基于直线检测的条码图像几何校正研究  

Research on Geometric Correction of Bar Code Image Based on Line Detection

作　　者：庞惠文 张增红 李梅[2] 肖圆圆[3] 

机构地区：[1]山西轻工职业技术学院轻工工程系,山西太原030013 [2]运城学院机电工程系,山西运城044000 [3]绵阳职业技术学院艺术系,四川绵阳621000

出　　处：《包装学报》2016年第2期57-61,共5页Packaging Journal

基　　金：中央财政支持高等职业学校提升专业服务产业能力基金资助项目(教职成厅函[2011]71号)

摘　　要：CCD相机采集的图像会产生一定程度的几何变形,需要进行几何校正。常用的图像几何校正方法不适合条码图像的校正,因此,结合已有的边缘检测算法和直线提取算法,提出了基于直线检测的条码图像的几何校正算法。基于直线检测的条码图像校正方法是在已除去背景的图像中,寻找目标的条形码边缘对应的直线,以此确定其偏移角度,并进行扭转校正。实验结果表明,有偏转角度的条码图像运用了所提方法校正后,都得到了有效的几何校正。Geometry deformation occurred to images taken by CCD camera needs geometric correction. Geometric correction algorithm of bar code image based on line detection is proposed combined with edge detection algorithm and line extraction algorithm since common image geometric correction methods do not apply to the correction of barcode images.Image correction based on line detection is to detect the corresponding straight line at the edge of targeted barcode in the image without background, aiming to determine the deviation angle, and then proceed with the torsion correction. The experimental results show that the bar code images with deflection angle have been effectively corrected by the proposed method.

关 键 词：印刷技术 条码图像 几何校正 边缘检测 直线检测 

分 类 号：TS801.8[轻工技术与工程]