炭素制品缺陷的X射线自动检测技术研究

周 贤; 刘义伦

doi:null

您当前的位置：

首页 >

文章列表页 >

炭素制品缺陷的X射线自动检测技术研究

更新时间：2020-08-12

- 炭素制品缺陷的X射线自动检测技术研究
- X-ray automatic inspection techniques for carbon product defects
- 光学精密工程 2006年14卷第3期页码：503-508
- 作者机构：
  
  中南大学机电工程学院,湖南长沙,410083
- 作者简介：
- 基金信息：
- DOI：
  中图分类号： TP391
- 收稿日期：2005-10-22，
  
  修回日期：2006-05-14，
  
  网络出版日期：2006-06-30，
  
  纸质出版日期：2006-06-30
- 稿件说明：
移动端阅览
周贤, 刘义伦. 炭素制品缺陷的X射线自动检测技术研究[J]. 光学精密工程, 2006,14(3):503-508.

ZHOU Xian, LIU Yi-lun. X-ray automatic inspection techniques for carbon product defects[J]. Optics and precision engineering, 2006, 14(3): 503-508.
周贤, 刘义伦. 炭素制品缺陷的X射线自动检测技术研究[J]. 光学精密工程, 2006,14(3):503-508. DOI：

ZHOU Xian, LIU Yi-lun. X-ray automatic inspection techniques for carbon product defects[J]. Optics and precision engineering, 2006, 14(3): 503-508. DOI：

摘要

针对炭素制品X光图像的特点

对其缺陷的提取与识别技术进行了研究

给出了目标边界提取算法和基于小波变换的图像增强算法

实现了图像的背景去除及增强处理。在此基础上

为排除噪声干扰的影响

采用数学形态学和迭代阈值分割相结合的方法从背景去除后的图像中提取出缺陷区域

取得了良好的效果。对缺陷特征选择及识别方法进行了研究

设计了基于遗传策略的特征选择和基于BP神经网络的缺陷识别算法

计算表明:缺陷正确识别率可达95%以上。采用上述技术开发完成了一套炭素制品缺陷X射线自动检测系统。

Abstract

Based on the characteristic of X-ray detection images of carbon product

the defect extraction and recognition techniques were present to remove background and enhance image successfully with target boundary extraction algorithm and image enhancement algorithm on wavelet transform. In order to eliminate the effect of noise

a mathematical morphology combining with iteration threshold segmentation method was adopted to extract defect from a image without background

it was proved to be very effective. An algorithm based on genetic strategy and BP neural network is proposed for the selection and recognition of defects

the results show that the veracity of defects recognition is 95%. These techniques are then combined into an automatic inspection system for carbon material.

关键词

Keywords

references

. 周贤,刘义伦,龚海飞. 炭素材料内部缺陷检测方法的探讨[J]. 无损检测,2005,27(3):132-134. ZHOU X,LIU Y L,GONG H F. Discussion on detection methods of carbon material flaws[J]. NDT, 2005,27(3):132-134.(in Chinese)

. 刘恒殊,黄廉卿. 基于模糊集理论的CR图像处理方法[J]. 光学精密工程,2002,10(2):188-192. LIU H SH,HUANG L Q. Processing method of CR image based on fuzzy set theory[J]. Optics and Precision Engineering, 2002,10(2):188-192.(in Chinese)

. TILLACK G R,NOCKEMANM C,BELLON C. X-ray modeling for industrial applications[J]. NDT and E International, 2000, 37: 481-488.

. FERNANDES B T,SILVA I,GAYDECKI P A. Vector extraction from digital images of steel bars produced by an inductive scanning system using a differential gradient method combined with a modified Hough transform[J]. NDT and E International, 2000, 33: 69-75.

. HALM S R. Radiographic interpretation[J]. British Journal of NDT, 1992, 34(5):234-238.

. BERGER H. 100 years of X-rays-industrial use of NDT and the role of ASNT[J]. Materials Evaluation, 1995, 53(11): 1253-1260.

. 李平,陈锐. 石墨制品的超声波无损检测技术[J]. 炭素,1997,(4):32-36. LI P, CHEN R. Ultrasonic nondestructive testing technique for graphite products[J]. Carbon. 1997(4):32-36.(in Chinese)

. 任大海,尤政,孙长库,等. 焊接缺陷实时成像自动检测技术的研究[J]. 清华大学学报,2001,41(2):25-29. REN D H,YOU ZH,SUN CH K,et al. Automatic inspection techniques by real-time radiography for weld defects[J]. J Tsinghua Univ, 2001,41(2):25-29.(in Chinese)

. 张雷,杨润魁,黄廉卿. 数字X光医学影像压缩技术[J]. 光学精密工程,2004,12(6):632-637. ZHANG L,YANG R K,HUANG L Q. Computed X radiography image compression[J]. Optics and Precision Engineering, 2004,12(6):632-637.(in Chinese)

. FLETCHER M J, SPEAKE J H. Trends in automated non-destructive testing weld and metal fabrication[J]. 1989,57(6): 239-242.

. JAIN A K, DUBUISSON M P. Segmentation of X-ray and c-scan images of fiber reinforced composite materials[J]. Pattern Recognition, 1992, 25(3): 257-270.

. STRICKLAND R K, HAHN H. Wavelet transforms methods for objects detection and recovery[J]. IEEE Trans Image Process, 1997,6(5): 724-735.

. KAZANTSEV I G, LEMAHIEU I. Reconstruction of elongated structures using ridge functions and natural pixels[J]. Inverse Problem, 2000,16(2): 505-517.

浏览量

365

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

DRT Net：面向特征增强的双残差Res-Transformer肺炎识别模型

基于Anscombe变换的X射线图像序列盲源分离降噪

微焦点X射线图像乘性加性混合噪声的去除