基于模式识别技术的条形码识别方法研究及应用

发布时间：2018-12-23 20:20

【摘要】：目前油田的物资管理工作,基本依靠物资管理系统,而物资管理系统中最重要的环节就是条形码扫描功能。由于天气以及油田现场环境等因素,严重影响了条形码扫描的速度以及准确率,导致物资管理工作效率低下。尤其在工具量较大的情况下,经常会严重影响工具的流通,从而造成企业的损失。为此本论文基于模式识别技术改进条形码识别技术,分析并解决了条形码图像由于倾斜、噪声、脏污而不能识别的问题。经仿真与实际测试,该设计可以有效改进条形码识别技术,使油田物资管理工作更加高效化、自动化。本文的主要研究内容包括:(1)基于Hough变换的图像倾斜校正:针对扫描到的倾斜条形码图像,本文采用基于Hough变换的图像倾斜校正方法,可以解决由于图像倾斜而导致条形码不能识别的问题。(2)基于支持向量机改进中值滤波算法:针对条形码图像的噪声问题,本文首先采用中值滤波算法处理条形码图像。通过采用中值滤波算法处理条形码图像,可以去除条形码图像上面的黑白孤立噪声点。然后针对中值滤波算法的优缺点、支持向量机的图像滤波去噪算法优缺点以及椒盐噪声的特征,本文采取基于支持向量机的方法需要改进中值滤波算法。仿真实验证明本文提出的算法可以更有效地消除椒盐噪声,并且处理结果具有较高的SNR。(3)基于自适应阈值选取的Ostu算法:本文首先采用Ostu算法以及Bernsen算法处理图像。根据油田现场的实际情况,以及比较这两种算法的处理效果与运行时间,本文选用Ostu算法处理条形码图像。然后针对Ostu算法的运行时间不能满足油田现场物资管理工作所需,对Ostu算法进行改进。本文采用自适应阈值选取的方法改进Ostu算法,改进效果分析表明改进后的算法可以加快处理时间,提高工作效率。
[Abstract]:At present, the material management work of oil field basically depends on the material management system, and the most important link in the material management system is the bar code scanning function. The speed and accuracy of barcode scanning are seriously affected by the weather and oil field environment, which leads to the inefficiency of material management. Especially in the case of a large number of tools, often seriously affect the circulation of tools, resulting in the loss of enterprises. This paper improves barcode recognition technology based on pattern recognition technology and analyzes and solves the problem that barcode image can not be recognized because of tilt noise and dirty pollution. Through simulation and practical test, the design can effectively improve the barcode recognition technology and make the oil field material management more efficient and automatic. The main contents of this paper are as follows: (1) Image tilt correction based on Hough transform: for the scanned skew bar code image, this paper adopts the image tilt correction method based on Hough transform. It can solve the problem that barcode can not be recognized because of image tilt. (2) improved median filter algorithm based on support vector machine: for the noise problem of barcode image, this paper first uses median filter algorithm to deal with barcode image. By using median filtering algorithm to process barcode images, the black and white isolated noise points on barcode images can be removed. Then, aiming at the advantages and disadvantages of median filtering algorithm, the advantages and disadvantages of SVM image filter denoising algorithm and the characteristics of salt and pepper noise, this paper adopts the method based on support vector machine to improve the median filtering algorithm. The simulation results show that the proposed algorithm can eliminate salt and pepper noise more effectively, and the processing results have higher SNR. (3) Ostu algorithm based on adaptive threshold selection. Firstly, Ostu algorithm and Bernsen algorithm are used to process images. According to the actual situation of oil field and comparing the processing effect and running time of the two algorithms, this paper selects Ostu algorithm to deal with barcode image. Then the Ostu algorithm is improved because the running time of Ostu algorithm can not meet the requirement of oil field material management. In this paper, the adaptive threshold selection method is used to improve the Ostu algorithm. The result of the improved algorithm shows that the improved algorithm can speed up the processing time and improve the working efficiency.
【学位授予单位】：东北石油大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TE4;TP391.4

【参考文献】