基于计算机视觉的圆锯片图像特征提取  

Circular Saw Blade Image Feature Extraction Based on Computer Vision

作　　者：赵玲[1] 卜范骞[1] 

机构地区：[1]莱芜职业技术学院智能制造系,山东济南271100

出　　处：《工业控制计算机》2024年第10期119-120,131,共3页Industrial Control Computer

摘　　要：针对圆锯片图像特征,将图像的预处理、轮廓提取、轮廓优化、锯片圆心的检测及内外圆直径、圆度误差的测量、齿尖及齿尖两段直线的检测、齿尖角的测量、径向前角及径向后角的测量算法集成。根据所研究的圆锯片图像齿尖特征,改进了直线的Hough变换,即首先将各个齿尖位置跟齿尖周围共线的像素点检测到,然后再用Hough变换逐次检测齿尖周围的两段直线。针对圆的常规Hough变换检测时间长的问题,改进了圆的Hough变换,即首先将圆上边缘像素点总数三等分,然后从每一部分边缘点各取一个像素点,利用不共线的三个像素点确定圆心。这样在保证精度的前提下,大大缩短了检测时间,为零件的在线检测奠定了基础。According to the image features of the circular,the image preprocessing,contour extraction,contour optimization,detection of the center of the saw blade,measurement of the diameter of the inner and outer circles,measurement of roundness error,detection of the tip and two straight lines of the tooth tip,measurement of the tip angle,and measurement algorithms of radial rake angle and radial back angle are integrated in this paper.According to the characteristics of the tooth tip of the circular saw blade image,the Hough transform of the straight line is improved,the collinear pixel points around the tooth tip at each tip are detected first,and then the two straight lines around the tooth tip were detected by the Hough transform one by one.In order to solve the long time of the detection of the conventional Hough transform of the circle,the Hough transform of the circle is improved.The total number of pixels at the edge of the circle is divided into three equal parts,and then one pixel is taken from each part of the edge point,and the three pixels that are not collinear are used to determine the center of the circle.This greatly shortens the detection time on the premise of ensuring accuracy and lays the foundation for the on-line inspection of parts.

关 键 词：特征 HOUGH变换 在线检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TG717[自动化与计算机技术—计算机科学与技术]