电子路牌显示质量缺陷的快速准确识别方法

全文下载： 202102013.pdf

文章编号： 1672-6987（2021）02-0090-07； DOI： 10.16351/j.1672-6987.2021.02.013

曹梦龙， 靳利文(青岛科技大学 自动化与电子工程学院，山东 青岛 266061)

摘要： 针对电子路牌显示质量缺陷自动光学检测时，遮阳板反光干扰与目标区域低对比度难以分割的问题，提出一种基于面积约束的阈值分割新方法。构造利用像素点频域和时域信息的双边滤波方式，对原始图像进行处理；在直方图阈值分割法基础上引入面积约束条件，并根据先验信息限制阈值取值范围；最后通过图像矩实现对缺陷目标的识别。实验结果表明：该方法成功实现了对电子路牌显示质量缺陷的实时、快速、准确检测，为电子路牌缺陷检测提供了一种有效的解决方法。

关键词： 电子路牌； 双边滤波； 阈值分割； 面积约束

中图分类号： TP 391文献标志码： A

引用格式： 曹梦龙， 靳利文. 电子路牌显示质量缺陷的快速准确识别方法［J］. 青岛科技大学学报(自然科学版), 2021, 42(2): 9096.

CAO Menglong, JIN Liwen. Rapid and accurate identification method of defects in electronic road signs［J］. Journal of Qingdao University of Science and Technology（Natural Science Edition), 2021， 42(2)： 9096.

Rapid and Accurate Identification Method of Defects in Electronic Road SignsCAO Menglong, JIN Liwen

(College of Automation and Electronic Engineering, Qingdao University of Science and Technology, Qingdao 266061, China)

Abstract: Aiming at the problem of automatic optical detection of electronic street sign display quality defects, it is difficult to separate the reflection interference of the sun visor and the low contrast of the target area, This paper proposed a new threshold segmentation method based on area constraints. This paper constructed a bilateral filtering method using pixel frequency and time domain information to process the original image. This paper introduced an area constraint condition based on the histogram threshold segmentation method, and limit the threshold value range based on prior information. Finally this paper realized defective targets using image moments. The experimental results show that the method successfully realizes the realtime, fast and accurate detection of electronic street sign display quality defects, and provides an effective solution for the detection of electronic street sign defects.

Key words: electronic road signs; bilateral filtering; threshold segmentation; area constraint

收稿日期： 20200428

基金项目： 国家重点研发计划项目（2019YFA07065023）.

作者简介： 曹梦龙（1971—），男，副教授.