摘要
生瓷片生产过程中,需对冲孔加工后的生瓷片进行质量检测。为满足系统实时性与精度要求,本文从降低算法的复杂性和提高识别的准确率两个角度,优化了传统图像检测算法,采用可分、递归,实现改进的均值滤波器和行程编码法优化的区域特征提取,减少运算量和资源占用;并使用动态阈值分割、对前/后景采用不同定义的连通域提取、引入真圆度等方法,提高识别准确率。结果表明:完成一个包含1088个微孔的瓷片的检测时间为54ms。
In the process of producing raw porcelain chips,it is necessary to check the quality of raw porcelain chips after punching processing.In order to meet the real-time and precision requirements of the system,this paper optimizes the traditional image detection algorithm from the two perspectives of reducing the complexity of the algorithm and improving the recognition accuracy.It adopts divisible and recursive methods to realize improved mean filter and stroke coding method to optimize regional feature extraction,reducing the amount of computation and resource occupation.The recognition accuracy is improved by using dynamic threshold segmentation,connecting domain extraction with different definitions of front/back scene,and roundness.The results show that the detection time of a ceramic with 1088 microholes is 54ms.
作者
徐阳
罗福源
徐鹏
XU Yang;LUO Fuyuan;XU Peng
出处
《计量与测试技术》
2023年第11期72-76,81,共6页
Metrology & Measurement Technique
关键词
机器视觉
微小群孔
缺陷检测
阈值分割
特征提取
machine vision
multiple micro holes
defect detection
threshold segmentation
feature extraction