当前位置>主页 > 期刊在线 > 智能制造 >

智能制造21年7期

基于 HALCON 视觉系统的遗传算法特征参数选取方法
郭志龙,赵辉
(兰州石化职业技术学院,甘肃 兰州 730060)

摘  要:针对不同构建取向成型的 3D 打印试验件,将其作为图像识别及检测的本体,对比 3D 打印模型参数,通过HALCON 视觉系统进行图像灰度化、图像增强及 ROI 选取等前期图像预处理,再针对不同类型的缺陷检测特征,采用基于遗传算法的特征参数选取方法。选择最优的亮度、对比度、伽马、锐度及照明等 HALCON 视觉系统参数并进行匹配检测,同时对该遗传算法进行改进处理,以验证改进后的匹配程度,取得了良好的产品缺陷检测效果。


关键词:3D 打印;HALCON 软件;视觉系统参数;预处理;遗传算法;最优;匹配度



DOI:10.19850/j.cnki.2096-4706.2021.07.041


课题项目:2018 年度教育科学“十三五” 规划高校与职业院校一般课题(GS〔2018〕GHB GZ050)


中图分类号:TP273.4                                    文献标识码: A                                    文章编号:2096-4706(2021)07-0158-04


Feature Parameter Selection Method of Genetic Algorithm Based on HALCON Vision System

GUO Zhilong,ZHAO Hui

(Lanzhou Petrochemical Polytechnic,Lanzhou 730060,China)

Abstract:Aiming at the 3D printing test pieces formed in different construction orientations,taking them as the ontology of image recognition and detection,compare the 3D printing model parameters,carry out early image preprocessing such as image graying,image enhancement and ROI selection through HALCON vision system,and then adopt the feature parameter selection method based on genetic algorithm for different types of defect detection features. Select HALCON vision system parameters such as the optimal brightness, contrast,gamma,sharpness and illumination for matching detection. At the same time,improve the genetic algorithm to verify the improved matching degree and achieve good product defect detection effect.

Keywords:3D printing;HALCON software;vision system parameter;preprocessing;genetic algorithm;optimal;matching degree


参考文献:

[1] KUANG Z P,DAI G M,WAN R J,et al. Osteogenic and antibacterial dual functions of a novel levofloxacin loaded mesoporous silica microspheres/nano-hydroxyapatite/polyurethane composite scaffold [J].Genes & Diseases,2019(2):193-202.

[2] 徐影,吕毅,王建波,等 . 深冷处理对 3D 打印 ABS 高分 子材料组织性能的影响 [J]. 矿冶工程,2020,40(4):142-144.

[3] 汪成龙,黄余凤,陈铭泉,等 . 基于 halcon 的字符缺陷 检测系统 [J]. 制造业自动化,2018,40(9):38-41.

[4] 舒奇,黄家才 . 基于 Halcon 的机器人手眼标定方法研究 [J]. 南京工程学院学报(自然科学版),2019,17(1):45-49.

[5] 田春林,陈李博,马国庆,等 . 基于 Halcon 的工业机器 人手眼标定方法研究 [J]. 制造业自动化,2018,40(3):16-18+ 46

[6] ZAVRTANIK V,KRISTAN M,SKČAJ D. Reconstruction by inpainting for visual anomaly detection [J].Pattern Recognition, 2020,112(2):107706.

[7] LIU C,ZHOU A M,ZHANG G X. Automatic clustering method based on evolutionary optimization [J].IET Computer Vision, 2013,7(4):258-271.

[8] CHAOUCH I,DRISS O B,GHEDIRA K. Applied Intelligence;Studies from University of Manouba Provide New Data on Applied Intelligence(A novel dynamic assignment rule for the distributed job shop scheduling problem using a hybrid ant-based algorithm) [J].Journal of Robotics & Machine Learning,2018(49): 1903-1924

[9] KRUSE J,CONNOR A M,MARKS S.An Interactive MultiAgent System for Game Design [J].The Computer Games Journal,2021 (10):41-63.


作者简介:郭志龙(1989.07—),男,汉族,甘肃天水人, 工程师,硕士研究生,研究方向:多智能体协作控制、设备故障检 测与诊断;赵辉(1990.09—),女,汉族,甘肃金昌人,助教, 硕士研究生,研究方向:汽车智能化制造、物联网系统及新能源汽 车应用技术。