|
摘要：
|
| 由于工业CT图像结构复杂，存在各种伪影，并且灰度分布呈现区域性质等特点，难以准确找出分割阈值，为此提出了一种适用于工业CT图像的分割算法。首先利用最大类间方差法和图像处理方法处理了外层伪影，然后利用聚类迭代的方法处理中心空气，得到感兴趣的区域。实验结果表明，此算法能够对具有先验知识的工业断层图像准确地提取感兴趣的区域。 |
| Based on structure characteristics and gray distribution of industry CT image, there are different kinds of artifacts as well as grey with regional property in reconstructive tomography .It’s hard to find the segmentation threshold exactly .So a kind of segmentation method was designed for industry CT image .This method can make use of Ostu segmentation method and the method of image process in order to elimitate the outside artifacts, and the center air by approaching clustering . Finally, we can get result of region of interest .The result of algorithm experiment show: This method can pick up region of interest correctly for industry CT image with prior knowledge. |
|
参考文献：
|
[1] 章毓晋. 图像工程（中册）: 图像分析[M]. 北京: 清华大学出版社, 2005.
Zhang YJ. Image engineering (middle volume): Image analysis[M]. Beijing: Tsinghua University Press, 2005. |
|
服务与反馈：
|
|
【文章下载】【发表评论】【查看评论】【加入收藏】
|
| 提示：您还未登录，请登录！ |
|
