摘要
研究了复杂背景图片中条码的定位及识别技术.通过特征提取、形态学的方法在低分辨率图像中对条码进行粗定位,以粗定位的结果为研究对象,采用边缘提取及灰度投影的方法对条码进行精确定位.提出了一种新的二值化算法,算法利用条码的固有特征得到图像的全局信息,利用全局与局部相结合的方法获取子图像块的最优阈值.实验结果表明,本文算法能有效地去除复杂背景对条码识别的影响,且算法具有较低的复杂度,能在嵌入式平台上实时运行.
Barcode localization and recognition technology was is explored in this paper. Feature extraction and math morphological methods are used to find the interesting regions in a down-sampled image first, and then edge detection and gray projection are used to locate the barcode accurately. A new binarization algorithm is proposed. A global threshold and mean pixel width of bar and space is calculated by the features of barcode, global and local information is used to get an optimal threshold of a subregion. The experiment result shows that the algorithm has a good per formance to eliminates the effect of complex background in the process of barcode recognition. At the same time, our algorithm has low complexity and can be used in embedded system real-time.
出处
《湖南工程学院学报（自然科学版）》
2012年第3期42-45,共4页
Journal of Hunan Institute of Engineering(Natural Science Edition)
基金
国家自然科学基金项目(60972037)
省部产学研项目(2009B090300267)
关键词
条形码
图像识别
特征提取
数学形态学
二值化
image recognition~ feature extractiom mathematical morphology~ binarization
