二维直方图阈值法分割工业CT图像的研究  

Segmentation of Industrial Computerized Tomography Image Based on Two-Dimensional Histogram Method

作　　者：向新程[1] 丛鹏[1] 陈景运[1] 

机构地区：[1]清华大学核能与新能源技术研究院,北京100084

出　　处：《原子能科学技术》2007年第3期347-350,共4页Atomic Energy Science and Technology

摘　　要：高精度工业CT图像因其包含物体三维信息而被广泛重视。尤其对于某些工件的缺陷检测,可利用图像分割技术进一步识别、突显缺陷,提高检测效率。针对一类工件的缺陷,对其高精度CT图像先后采用了比特窗算法进行预处理和二维直方图阈值法进行分割,突出显示了该类缺陷。该类方法对类似问题有广泛应用。The high resolution industrial CT image is useful for its three-dimensional information of the object. In the testing of the defect of workpiece, the testing efficiency by using image segmentation method to recognize and reveal the defect is enhanced. The bit-window method in the pre-process of the CT image and two-dimensional histogram method in the image segmentation were successively adopted and the defect of a inspected workpiece was revealed very well. Further more, these methods can solve similar problems.

关 键 词：比特窗 二维直方图阈值法 图像分割 

分 类 号：TL811[核科学技术—核技术及应用]