An Improved Region Growing Algorithm and its Application of Foreign Substances Inspection
-
摘要: 提出一种基于区域生长的图像分割算法, 并达到异物识别的目的。该算法根据不同X射线图像直方图的特征, 自动选取出所需要的种子点, 并且通过对已进行过区域生长的部分进行概率统计, 得出合适的阈值, 进行种子区域生长。相对于一般的根据每幅图像手动选取种子点以及阈值的种子区域生长法(SRG), 该算法能够快速有效地进行图像分割, 异物识别率高。Abstract: An image segmentation algorithm based on automated seeded region growing was advanced to achieve the aim of foreign substances inspection. According to the feature of various X-ray image’s histogram, this algorithm automatically selected required seeds, and through the relative statistical characteristics it acquired the proper threshold value. Comparing with ordinary manual seeded and threshold value (SRG), this algorithm could segment images more effectively and achieve better effect.
-
Keywords:
- Region growing /
- Automated seeded /
- Histogram /
- Statistical characteristics /
- Image processing
-
-
[1] Rolf Adams, Leanne Bischof. Seeded region growing[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1994, 16(6): 641-647. [2] Weihong Cui, Zequn Guan. An improved region growing algorithm for image segmentation[J]. IEEE computer society, 2008 International Conference on Computer Science and Software Engineering, [3] 冈萨雷斯, 伍兹.数字图像处理[M].北京: 电子工业出版社, 2004. [4] 张晓梅, 王诚梅, 韩琥.基于区域的烟尘图像分割方法[J].计算机工程与应用, 2008, 44(13): 193-195. [5] 秦襄培.Matlab图像处理与界面编程[M].北京: 电子工业出版社, 2009.
计量
- 文章访问数: 0
- HTML全文浏览量: 0
- PDF下载量: 0