Real-time detection of cartridge quality based on X ray
-
摘要: 为了能实时而准确地检测药筒质量是否合格,通过无损检测的方法,首先获取药筒的X射线图片,然后对其采取一些预处理算法,以提高图像的信噪比,并对图片内部特征分别采取文中提出的局部平滑度的极小值算法和局部均值极小法,提取相应的特征参数,并将其与标准的特征参数比较,进而可以判断药筒质量是否合格。该算法具有快速、准确、高效的优点,在现场测试中取得了很好的实际应用效果。Abstract: In order to detect the quality of the cartridge precisely in real-time, it is needed to acquire the X ray photo of the cartridge by Non Destructive Testing(NDT) method, then some preprocess algorithms are used to enhance the signal to noise ratio (SNR) of the image. The local average minimum algorithm and the local smoothness minimum algorithm are proposed to extract the feature parameters of the cartridge, which are compared with the normal cartridge parameter, and an identification report about the cartridge quality will be given. The method is fast, precise and efficient.
-
Keywords:
- NDT /
- X-ray /
- local smoothness /
- local average /
- cartridge
-
-
[1]刘传才,杨静宇.一种新的图像纹理表示方法[J].计算机学报,2001,24(11):1302-1309.
LIU Chuan-cai,YANG Jing-yu.A new method for the image texture representation[J]. Chinese Journal of Computer,2001,24(11):1302-1309.(inChinese with an English abstract)
[2]胡小锋,赵辉.Visual C++/MATLAB图像处理与识剐使用案例精选[M].北京:人民邮电出版社,2004.
HU Xiao-feng,ZHAO hui.Visual C++/matlab image processing and recognition using cases[M]. Beijing :Posts&Telecom Press,2004.(in Chinese)
[3]容观澳.计算机图像处理[M].北京:清华大学出版社,2002.
RONG Guan-ao.Computer image processing[M]. Beijing:Tsinghua University Press,2002.(in Chinese)
[4]降雨志,张义顺.投影法和差影法在焊缝跟踪图像识别中的应用[J].沈阳工业大学学报,2005,27(5):571-573.
JIANG Yu-zhi,ZHANG Yi-shun. Application of both projection and difference image methods in welding seam recognition system[J]. Journal of ShengYang University of Technology,2005,27(5):571-573. (in Chiniese with an English abstract)
[5]杨淑莹.VC++图像处理设计[M].北京:清华大学出版社,2003.
YANG Shu-ying.The program design in image processing based on VC++[M].Beijing:The Press of Tsinghua University ,2003.(in Chinese)
[6]毛磊,孔凡让.基于图像处理与神经网络的钢丝绳表面缺陷检测[J].测控技术,2007,26(7):8-9.
MAO Lei,KONG Fan-rang. Detection of external defects in steel wire rope based on image processing and neural network[J]. Measurement & Contron Technology,2007,26(7):8-9.(in Chiniese with an English abstract)
计量
- 文章访问数: 2735
- HTML全文浏览量: 149
- PDF下载量: 1007