• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Annual Symposium on Quantitative Nondestructive Evaluation; 19980719-24; Snowbird,UT(US) >AUTOMATED INTERPRETATION FOR ULTRASONIC NDT DATA
【24h】

AUTOMATED INTERPRETATION FOR ULTRASONIC NDT DATA

机译：超声NDT数据的自动解释

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The expert system provides a powerful platform for containing the core NDT knowledge for classification and sizing of flaws. Introducing knowledge of specific NDT applications can expand the expert system and the Windows based dialog editing facility provides a platform for achieving this without interacting directly with the CLIPS coding. Three segmentation methods have been evaluated regarding their ability to extract regions of interest from a 3D ultrasonic image. The ability to do this successfully can have significant impact on how well the expert system can form correct flaw hypotheses. For the image considered within this paper using the average contrast measure to determine the threshold value from the co-occurrence matrix provided the best result, although overall the measures performed quite poorly. Use of the morphological operators on this image also performed comparatively well with the result from the average contrast measure, although this is an extremely user intensive technique and is not readily applicable to use within an automated system. Histogram thresholding was the segmentation method used to provide the suspect regions for input to the expert system. Although this method did oversegment the flaw into more than one region, rules within the expert system could compensate for this.
机译：专家系统提供了一个强大的平台，用于包含核心NDT知识，以便对缺陷进行分类和确定大小。引入有关特定NDT应用程序的知识可以扩展专家系统，并且基于Windows的对话框编辑工具提供了一个实现此平台的平台，而无需与CLIPS编码直接交互。关于从3D超声图像中提取感兴趣区域的能力，已经评估了三种分割方法。成功执行此操作的能力可能会对专家系统形成正确的缺陷假设的能力产生重大影响。对于本文中考虑的图像，使用平均对比度度量从同现矩阵确定阈值可提供最佳结果，尽管总体而言，这些度量的执行效果很差。从平均对比度测量的结果来看，在此图像上使用形态运算符的效果也相对较好，尽管这是一项非常耗费用户的技术，并且不易在自动化系统中使用。直方图阈值化是用于提供可疑区域以输入到专家系统的分割方法。尽管此方法确实将缺陷过度分割到多个区域，但是专家系统中的规则可以弥补这一缺陷。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号