首页> 外文会议>15th world conference on non-destructive testing >Weld Defect Extraction and Classification in Radiographic Testing Based Artificial Neural Networks
【24h】

Weld Defect Extraction and Classification in Radiographic Testing Based Artificial Neural Networks

机译:基于人工神经网络的射线照相测试中焊缝缺陷的提取与分类

获取原文

摘要

This paper contains two parts: In the first part, we show the effectiveness of usingneural paradigms to detect edges in X-Ray images which are used in Non DestructiveTesting. The developed classifier consisted of a multilayer feed forward networkwindow in which the center pixel was classified using gray scale information withinthe window. The aim of the work in the second part, is to construct a set of welddefect descriptors in X-ray images and their recognition by the neural classifier.These descriptors are based on the geometric invariant moments which are insensitiveregarding usual geometrical transformations. Once the geometric invariant featurescomputed, a neural network classifier trained by back-propagation has to classify thedefect-images in planer or volumetric defect classes.
机译:本文包含两个部分:在第一部分中,我们展示了使用的有效性 神经范例以检测X射线图像中用于无损检测的边缘 测试。开发的分类器由多层前馈网络组成 在其中使用灰度信息对中心像素进行分类的窗口 窗户。第二部分的工作目的是构建一组焊缝 X射线图像中的缺陷描述符及其通过神经分类器的识别。 这些描述符基于不敏感的几何不变矩 关于通常的几何变换。一旦几何不变特征 计算后,经过反向传播训练的神经网络分类器必须对神经网络分类器进行分类。 平面或体积缺陷类别中的缺陷图像。

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