基于图像匹配的透明车灯部件检测

doi:10.3969/j.issn.1000-1158.2018.03.13

摘要
图/表
参考文献(12)
相关文章 (15)

全文: PDF (392 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对车灯总成的完整性检测,提出了基于图像金字塔和归一化互相关函数结合的分层匹配算法。利用直方图均衡化和锐化滤波增强图像对比度以及边缘细节信息,再运用归一化互相关相似度函数实现车灯透明部件的匹配,同时采用金字塔分层来提高图像匹配的速度。通过实验确定了合适的匹配窗口大小、金字塔层数、相似度阈值等,实现了车灯透明部件的安装检测。实验结果表明:该方法对车灯透明部件的成功检测率可达到约95%,检测时间约50 ms,具有较高的准确率和实时性。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	朱婉仪
	穆平安
	戴曙光

关键词 ：计量学, 车灯部件检测, 图像匹配, 直方图均衡化, 归一化互相关, 金字塔分层

Abstract：To satisfy the detection of transparent lamp parts in the car lights assembly,a hierarchical matching algorithm based on the image pyramid was proposed combining with the normalized cross correlation method. Histogram equalization and sharpening filtering are used to improve the gray contrast and to enhance the edge of the image. Then the image matching is realized by the normalized cross correlation similarity function, and through pyramid algorithm to improve the speed of image matching. The appropriate matching window size, pyramid layer number and similarity threshold is determined to achieve the installation of transparent parts of the lamp detection through the experiment. It has been proved that the detection rate of the transparent part can reach 95% and the average detection time is 50ms, which has high accuracy and speed.

Key words： metrology lamp parts inspection image matching histogram equalization normalized cross-correlation pyramid delaminating

收稿日期: 2017-04-13 发布日期: 2018-04-12

PACS:

TB96

通讯作者: 朱婉仪 E-mail: 18818224608@163.com

作者简介: 朱婉仪（1994-）,女,湖南衡阳人,上海理工大学硕士研究生,主要研究方向为机器视觉测试技术。18818224608@163.com

引用本文:

朱婉仪,穆平安,戴曙光. 基于图像匹配的透明车灯部件检测[J]. 计量学报, 2018, 39(3): 353-358.
ZHU Wan-yi,MU Ping-an,DAI Shu-guang. Detection of Transparent Lamp Parts Based on Image Matching. Acta Metrologica Sinica, 2018, 39(3): 353-358.

链接本文:

http://jlxb.china-csm.org:81/Jwk_jlxb/CN/10.3969/j.issn.1000-1158.2018.03.13 或 http://jlxb.china-csm.org:81/Jwk_jlxb/CN/Y2018/V39/I3/353