实际边缘轮廓下的螺栓综合尺寸质量检测  被引量:6

Quality Inspection of Overall Bolt Dimension with Actual Edge Profile

在线阅读下载全文

作  者:孔明[1] 徐志玲[1] 徐勇 王琨 刘子豪 KONG Ming;XU Zhi-ling;XU Yong;WANG Kun;LIU Zi-hao(China Jiliang University,Hangzhou,Zhejiang 310018,China;Taizhou Fangyuan Quality Inspection Corporation,Taizhou,Zhejiang 318000,China;Aozhan Industrial Corporation,Hangzhou,Zhejiang 311221,China;Jiaxing University,Jiaxing,Zhejiang 314033,China)

机构地区:[1]中国计量大学,浙江杭州310018 [2]台州方圆质检有限公司,浙江台州318000 [3]奥展实业有限公司,浙江杭州311221 [4]嘉兴学院,浙江嘉兴314033

出  处:《计量学报》2021年第6期724-730,共7页Acta Metrologica Sinica

基  金:浙江省公益技术研究计划(LGG20FD30006)。

摘  要:为了检测螺栓的尺寸质量,提出一种基于实际边缘轮廓的螺栓综合尺寸质量检测算法。该算法首先根据螺栓图纸公差要求,设计螺栓综合尺寸公差带图作为模板图;然后对螺栓图像进行预处理,包括三段线性灰度增强、中值滤波和最大类间方差法等;进一步将螺栓图像旋转为水平位姿并进行canny边缘检测,将螺栓图像和模板图裁剪,得到具有相同位姿、尺寸大小和白色边缘轮廓的二值图像;最后将处理好的螺栓图像和模板图做图像加法运算,比较模板图与经过运算后图像的白色像素点个数,判别螺栓的综合尺寸是否合格。进一步地提出了一种离散序列图检测方法,实现了螺栓全周的综合尺寸质量检测。In order to detect the dimensional quality of bolts,a comprehensive quality inspection algorithm with the actual edge profile was proposed.The algorithm starts with designing tolerance zone diagrams for bolt dimension as template diagrams according to the tolerance requirements on bolt drawing.Then performing preprocessing to bolt images,including three-piecewise linear grayscale enhancement,median filtering,and maximum inter-class variance,etc.Further rotating the bolt images to a horizontal position and performing canny edge detection,cropping the bolt images and the template images to obtain a binary image with the same position,size and white edge contour.Finally,carrying out image addition operation to the processed bolt images and template images.Whether the overall size of the bolt is qualified can be determined by comparing the number of white pixels of the template images and the images after image operation.Furthermore,a discrete sequence diagram detection method was proposed to realize the comprehensive dimensional quality detection of bolts all around.

关 键 词:计量学 螺栓尺寸 边缘轮廓 图像处理 尺寸公差 

分 类 号:TB92[一般工业技术—计量学]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象