数据与计算发展前沿 ›› 2020, Vol. 2 ›› Issue (4): 132-141.

doi: 10.11871/jfdc.issn.2096-742X.2020.04.011

所属专题： 下一代互联网络技术与应用

• 技术与应用 • 上一篇    下一篇

一种基于特征点的卷烟商标纸配准方法

王凯华1,2(),李晓辉3(),周明珠3(),罗泽1(),邢军3,*()   

  1. 1.中国科学院计算机网络信息中心,北京 100190
    2.中国科学院大学,北京 100049
    3.国家烟草质量监督检验中心,河南 郑州 450001
  • 收稿日期:2020-03-17 出版日期:2020-08-20 发布日期:2020-09-10
  • 通讯作者: 邢军
  • 作者简介:王凯华,中国科学院计算机网络信息中心,中国科学院大学,硕士研究生,主要研究方向为智能图像和视频分析。
    本文中负责实验设计和实现。
    Wang Kaihua is a master student at Computer Network Information Center of the Chinese Academy of Sciences / University of the Chinese Academy of Sciences. The main research directions are intelligent image and video analysis.
    In this paper, he is responsible for experimental design and implementation.
    E-mail: wangkaihua@cnic.cn|李晓辉,国家烟草质量监督检验中心,高级工程师,主要研究方向为烟草物理性能检测技术与标准研究
    本文中负责流程设计和应用分析。
    Li Xiaohui is a senior engineer at China National Tobacco Quality Supervision & Test Center. Her research direction is on testing technologies and standards of tobacco physical properties.
    In this paper, she is responsible for the process design and application analysis.
    E-mail: lixh@ztri.com.cn|周明珠,国家烟草质量监督检验中心,高级工程师,主要研究方向为烟草物理性能检测技术与标准研究。
    本文中负责方法介绍。
    Zhou Mingzhu is a senior engineer at China National Tobacco Quality Supervision & Test Center. Her research direction is on testing technologies and standards of tobacco physical properties.
    In this paper, she is responsible for the method introduction.
    E-mail: zhoumz@ztri.com.cn|罗泽,中国科学院计算机网络信息中心,研究员,博士生导师,主要研究方向为海量数据分布处理理论和方法,数据挖掘和机器学习理论、方法和应用。
    本文中负责结论展望。
    Luo Ze, Ph.D. Supervisor, is a research fellow at Computer Network Information Center of the Chinese Academy of Sciences. His research directions are massive data distribution processing theory and method, data mining and machine learning theory, method and application.
    In this paper, he is responsible for the conclusion outlook.
    E-mail: luoze@cnic.cn|邢军,国家烟草质量监督检验中心,研究员,主要研究方向为检测技术与标准研究。
    本文中负责现状分析。
    Xing Jun is a research fellow at China National Tobacco Quality Supervision & Test Center. Her research direction is research on testing technologies and standards.
    In this paper, she is responsible for the current studying status analysis.
    E-mail: xingj@ztri.com.cn
  • 基金资助:
    中国烟草总公司科技重大专项([110201901026(SJ-05)])

A Method of Cigarette Wrapping Paper Image Registration Based on Feature Points

Wang Kaihua1,2(),Li Xiaohui3(),Zhou Mingzhu3(),Luo Ze1(),Xing Jun3,*()   

  1. 1. Computer Network Information Center, Chinese Academy of Sciences, Beijing 100190, China
    2. University of Chinese Academy of Sciences, Beijing 100049, China
    3. China National Tobacco Quality Supervision & Test Center, Zhengzhou, Henan 450001, China;
  • Received:2020-03-17 Online:2020-08-20 Published:2020-09-10
  • Contact: Xing Jun

RichHTML

6

PDF

282

摘要：

【目的】利用改进的尺度不变特征变换（Scale-Invariant Feature Transform, SIFT）算法提取的匹配特征对卷烟商标纸图像进行细粒度配准,达到提升配准精度和区分真伪卷烟商标纸图像的目的。【方法】通过对图像分块处理、剔除不稳健特征点、单应性矩阵粗配准后根据匹配点距离进行约束筛选匹配对,并提出根据细粒度配准后的匹配点距离均值进行评价,最终实现并改进了基于特征点的卷烟商标纸细粒度图像配准方法。【结果】基于本文改进的特征点检测方法可以提取到更均衡的特征点,提高推定匹配率,提出的配准结果评估标准能有效评估配准质量,粗配准筛选匹配点可以提高图像细粒度配准的精度,并可以对卷烟商标纸图像进行区分。【局限】目前的改进集中在匹配对的筛选,在细粒度配准方法研究上仍有改进的空间。【结论】基于改进的SIFT算法提取的特征点,提出了先粗配准后细配准的图像细粒度配准策略,经实验证明此策略可以提升图像配准精度,并可以达到区分卷烟商标纸图像的目的。

关键词: SIFT, 特征检测, 特征点匹配, 单应性变换, 图像配准

Abstract:

[Objective] In order to improve the registration accuracy and authenticate the wrapping paper images, fine-grained registration is performed on the cigarette packaging images by using the matching feature points extracted from the optimized SIFT (Scale-invariant feature transform) algorithm. [Methods] After performing block processing on the images, removing unstable feature points, and using coarse registration of the homography matrix to filtrate paired points by distance constraint, an evaluation approach is proposed based on average distance betwteen fine-grained matching pairs to improve the registration performance based on SIFT features. [Results] The experimental results show that the improved feature point extraction method in this paper can extract more balanced feature points and improve the estimated matching rate. The proposed registration evaluation standard can effectively evaluate the registration quality, and the coarse registered matching points can improve the accuracy of fine-grained registration of the image and authenticate the wrapping paper images. [Limitations] The current improvement is focused on the selection of matching pairs and there is still room for improvement in the research of fine-grained registration methods. [Conclusions] Experiments prove that this strategy can improve registration accuracy and achieve the purpose of authenticating cigarette wrapping paper images.

Key words: SIFT, feature detection, feature matching, homography transformation, image registration