基于游程编码思想的条码定位与识别方法

在复杂背景中的定位条码是图像式条码识别系统中的一个关键步骤,如何在复杂的背景中快速、白动地检测出条形码是该文的主要研究内容。首先介绍了条型码的编码结构,然后提出了一种基于游程编码思想的条码定位与识别方法,最后通过给出相关实验结果,验证了该算法的可行性和实用性。     

基于游程编码思想的条码定位与识别方法

在复杂背景中的定位条码是图像式条码识别系统中的一个关键步骤,在复杂的背景中快速、自动地检测出条形码是文章研究的主要内容。该文首先介绍了条型码的编码结构,然后提出了一种基于游程编码思想的条码定位与识别方法,最后通过相关实验结果验证了该算法的可行性和实用性。     

复杂图像背景下的一维条码定位研究

复杂背景中的条码自动定位在图像式条码识别系统中是一个关键步骤,如何在复杂的背景中快速、自动地检测出条形码是研究的主要内容。主要利用条码区域的边缘密度高、以及边缘方向单一特征检测条码区域,研究了复杂图像背景中条码定位的步骤并提出一种基于边缘梯度方向的边缘提取法,得到了较好的效果。     

一种层次化的Data Matrix实时解码方法

在实际应用中,Data Matrix二维条形码的解码效果会受到不均匀光照和复杂背景的影响。为此,提出一种层次化的Data Matrix实时解码方法。利用大津法将条码图像的前景和背景进行分割,得到二值化的条码图像,对其进行区域定位和精确定位。在此基础上,设计基于数字信号处理的二维条形码快速解码终端。实验结果表明,该解码方法具有较好的实时性和较高的鲁棒性。     

Automated barcode recognition for smart identification and inspection automation

Barcodes have been widely used in many industrial products for automatic identification in data collection and inventory control purposes. It is well known that in many stores laser bar-code readers are used at check-out counters. However, there is a major constraint when this tool is used. That is, unlike traditional camera-based picturing, the distance between the laser reader (sensor) and the target object is close to zero when the reader is applied. This may result in inconvenience in inspection automation because the human operator has to manipulate either the sensor or the objects. For the purpose of in-store or inspection automation, the human operator needs to be removed from the process, i.e. a robot with visual capability is required to play an important role in such a system. Moreover, an automated system is required to find, locate and decode barcodes on various document images even with low resolution compared with laser barcode readers (about 15,000 dots per inch) and can handle damaged bar codes. This paper proposes a smart barcode detection and recognition system (SBDR) based on fast hierarchical Hough transform (HHT). The back-propagation neural network (BPNN) is selected as a powerful tool to perform the recognition process. The paper presents an effective method to utilize the specific graphic features of barcodes for positioning and recognition purposes even in case of distorted barcodes. The first step the system has to perform is to locate the position and orientation of the barcode in the required material document image. Secondly, the proposed system has to segment the barcode. Finally, a trained back-propagation neural network is used to perform the barcode recognition task. Experiments have been conducted to corroborate the efficiency of the proposed method.     

复杂背景下PDF417条码定位研究

PDF417作为一种广泛应用的二维条码,其检测定位成为条码识别过程中的关键,而复杂背景下条码的定位技术目前还不够完善。本文针对这一问题进行了研究,提出了一种基于数学形态学运算和Canny边缘检测的定位算法,并在此基础上作区域精确化处理,实现了条码的准确定位。实验结果表明,该方法能够有效提取出各种低对比度、边缘模糊、倾斜、商标干扰等复杂背景下的条码图像。     

Stabilization and extraction of 2D barcodes for camera phones

With the ubiquity of cellular phones, mobile applications with 2D barcodes have drawn a lot of attentions in recent years. When a user takes a barcode image with the camera in a mobile device, the captured image tends to be blurred due to camera shaking when the user presses the shutter. In addition, the captured image includes part of the complex background of the page with the barcode. In this paper, we point out that the above two issues, which have not been identified in previous works, deteriorate the accuracy of barcode recognition in the mobile computing. We then propose an efficient and effective algorithm to restore and extract 2D barcode from a complex background in a camera-shaken image. Compared with previous approaches, our algorithm outperforms in not only smaller running time but also higher accuracy of the barcode recognition in the mobile computing.     

基于DSP的PDF417快速解码终端的设计与实现

设计并实现了一种基于DSP平台的PDF417条码快速解码终端。该终端能对摄像头采集的含有PDF417条码的图像进行复杂背景下条码区域提取、条码畸变校正等条码图像预处理,实现了复杂背景和不均匀光照条件下条码的全方位快速读取。     

实时卡片字符识别与校验系统的设计与实现

论文设计并实现了一种基于多阈值字符分割、多分类器融合分类和相似判断的IC卡字符识别与校验系统。为了减少分割可能带来的错误识别,系统采用了对多阈值下的识别结果取平均的方法来提高识别的稳定性。通过分类结果、分类权向量和给定字符之间的相似性判断最终确定打印的字符是否正确。相似性规则中的阈值由实验确定。硬件上采用了性价比较高的设备,实验结果表明文中的系统能满足实际要求。     

Pervasive 2D Barcodes for Camera Phone Applications

In a previous study, we evaluated six 2D barcodes using eight criteria for standardization potential: omnidirectional symbol reading, support for low-resolution cameras, reading robustness under different lighting conditions, barcode reading distance, error correction capability, security, support for multiple character sets, and data capacity. We also considered the fidelity of the camera phone's captured image as a metric for gauging reading reliability. Here, we review the six 2D barcodes and then use an additional metric - a first-read rate - to quantitatively verify our earlier results and better gauge reading reliability.     

基于TMS320DM642静脉识别的储物柜系统设计

在现有静脉识别算法的基础上,提出了一种基于TMS320 DM642的静脉识别控制储物柜的方法.该系统主要由电磁锁、电锁控制器和中央处理器组成.存储过程中采集静脉,生成的ID需要与储物柜号绑定,柜门打开;取出过程中,需要静脉验证,打开与其绑定的柜门并删除对应的静脉信息.在原有的基础上,静脉验证取代条形码,这种方法比条形码验证更安全,不易丢失.     

一种高速可靠的彩票数字识别系统

提出了一种彩票数字识别系统。该系统采用了图象处理中的邻域平均法、阈值选取及图象分割技术。应用结果表明 ,该方法简易、可靠、识别速度快、精度高。     

Text detection and recognition in images and video frames

This paper presents a new method for detecting and recognizing text in complex images and video frames. Text detection is performed in a two-step approach that combines the speed of a text localization step, enabling text size normalization, with the strength of a machine learning text verification step applied on background independent features. Text recognition, applied on the detected text lines, is addressed by a text segmentation step followed by an traditional OCR algorithm within a multi-hypotheses framework relying on multiple segments, language modeling and OCR statistics. Experiments conducted on large databases of real broadcast documents demonstrate the validity of our approach.     

复杂环境中多信息融合的手势识别

针对复杂环境中的手势识别问题,提出了一种融合深度信息和红外信息的手势识别方法。首先利用Kinect摄像头的深度信息进行动态实时手势分割,然后融合红外图像复原手势区域。解决了实时手势分割和利用手势的空间分布特征进行手势识别时由于分割的手势区域有缺损或有人脸干扰时识别率低的问题。经实验验证,提出的方法不仅不受环境光线的影响,而且可以识别区分度较小的手势,对旋转、缩放、平移的手势识别也具有鲁棒性。对于区分度较大的手势,识别率高达100%。     

高速条码识别系统

针对目前条码识别算法实时性不够的不足,同时为解决具体应用中定位算法稳定性不高,景深不够等问题,提出了一种基于计算机视觉、FPGA和DSP混合设计的条码识别系统,其高速条码定位算法在FPGA上实现,定位时间仅为2.5ms,定位速度快,同时减少了后续DSP的计算量,大大提高了系统的实时性,并应用于轮胎条码识别系统.经实际试验表明,该系统达到16帧每秒的识别速度,且识别率高,对高速运动条码信息的在线检测应用具有较好的实用价值.     

Data Matrix二维条形码解码器图像预处理研究

DM码是一种常见的二维条形码,图像预处理是DM码解码器自动识别过程中的重要步骤。提出一种实用的DM码识别图像预处理方法。它没有使用传统的边缘检测和直线检测手段,因此受背景噪声、几何失真的影响较小。此外,使用了校正铁路线坐标,并按区域取样生成码流,显著提高了DM码的识别速度和识别率。实验结果表明,该算法可以克服DM码识别过程中易受噪声干扰、光照不均和几何失真等影响的问题。     

Embedding biometric identifiers in 2D barcodes for improved security

Two-dimensional (2D) barcode symbology is an emerging technology used for compactly storing and retrieving information. These barcodes can be found on the back of drivers' licenses and are encoded with secure text data. Standard 2D barcode such as PDF417 uses upper and lowercase alphabets, numeric digits and special characters for encoding. Some barcodes also include a compressed photo of the individual. The visual quality of the compressed image is usually poor and occupies a large amount of space which greatly reduces the capacity needed for encoding text. This paper presents a novel approach for embedding uncompressed images in a standard PDF417 2D barcode using a blind digital watermarking technique. The text is encoded in the standard PDF417 format with error correction, while the face and fingerprint images are watermarked in the encoded 2D barcode. Experimental results show that the proposed technique effectively increased the standard capacity of the PDF417 2D barcode without altering the contents of the encoded data. The results also show that the visual quality of the extracted photo image is high. The extracted fingerprint image when compared with the original fingerprint using an AFIS system yielded a high matching score.     

基于局部透视变换的圆柱体侧表面PDF417条码矫正方法

PDF417条码作为一种应用广泛的二维条码,正在逐步应用于标识各种商品。目前,对于生活中很常见的圆柱体商品,其侧表面PDF417条码的识别还没有研究报导。本文提出了一种基于局部透视变换的图像矫正算法,将圆柱体侧表面PDF417条码分割为若干子区域,然后对各个子区域分别进行透视变换,将各个子区域分别矫正之后再行拼接,最后得到平面PDF417条码的图像。实验结果表明,该方法可以有效地将圆柱体侧表面的PDF417条码矫正为平面PDF417条码,并且正确地解码出条码所含信息。     

金属表面Data Matrix条码高光区域的信息重构

为实现工业产品的可追溯性,直接将条码加工在零件表面的直接零件标识（Direct Part Marking,DPM）技术,在国内外受到了越来越多的关注。对于金属零件,由于其具有较高的反光性,由相机捕获的金属表面的条码图像常常产生局部高光现象,影响条码的正确读取。为此,针对金属表面激光标刻二维条码出现的局部高光现象,提出了基于五步重构模型的条码重构法,以重构高光区域的条码信息。对获得的条码图像进行倾斜校正,使“L”型实线边界位于图像左下角,对条码进行网格划分实现各个模块的定位。基于Modified Specular-Free（MSF）图像对高光区域进行检测。采用五步重构模型对条码的各个模块进行数值填充,对条码进行读取。实验表明,该算法能达到去除金属表面上条码局部高光的目的,并取得了较高的识读正确率。