论文部分内容阅读
提出一种基于区域生长的CT序列图像的分割算法。在第一张待分割目标区域中选取一个种子点,利用四领域的生长规则对种子点进行区域生长,得到一组点集,将这个点集投影到下一张CT图像中,得到一组新的点集,再提取该点集的轮廓,最后对该轮廓上的点进行四领域的区域生长,分割出最终的目标区域。实验结果表明,该分割算法不仅适用于简单的图像分割问题,而且对于背景复杂、光照不均匀的图像也能取得较好的分割效果。
A segmentation algorithm of CT sequence image based on region growth is proposed. Select a seed point in the first target area to be segmented, and use the growth rule of the four fields to grow the seed points regionally, get a set of point sets, and project this point set to the next CT image to get a new set of new Point set, and then extract the outline of the point set, and finally on the outline of the four-point area of the regional growth, segmentation of the final target area. Experimental results show that the proposed segmentation algorithm is not only suitable for simple image segmentation, but also achieves good segmentation results for complex background images with uneven illumination.