引用本文:周继彦,余正泓.基于图像处理的包装印刷缺陷检测方法[J].包装工程,2017,38(9):240-244.
【打印本页】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 3466次   下载 3261 本文二维码信息
码上扫一扫!
分享到: 微信 更多
基于图像处理的包装印刷缺陷检测方法
周继彦, 余正泓
广东省科技干部学院,珠海 519090
摘要:
目的 为了提高包装印刷品缺陷检测的准确度和适用性,基于图像处理设计一种包装印刷缺陷检测方法。方法 采用几个关键步骤包括图像配准、配准区域自动选取、缺陷检测等,并根据包装印刷品图像特征选择配准区域,同时给出一种快速图像配准算法,利用改进差影匹配算法实现缺陷检测;基于DSP和FPGA设计控制系统硬件平台,主要包括控制单元、图像采集与处理单元、成像单元等,并进行实验研究。结果 所述方法能够准确识别出细微的刀丝、拉条类缺陷。结论 该方法具有较高的可靠性、通用性,可实现包装印刷品缺陷的快速检测。
关键词:  图像配准  缺陷检测  DSP+FPGA
DOI:
分类号:
基金项目:2016年广东省自然科学基金博士科研启动基金(2016A030310306);2016年珠海市先进装备制造与材料成形技术重点实验室重点基金(201601A);2016年广东省高等职业教育特色专业计划基金(2016GZPP055)
Package Printing Defect Detection Method Based on Image Processing
ZHOU Ji-yan, YU Zheng-hong
Guangdong Institute of Science and Technology, Zhuhai 519090, China
Abstract:
The work aims to improve the accuracy and applicability of package printing defect detection, and design a detection method based on image processing. Several key steps, including image registration, automatic selection of registration areas, and defect detection were adopted. The registration area was selected according to the image features of package printing matter. At the same time, a fast image registration algorithm was presented. Then, the defect detection was achieved on the basis of the improved subtraction matching algorithm. The hardware platform of control system was designed based on DSP and FPGA, mainly including control unit, image acquisition and processing unit, imaging unit and so on. Finally, the experimental study was carried out. Experimental results showed that the described method could accurately identify the subtle defect, such as knife filament and brace. The method has higher reliability and universality, which can realize the rapid detection of package printing defects.
Key words:  image registration  defect detection  DSP+FPGA

关于我们 | 联系我们 | 投诉建议 | 隐私保护

您是第24463933位访问者    渝ICP备15012534号-2

版权所有:《包装工程》编辑部 2014 All Rights Reserved

邮编:400039 电话:023—68792836传真:023—68792396 Email: designartj@126.com

    

  
 

渝公网安备 50010702501717号