目标加入对感兴趣区域的竞争,自动重新确定各像素的连接度。采用虚拟中国人数据集的切片进行实验,结果表明,该算法能在待分割对象边界比较模糊的情况下取得理想的结果。

For segmening the slices images of the virtual human, a competition based relative fuzzy connectedness of multiple objects is proposed, The connectedness between the object and every pixel is computed through traditional function. When some area is over segmented, the method of relative fuzzy connected of multiple objects is effectively utilized, the connectedness of every pixel is recomputed. The experiments are carried out by using the slice images of virtual Chinese human No. 1 Dataset. The results demonstrate the effectiveness of the method , especially for those objects with blurred boundary.