当前位置>主页 > 期刊在线 > 计算机技术 >

计算机技术2020年22期

应用于方阵式DM 码定位的图像分割技术研究
康三顺
(北京瀚远医药科技有限公司,北京 100176)

摘  要:文章所研究的图像分割过程主要分为预处理、粗定位、精定位三部分;预处理的目的主要是减少图像信息量和提升图像质量;粗定位主要分图像分块、候选区域标记、区域生长;候选区域标记通过Scharr 边缘检测算子计算DM 码梯度方向和梯度幅值,通过梯度方向和梯度幅值进行统计直方图计算,确定候选区域;区域生成将在候选区域进行种子生长,完成粗定位;精定位部分结合由外向里,由里向外进行扫描完成DM 更精确的定位。


关键词:二维码;DM 码;图像分割



中图分类号:TP391.41         文献标识码:A         文章编号:2096-4706(2020)22-0091-04


Research on Image Segmentation Technology Applied to Square Matrix DM Code Location

KANG Sanshun

(Beijing Hanyuan Pharmaceutical Technology Co.,Ltd.,Beijing 100176,China)

Abstract:The image segmentation process studied in this paper is mainly divided into three parts:preprocessing,coarse positioning and fine positioning;the main purpose of preprocessing is to reduce the amount of image information and improve the image quality;the coarse positioning is mainly divided into image segmentation,candidate region labeling and region growing;the candidate region labeling calculated the DM code gradient direction and gradient amplitude by Scharr edge detection operator,and the candidate region is determined by statistical histogram calculation of gradient direction and gradient amplitude;in the region generation,seed growth will be carried out in the candidate region to complete the coarse positioning;in the fine positioning part,scanning from the outside in and from the inside out will be combined to complete the more accurate positioning of DM.

Keywords:QR code;DM code;image segmentation


参考文献:

[1] 张宁. 基于摄像方式的二维条码识别算法的研究 [D]. 南京:南京理工大学,2013.

[2] 于英政.QR 二维码相关技术的研究 [D]. 北京:北京交通大学,2014.

[3] 刘美,薛新松,刘广文,等. 对比度增强的彩色图像灰度化算法 [J]. 长春理工大学学报(自然科学版),2018,41(5):70-74.

[4] 王娟,王萍. 直接零件标志条码区域定位算法 [J]. 计算机辅助设计与图形学学报,2014,26(7):1159-1166.

[5] 王伟,何卫平,雷蕾,等. 污染及多视角下DataMatrix码精确定位 [J]. 计算机辅助设计与图形学学报,2013,25(9):1345-1353.

[6] ZHANG C N,MA L,MAO D. A 2D Barcode RecognitionSystem Based on Image Processing [C]//The 2011 InternationalConference on Electric and Electronics(EEIC 2011).Springer,2011:670-675.


作者简介:康三顺(1986.02—),男,汉族,甘肃天水人,高级算法工程师,本科,研究方向:图像识别、人工智能、计算机应用。