WO2022007431A1 - 一种Micro QR二维码的定位方法 - Google Patents

一种Micro QR二维码的定位方法 Download PDF

Info

Publication number
WO2022007431A1
WO2022007431A1 PCT/CN2021/081732 CN2021081732W WO2022007431A1 WO 2022007431 A1 WO2022007431 A1 WO 2022007431A1 CN 2021081732 W CN2021081732 W CN 2021081732W WO 2022007431 A1 WO2022007431 A1 WO 2022007431A1
Authority
WO
WIPO (PCT)
Prior art keywords
micro
code
image
point
dimensional code
Prior art date
Application number
PCT/CN2021/081732
Other languages
English (en)
French (fr)
Inventor
陈余泉
卢盛林
Original Assignee
广东奥普特科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 广东奥普特科技股份有限公司 filed Critical 广东奥普特科技股份有限公司
Publication of WO2022007431A1 publication Critical patent/WO2022007431A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06KGRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/14Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
    • G06K7/1404Methods for optical code recognition
    • G06K7/1408Methods for optical code recognition the method being specifically adapted for the type of code
    • G06K7/14172D bar codes
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06KGRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/14Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
    • G06K7/1404Methods for optical code recognition
    • G06K7/1439Methods for optical code recognition including a method step for retrieval of the optical code
    • G06K7/1443Methods for optical code recognition including a method step for retrieval of the optical code locating of the code in an image
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06KGRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/14Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
    • G06K7/1404Methods for optical code recognition
    • G06K7/1439Methods for optical code recognition including a method step for retrieval of the optical code
    • G06K7/1452Methods for optical code recognition including a method step for retrieval of the optical code detecting bar code edges
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/70Denoising; Smoothing
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/13Edge detection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20024Filtering details
    • G06T2207/20032Median filtering

Definitions

  • the invention belongs to the technical field of two-dimensional codes, and in particular relates to a positioning method of a Micro QR two-dimensional code.
  • QR code Quality Response Code, Quick Response Matrix Code
  • QR code Quick Response Code, Quick Response Matrix Code
  • Micro QR codes can be used as a supplement to the small-capacity storage methods of QR codes, thereby expanding the QR code. field of application of the code.
  • FIG. 1 a schematic diagram of the structure of the Micro QR code is shown.
  • the Micro QR code is only provided with a positioning block graphic, which will be used for saving
  • the clock area of the coded content of the two-dimensional code is set at the edge of the entire code area.
  • the traditional QR code positioning method mainly determines the scope of the code area according to the right-angled triangle relationship of the three positioning blocks. Therefore, the traditional QR code positioning method cannot locate the Micro QR code or has the defect of inaccurate positioning. .
  • the purpose of the present invention is: aiming at the deficiencies of the prior art, a kind of positioning method of Micro QR two-dimensional code is provided, and the Micro QR two-dimensional code can be effectively and accurately positioned by this method.
  • the present invention adopts the following technical solutions:
  • a method for locating a Micro QR two-dimensional code comprising the following steps:
  • Step 1 Perform edge detection on the Micro QR code image, after obtaining the connected domain, extract all contours according to the connected domain;
  • Step 2 Select the rectangular connected domain from all the contours obtained in step 1, and determine the center of the rectangular positioning block;
  • Step 3 according to the proportional relationship of the two-dimensional code positioning block, calculate the precise coordinates of the center of the positioning block;
  • Step 4 Taking the center point of the positioning block as the starting point, obtain the horizontal boundary point and the vertical boundary point of the code area along the horizontal and vertical directions of the clock area respectively, and according to the two boundary points and the center point of the positioning block , to determine the code area boundary of the Micro QR code.
  • the step 1 includes the following steps:
  • Step 1.1 filter out noise
  • Step 1.2 calculate the gradient intensity and direction of each pixel in the image
  • Step 1.3 Use non-maximum suppression to eliminate spurious responses caused by edge detection
  • Step 1.4 Use dual threshold detection to determine real and potential edges
  • Step 1.5 Complete edge detection by suppressing isolated weak edges and extract all contours, where the set of contours is ⁇ L1,L2,...,Ln ⁇ , and each contour Li consists of point columns ⁇ A0,A1,... , An ⁇ constitutes.
  • the step 2 is specifically to screen out a rectangular closed contour from all the obtained contours, which specifically includes the following steps:
  • Step 2.1 traverse the contour set ⁇ L1, L2,..., Ln ⁇ , calculate the distance between the starting point A0 and the ending point An of each contour Li, denoted as Distance, if Distance ⁇ D1pixel, the contour is a closed contour, where , D1 is the rectangle judgment parameter, and its size is determined by the size of the Micro QR code;
  • Step 2.2 Calculate the area and contour length of the extracted contour according to the properties of the rectangle, and then remove the unsatisfactory proportional relationship , where Area is the area of the contour, Length is the length of the contour, and D2 is a value close to 0. The specific value is determined according to the degree of ambiguity of the Micro QR code.
  • the step 3 specifically includes the following steps:
  • Step 3.1 traverse all rectangular closed contours obtained in step 2;
  • Step 3.2 extract the centroid (X, Y) of the closed contour of the rectangle;
  • Step 3.3 Calculate the angle between the closed contour of the rectangle and the horizontal direction, denoted as Angle;
  • Step 3.4 Taking the abscissa X of the centroid (X, Y) as the starting point, traverse the image in the horizontal direction according to the relationship of the positioning block 1:1:3:1:1, and obtain the abscissa X0 of the centroid correction point;
  • Step 3.5 Take the abscissa X0 of the centroid correction point as the starting point, traverse the image vertically, and obtain the ordinate Y0 of the center correction point;
  • Step 3.6 The precise position of the center of the rectangular frame is StartPoint(X0, Y0).
  • the acquisition of the boundary point in the horizontal direction in the step 4 specifically includes the following steps:
  • Step 4.1a Starting from the precise position StartPoint (X0, Y0) of the center of the rectangular frame, starting from the -90° direction of the included angle Angle, that is, rotating 90 degrees counterclockwise, and recording the three points that intersect with the edge of the code area.
  • StartPoint X0, Y0
  • -90° direction of the included angle Angle that is, rotating 90 degrees counterclockwise
  • the acquisition of the boundary point in the vertical direction in the step 4 specifically includes the following steps:
  • Step 4.1b Starting from the precise position StartPoint (X0, Y0) of the center of the rectangular frame, starting from the 180° direction of the included angle Angle, the three points that intersect with the edge of the code area are respectively recorded as A1, A2, A3;
  • the size is determined according to the ambiguity of the Micro QR code.
  • the step 1 further includes image preprocessing before edge detection is performed on the Micro QR code image, and the preprocessing includes the following steps :
  • Step 0.1 read the grayscale image of the Micro QR code
  • Step 0.2 the input image is smoothed by using a nonlinear filter to terminate the filtering, and the output of the two-dimensional median filter is:
  • f(x, y), g(x, y) are the original image and the processed image, respectively, and W is a two-dimensional template.
  • the step 4 after the step 4, it also includes post-processing on the two-dimensional code image, and the post-processing includes correcting the two-dimensional code image, and then after two After the value processing, the accurately positioned Micro QR two-dimensional code image is obtained.
  • Step 5.1 Map the center coordinates of the positioning block of the Micro QR code, the EngPointX coordinates of the horizontal boundary point and the EngPointY coordinates of the vertical boundary point to the centers of the three positioning blocks of the standard QR code, which satisfy the following relationship :
  • the image rotation correction coefficients a11, a12, a21, and a22 are obtained by solving, and after rotating the original Micro QR code according to affine transformation, the standard Micro QR two-dimensional code image is obtained by bilinear interpolation;
  • T is the segmentation threshold. Binarization is performed according to the segmentation threshold to obtain the Micro QR code image after accurate positioning.
  • the beneficial effects of the present invention are: compared with the prior art, the present invention can accurately locate the MicroQR two-dimensional code image, and completely cut out the code area image for the decoding program to call and decode; for the two-dimensional code image under complex background, Through contour extraction, the problem of inability to locate or inaccurate positioning of MicroQR codes for QR code images with poor image clarity and uneven illumination is solved, which is helpful for locating QR codes by using position locator features. .
  • Fig. 1 is the structural schematic diagram of Micro QR two-dimensional code
  • Fig. 2 is the working flow chart of the present invention
  • Fig. 3 is the original image of Micro QR two-dimensional code in the embodiment of the present invention.
  • Fig. 4 is the Micro QR two-dimensional code image after smooth processing in the embodiment of the present invention.
  • Fig. 5 is the Micro QR two-dimensional code image whose outline is extracted by Canny operator in the embodiment of the present invention.
  • FIG. 6 is an image of a rectangular frame positioning block in an embodiment of the present invention.
  • Fig. 8 is three positioning points of the code area image in the embodiment of the present invention.
  • Fig. 9 is the image of the code area in the embodiment of the present invention.
  • FIG. 10 is an image of the code region after binarization in the embodiment of the present invention.
  • the terms “installed”, “connected”, “connected”, “fixed” and other terms should be understood in a broad sense, for example, it may be a fixed connection or a detachable connection , or integrally connected; it can be a mechanical connection or an electrical connection; it can be a direct connection, or an indirect connection through an intermediate medium, or the internal communication between the two components.
  • installed e.g., it may be a fixed connection or a detachable connection , or integrally connected; it can be a mechanical connection or an electrical connection; it can be a direct connection, or an indirect connection through an intermediate medium, or the internal communication between the two components.
  • a method for locating a Micro QR code includes the following steps:
  • Step 1 Perform edge detection on the Micro QR code image, after obtaining the connected domain, extract all contours according to the connected domain;
  • Step 2 Select the rectangular connected domain from all the contours obtained in step 1, and determine the center of the rectangular positioning block;
  • Step 3 according to the proportional relationship of the two-dimensional code positioning block, calculate the precise coordinates of the center of the positioning block;
  • Step 4 Taking the center point of the positioning block as the starting point, obtain the horizontal boundary point and the vertical boundary point of the code area along the horizontal and vertical directions of the clock area respectively, and according to the two boundary points and the center point of the positioning block , to determine the code area boundary of the MicroQR code.
  • the present invention also includes preprocessing, wherein the preprocessing is to perform image smoothing processing on the image before step 1 .
  • the preprocessing includes the following steps:
  • Step 0.1 read the grayscale image of the Micro QR code
  • Step 0.2 the input image is smoothed by using a nonlinear filter to terminate the filtering, and the output of the two-dimensional median filter is:
  • f(x, y), g(x, y) are the original image and the processed image, respectively, and W is a two-dimensional template.
  • FIG. 3 it is the original image of the MicroQR two-dimensional code in the embodiment of the present invention.
  • Read the grayscale image shown in Figure 3 and apply a nonlinear filter median filter to the input image to smooth the image to eliminate the process of image acquisition, acquisition, transmission and conversion (such as imaging, copy scanning, transmission and display, etc.) , the image is disturbed by varying degrees of visible or invisible noise.
  • Median filtering can overcome the image blur problem caused by linear filter to a certain extent, while filtering out noise, the edge information of the image is better preserved.
  • the smooth filtering window used in the present invention is controlled according to the intensity of the noise.
  • the filtering window is [3*3].
  • the final smoothing result is shown in Figure 4.
  • FIG. 5 is an image of the contour extracted by the Canny operator in the embodiment of the present invention.
  • the contour of the image extracted in step 1.1 uses the canny operator to extract the contour, including the following steps:
  • Step 1.1 filter out noise
  • Step 1.2 calculate the gradient intensity and direction of each pixel in the image
  • Step 1.3 Use non-maximum suppression to eliminate spurious responses caused by edge detection
  • Step 1.5 Complete edge detection by suppressing isolated weak edges and extract all contours, where the set of contours is ⁇ L1,L2,...,Ln ⁇ , and each contour Li consists of point columns ⁇ A0,A1,... , An ⁇ constitutes.
  • FIG. 6 is an image of a rectangular frame positioning block in an embodiment of the present invention.
  • all the rectangular frame images in the image are selected from all the contours obtained in the previous step that conform to the rectangular positioning frame, where the contour set ⁇ L1, L2, ..., Ln ⁇ contains any contours.
  • Li consists of point columns ⁇ A0,A1,...,An ⁇ . Since the contour of the positioning block is a rectangular closed contour, here we mainly find the closed contour that constitutes the positioning block, which includes the following steps:
  • Step 2.2 Calculate the area and contour length of the extracted contour according to the properties of the rectangle, and then remove the unsatisfactory proportional relationship
  • the contour of where Area is the area of the contour, Length is the length of the contour, and D2 is a value close to 0.
  • the step specifically includes the following steps:
  • Step 3.1 traverse all rectangular closed contours obtained in step 2.2;
  • Step 3.2 extract the centroid (X, Y) of the closed contour of the rectangle;
  • Step 3.3 Calculate the angle between the closed contour of the rectangle and the horizontal direction, denoted as Angle;
  • Step 3.4 Taking the abscissa X of the centroid (X, Y) as the starting point, traverse the image in the horizontal direction according to the relationship of the positioning block 1:1:3:1:1, and obtain the abscissa X0 of the centroid correction point;
  • Step 3.5 Take the abscissa X0 of the centroid correction point as the starting point, traverse the image vertically, and obtain the ordinate Y0 of the center correction point;
  • Step 3.6 The precise position of the center of the rectangular frame is StartPoint(X0, Y0).
  • FIG. 7 is a search image of horizontal and vertical boundary points in an embodiment of the present invention.
  • the center of the positioning block is taken as the starting point, and the horizontal boundary point of the code area is found along the horizontal direction of the clock area. Then find the vertical boundary point of the code area along the vertical direction of the clock area. According to the center of the positioning block and two boundary points, the code area boundary of MicroQR is determined.
  • the acquisition of boundary points in the horizontal direction specifically includes the following steps:
  • Step 4.1a The clock area in the horizontal direction is flush with the black and white outermost layer of the positioning block. Starting from the precise position StartPoint (X0, Y0) of the center of the rectangular frame, starting from the -90° direction of the included angle, that is, along the Rotate 90 degrees counterclockwise, intersect with the white edge three times, and record the focus as A1, A2, and A3 respectively;
  • the acquisition of boundary points in the vertical direction includes the following steps:
  • Step 4.1b Starting from the precise position StartPoint (X0, Y0) of the center of the rectangular frame, starting from the 180° direction of the included angle Angle, the three points that intersect with the edge of the code area are respectively recorded as A1, A2, A3;
  • the size is determined according to the ambiguity of the Micro QR code.
  • the distance between the boundary point in the vertical direction and the center of the positioning block and the distance between the boundary point in the horizontal direction and the center of the positioning block are calculated respectively.
  • There are three positioning points in the area image, and the corresponding distance can be obtained by using the distance formula only according to the coordinates of the horizontal boundary point and the vertical boundary point. In order to simplify and improve the calculation speed of the algorithm, the following formula is used. If pattern X pattern Y, it means that the boundary point in the horizontal direction and the boundary point in the vertical direction are correct, and then proceed to the next step; otherwise, continue to traverse the next rectangular frame .
  • the embodiment of the present invention also includes post-processing, that is, after step 4, the image is corrected and binarized to obtain a precisely positioned Micro QR two-dimensional code image.
  • post-processing if the image is in a tilted state, Then it is also necessary to use the affine transformation method for calculation.
  • the specific calculation process is as follows:
  • Fig. 9 is the code area image in the embodiment of the present invention, and it is subjected to affine transformation, which mainly includes the following steps:
  • Step 5.1 Map the center coordinates (X0, Y0) of the positioning block of the Micro QR code, the EngPointX coordinate of the horizontal boundary point and the EngPointY coordinate of the vertical boundary point to the center of the three positioning blocks of the standard QR two-dimensional code, Among them, the corresponding points of (X0, Y0) are (3.5, 3.5), and the corresponding points of EngPointX and EngPointY are (0.5, 7+pattern X ⁇ 2), (7+pattern X ⁇ 2, 0.5), respectively.
  • the standard two-dimensional The block size of the codes is 1, and they satisfy the following relation:
  • the image rotation correction coefficients a11, a12, a21, a22, b1, and b2 are obtained by solving, and after rotating the original Micro QR code according to the affine transformation, the standard Micro QR two-dimensional code image is obtained by bilinear interpolation;
  • the maximum inter-class variance threshold method is an adaptive threshold determination method. The algorithm assumes that image pixels can be divided into background and target parts according to a threshold. Then, the optimal threshold is calculated to distinguish the two types of pixels, so that the two types of pixels have the maximum discrimination degree. Since the code area image is a black and white coded image, it is most suitable to use the maximum inter-class variance threshold method.
  • T is the segmentation threshold. Binarization is performed according to the separation threshold to obtain the Micro QR code image after precise positioning.
  • the acquired code area image is deliberately selected to exceed the edge of the code area by 2 pixels on the top, bottom, left and right, that is, the microQR two-dimensional code area that meets the international standard. size.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Health & Medical Sciences (AREA)
  • Electromagnetism (AREA)
  • General Health & Medical Sciences (AREA)
  • Toxicology (AREA)
  • Artificial Intelligence (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

一种Micro QR二维码的定位方法,包括步骤1、对Micro QR二维码图像进行边缘检测,获取连通域后,再根据连通域提取所有轮廓;步骤2、从步骤1中得到的所有轮廓中挑选出矩形连通域,确定矩形定位块中心;步骤3、根据二维码定位块的比例关系,计算出定位块中心的精确坐标;步骤4、以定位块的中心点为起点,分别沿时钟区的水平和垂直方向获取码区的水平方向的边界点和垂直方向的边界点,并根据这两个边界点和定位块的中心点,确定Micro QR二维码的码区边界。通过该方法有效解决了Micro QR二维码无法定位或定位不够准确的问题。

Description

一种Micro QR二维码的定位方法
本申请要求于2020年7月7日提交中国专利局、申请号为202010643510.6、发明名称为“一种Micro QR二维码的定位方法”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本发明属于二维码技术领域,具体涉及一种Micro QR二维码的定位方法。
背景技术
QR码(Quick Response Code,快速响应矩阵码)是二维码的一种,不仅存储量大、成本低,还可以表示汉字及图像等多种文字信息、其保密防伪性强而且使用非常方便,因此,已被各行业普遍采用。而为了使镶嵌在物体上的二维码更“微型化”,行业内提出了一种Micro QR二维码,相对于QR码,Micro QR二维码只需要一个查找符号,拥有更小的体积和信息头开销,可以存储20bits到128bits容量的数据,此容量与QR码的存储量间没有交集,因此,Micro QR二维码可以作为QR码小容量存储方式的一种补充,从而扩大了QR码的应用领域。
如附图1中所示,展示了Micro QR二维码的结构示意图,与QR码不同的是,为了减少打印的尺寸,Micro QR二维码只设置有一个定位块图形,并且将用于保存二维码编码内容的时钟区设置在整个码区的边缘处。然而传统的QR码定位方法主要是根据三个定位块的直角三角关系来确定码区范围的,因此,采用传统的QR码定位方法无法对Micro QR二维码进行定位或存在定位不准确的缺陷。
有鉴于此,上述问题亟待解决。
发明内容
本发明的目的在于:针对现有技术的不足,而提供的一种Micro QR二维码的定位方法,通过该方法能对Micro QR二维码进行有效准确的定位。
为实现上述目的,本发明采用如下技术方案:
一种Micro QR二维码的定位方法,包括以下步骤:
步骤1、对Micro QR二维码图像进行边缘检测,获取连通域后,再根据连通域提取所有轮廓;
步骤2、从步骤1中得到的所有轮廓中挑选出矩形连通域,确定矩形定位块中心;
步骤3、根据二维码定位块的比例关系,计算出定位块中心的精确坐标;
步骤4、以定位块的中心点为起点,分别沿时钟区的水平和垂直方向获取码区的水平方向的边界点和垂直方向的边界点,并根据这两个边界点和定位块的中心点,确定Micro QR二维码的码区边界。
作为对本发明中所述的Micro QR二维码的定位方法的改进,所述步骤1包括以下步骤:
步骤1.1、滤除噪声;
步骤1.2、计算图像中每个像素点的梯度强度和方向;
步骤1.3、使用非极大值抑制,消除边缘检测带来的杂散响应;
步骤1.4、使用双阈值检测确定真实的和潜在的边缘;
步骤1.5、通过抑制孤立的弱边缘完成边缘检测并提取所有轮廓,其中,轮廓的集合为{L1,L2,...,Ln},每个轮廓Li由点列{A0,A1,...,An}构成。
作为对本发明中所述的Micro QR二维码的定位方法的改进,所述步骤2具体为在获取的全部轮廓中筛选出矩形封闭的轮廓,具体包括以下步骤:
步骤2.1、遍历轮廓集合{L1,L2,...,Ln},计算每个轮廓Li的始点A0与终点An的距离,记为Distance,若Distance<D1pixel,则该轮廓为封闭的轮廓,其中,D1为矩形判断参数,其大小由Micro QR二维码的尺寸确定;
步骤2.2、根据矩形的性质,计算提取出的轮廓的面积与轮廓长度,再去除不满足比例关系式
Figure PCTCN2021081732-appb-000001
的轮廓,其中,Area为轮 廓的面积,Length为轮廓的长度,D2为接近于0的值,具体数值大小根据Micro QR二维码的模糊程度确定。
作为对本发明中所述的Micro QR二维码的定位方法的改进,所述步骤3具体包括以下步骤:
步骤3.1、遍历所述步骤2中获取的所有矩形封闭轮廓;
步骤3.2、提取矩形封闭轮廓的质心(X,Y);
步骤3.3、计算矩形封闭轮廓与水平方向的夹角,记为Angle;
步骤3.4、以质心(X,Y)的横坐标X为起点,根据定位块1:1:3:1:1的关系沿横向遍历图像,得到质心校正点的横坐标X0;
步骤3.5、以质心校正点的横坐标X0为起点,纵向遍历图像,得到中心的校正点纵坐标Y0;
步骤3.6、矩形框中心的精确位置为StartPoint(X0,Y0)。
作为对本发明中所述的Micro QR二维码的定位方法的改进,所述步骤4中水平方向的边界点获取具体包括以下步骤:
步骤4.1a、从矩形框中心的精确位置StartPoint(X0,Y0)出发,沿夹角Angle的-90°方向出发,即沿逆时针方向旋转90度,与码区的边缘相交的三点分别记为A1,A2,A3;
步骤4.2a、取新的起点,记为NewStartPoint(X,Y),其坐标设置为NewStartPoint=0.5×(A2+A3),再从NewStartPoint(X,Y)出发,转向90°,沿夹角Angle遍历整幅图像,与码区边缘相交,交点依次记为B1,B2,B3...Bn;
步骤4.3a、将B1和B2的长度B1B2设置为基准,并设置pattern X=0,计算B2到B4的长度,记为B2B4,若
Figure PCTCN2021081732-appb-000002
成立,记pattern X=pattern X+1,并继续计算后续点列,直到不满足上述关系式,停止遍历,得到水平方向的边界点坐标EngPointX,其中,D3为接近于0的值,具体数值大小根据Micro QR二维码的模糊程度确定。
作为对本发明中所述的Micro QR二维码的定位方法的改进,所述步骤4中垂直方向的边界点获取具体包括以下步骤:
步骤4.1b、从矩形框中心的精确位置StartPoint(X0,Y0)出发,沿夹角Angle的180°方向出发,与码区的边缘相交的三点分别记为A1,A2,A3;
步骤4.2b、取新的起点,记为NewStartPoint(X,Y),其坐标设置为NewStartPoint=0.5×(A2+A3),再从NewStartPoint(X,Y)出发,转向90°,沿夹角Angle的90°方向遍历整幅图像,与码区边缘相交,交点依次记为B1,B2,B3...Bn;
步骤4.3b、将B1和B2的长度B1B2设置为基准,并设置pattern Y=0,计算B2到B4的长度,记为B2B4,若
Figure PCTCN2021081732-appb-000003
成立,记pattern Y=pattern Y+1,并继续计算后续点列,直到不满足上述关系式,停止遍历,最后得到垂直方向的边界点坐标EngPointY,其中,D3为接近于0的值,具体数值大小根据Micro QR二维码的模糊程度确定。
作为对本发明中所述的Micro QR二维码的定位方法的改进,在获取水平方向的边界点和垂直方向的边界点后,分别计算垂直方向的边界点与定位块中心的距离以及水平方向的边界点与定位块中心的距离,若pattern X=pattern Y,则表示水平方向的边界点和垂直方向的边界点是正确的,继续执行下一步骤;否则继续遍历下一个矩形框。
作为对本发明中所述的Micro QR二维码的定位方法的改进,所述步骤1在对Micro QR二维码图像进行边缘检测前,还包括对图像的预处理,所述预处理包括以下步骤:
步骤0.1、读取Micro QR二维码的灰度图像;
步骤0.2、对输入的图像采用非线性滤波器终止滤波进行图像平滑处理,其二维中值滤波输出为:
g(x,y)=med{f(x-k,y-l),(k,l∈W)},
其中,f(x,y),g(x,y)分别为原始图像和处理后的图像,W为二维模板。
作为对本发明中所述的Micro QR二维码的定位方法的改进,所述步 骤4之后还包括对二维码图像的后处理,所述后处理包括对二维码图像进行校正,再经过二值化处理后得到精确定位后的Micro QR二维码图像。
作为对本发明中所述的Micro QR二维码的定位方法的改进,所述后处理中若图像为倾斜状态,则按照以下步骤进行校正:
步骤5.1、将Micro QR二维码的定位块中心坐标和水平方向的边界点EngPointX坐标以及垂直方向的边界点EngPointY坐标映射到标准QR二维码的三个定位块的中心,它们满足以下关系式:
Figure PCTCN2021081732-appb-000004
通过求解得到图像旋转校正系数a11、a12、a21、a22,将原始的Micro QR二维码按照仿射变换旋转后,利用双线性插值法得到标准的Micro QR二维码图像;
步骤5.2、将Micro QR二维码图像像素分为背景和目标两部分,按照0到255个灰度值遍历,计算并找到前景和背景最大类间方差variance=w0×w1×(u0-u1)×(u0-u1),其中,w0为分开后前景像素点数占图像的比例;u0为分开后前景像素点的平均灰度;w1为分开后背景像素点数占图像的比例;u1为分开后背景像素点的平均灰度,当找到最大方差时停止遍历,此时的T为分割阈值,根据分隔阈值进行二值化处理,得到精确定位后的Micro QR二维码图像。
本发明的有益效果在于:与现有技术相比,本发明能够对MicroQR二维码图像准确定位,并且完整切出码区图像,供解码程序调用解码;对于复杂背景下的二维码图像,通过轮廓提取,解决了对于图像清晰度较差和光照不均匀的二维码图像对MicroQR二维码无法进行定位或定位不准确的问题,有助于利用位置定位符特征对二维码进行定位。
附图说明
此处所说明的附图用来提供对本发明的进一步理解,构成本发明的一 部分,本发明的示意性实施方式及其说明用于解释本发明,并不构成对本发明的不当限定。在附图中:
图1为Micro QR二维码的结构示意图;
图2为本发明的工作流程图;
图3为本发明实施例中的Micro QR二维码原图像;
图4为本发明实施例中经过平滑处理后的Micro QR二维码图像;
图5为本发明实施例中经过Canny算子提取轮廓的Micro QR二维码图像;
图6为本发明实施例中矩形框定位块的图像;
图7为本发明实施例中边界探索的图像;
图8为本发明实施例中码区图像的三个定位点;
图9为本发明实施例中码区的图像;
图10为本发明实施例中码区二值化后的图像。
具体实施方式
如在说明书及权利要求当中使用了某些词汇来指称特定组件,本领域技术人员应可理解,硬件制造商可能会用不同名词来称呼同一个组件。本说明书及权利要求并不以名称的差异来作为区分组件的方式,而是以组件在功能上的差异来作为区分的准则。如在通篇说明书及权利要求当中所提及的“包含”为一开放式用语,故应解释成“包含但不限定于”。“大致”是指在可接受的误差范围内,本领域技术人员能够在一定误差范围内解决所述技术问题,基本达到所述技术效果。
在本发明的描述中,需要理解的是,术语“上”、“下”、“前”、“后”、“左”、“右”、水平”等指示的方位或位置关系为基于附图所示的方位或位置关系,仅是为了便于描述本发明和简化描述,而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作,因此不能理解为对本发明的限制。
在本发明中,除非另有明确的规定和限定,术语“安装”、“相连”、“连接”、“固定”等术语应做广义理解,例如,可以是固定连接,也可以是可 拆卸连接,或一体地连接;可以是机械连接,也可以是电连接;可以是直接相连,也可以通过中间媒介间接相连,可以是两个元件内部的连通。对于本领域的普通技术人员而言,可以根据具体情况理解上述术语在本发明中的具体含义。
以下结合附图对本发明作进一步详细说明,但不作为对本发明的限定。
如图2所示,一种Micro QR二维码的定位方法,包括以下步骤:
步骤1、对Micro QR二维码图像进行边缘检测,获取连通域后,再根据连通域提取所有轮廓;
步骤2、从步骤1中得到的所有轮廓中挑选出矩形连通域,确定矩形定位块中心;
步骤3、根据二维码定位块的比例关系,计算出定位块中心的精确坐标;
步骤4、以定位块的中心点为起点,分别沿时钟区的水平和垂直方向获取码区的水平方向的边界点和垂直方向的边界点,并根据这两个边界点和定位块的中心点,确定MicroQR二维码的码区边界。
另外,需要说明的是,本发明还包括预处理,其中,预处理为在步骤1前对图像进行图像平滑处理。
其中,预处理包括以下步骤:
步骤0.1、读取Micro QR二维码的灰度图像;
步骤0.2、对输入的图像采用非线性滤波器终止滤波进行图像平滑处理,其二维中值滤波输出为:
g(x,y)=med{f(x-k,y-l),(k,l∈W)},
其中,f(x,y),g(x,y)分别为原始图像和处理后的图像,W为二维模板。
如图3所示,为本发明实施方式中MicroQR二维码原图像。读取图3所示的灰度图像,对输入图像采用非线性滤波器中值滤波进行图像平滑处理,消除图像在采集、获取、传送和转换(如成像、复制扫描、传输以及显示等)过程中,图像被不同程度的可见或不可见的噪声干扰。中值滤波,在一定程度上可以克服线性滤波器所带来的图像模糊问题,在滤除噪声的同时,较好地保留了图像的边缘信息。二维中值滤波输出为g(x,y)= med{f(x-k,y-l),(k,l∈W)},其中,f(x,y),g(x,y)分别为原始图像和处理后的图像,W为二维模板。
于本发明中采用的平滑滤波的窗口是根据噪声的强度来控制的,针对图3所示MicroQR二维码,滤波窗口为[3*3]。最终平滑结果如图4所示。
其中,图5为本发明实施例中Canny算子提取轮廓的图像,本发明中步骤1.1中提取图像轮廓使用了canny算子提取轮廓,包括以下步骤:
步骤1.1、滤除噪声;
步骤1.2、计算图像中每个像素点的梯度强度和方向;
步骤1.3、使用非极大值抑制,消除边缘检测带来的杂散响应;
步骤1.4、使用双阈值检测确定真实的和潜在的边缘,于本实施例中,低阈值=10,高阈值=30;
步骤1.5、通过抑制孤立的弱边缘完成边缘检测并提取所有轮廓,其中,轮廓的集合为{L1,L2,...,Ln},每个轮廓Li由点列{A0,A1,...,An}构成。
图6为本发明实施例中矩形框定位块图像。如图6所示,图像里所有矩形框图像为从上一步骤中获取的全部轮廓中挑选出符合矩形定位框的轮廓,其中轮廓集合{L1,L2,...,Ln}包含的任意轮廓Li由点列{A0,A1,...,An}构成,鉴于定位块的轮廓是矩形封闭轮廓,这里主要寻找构成定位块的封闭轮廓,具体包括以下步骤:
步骤2.1、遍历轮廓集合{L1,L2,...,Ln},计算每个轮廓Li的始点A0与终点An的距离,记为Distance,若Distance<D1pixel,则该轮廓为封闭的轮廓,其中,D1为矩形判断参数,其大小由Micro QR二维码的尺寸确定,本实施例中的D1=0.1;
步骤2.2、根据矩形的性质,计算提取出的轮廓的面积与轮廓长度,再去除不满足比例关系式
Figure PCTCN2021081732-appb-000005
的轮廓,其中,Area为轮廓的面积,Length为轮廓的长度,D2为接近于0的值,具体数值大小根据Micro QR二维码的模糊程度确定,本实施例中的D2=0.1。
优选的,步骤具体包括以下步骤:
步骤3.1、遍历所述步骤2.2获取的所有矩形封闭轮廓;
步骤3.2、提取矩形封闭轮廓的质心(X,Y);
步骤3.3、计算矩形封闭轮廓与水平方向的夹角,记为Angle;
步骤3.4、以质心(X,Y)的横坐标X为起点,根据定位块1:1:3:1:1的关系沿横向遍历图像,得到质心校正点的横坐标X0;
步骤3.5、以质心校正点的横坐标X0为起点,纵向遍历图像,得到中心的校正点纵坐标Y0;
步骤3.6、矩形框中心的精确位置为StartPoint(X0,Y0)。
图7为本发明实施例中水平与垂直边界点探索图像。根据MicroQR二维码时钟区的特性,以定位块中心为起点,沿着时钟区水平方向寻找码区水平方向边界点。再沿着时钟区垂直方向寻找码区垂直方向边界点。根据定位块的中心以及两个边界点,确定出MicroQR的码区边界。其中,水平方向的边界点获取具体包括以下步骤:
步骤4.1a、水平方向时钟区与定位块黑白黑的最外层是平齐的,从矩形框中心的精确位置StartPoint(X0,Y0)出发,沿夹角Angle的-90°方向出发,即沿逆时针方向旋转90度,与白色的边缘相交三次,焦点分别记为A1,A2,A3;
步骤4.2a、取新的起点,记为NewStartPoint(X,Y),其坐标设置为NewStartPoint=0.5×(A2+A3),再从NewStartPoint(X,Y)出发,转向90°,沿夹角Angle遍历整幅图像,与码区白色的边缘相交,交点依次记为B1,B2,B3...;
步骤4.3a、将B1和B2的长度B1B2设置为基准,并设置pattern X=0,计算B2到B4的长度,记为B2B4,若
Figure PCTCN2021081732-appb-000006
成立,记pattern X=pattern X+1,并继续计算后续点列,直到不满足上述关系式,停止遍历,得到水平方向的边界点坐标EngPointX,其中,D3为接近于0的值,具体数值大小根据Micro QR二维码的模糊程度确定,本实施例中D3=0.1。
垂直方向的边界点获取具体包括以下步骤:
步骤4.1b、从矩形框中心的精确位置StartPoint(X0,Y0)出发,沿夹角Angle的180°方向出发,与码区的边缘相交的三点分别记为A1,A2,A3;
步骤4.2b、取新的起点,记为NewStartPoint(X,Y),其坐标设置为NewStartPoint=0.5×(A2+A3),再从NewStartPoint(X,Y)出发,转向90°,沿夹角Angle的90°方向遍历整幅图像,与码区边缘相交,交点依次记为B1,B2,B3...Bn;
步骤4.3b、将B1和B2的长度B1B2设置为基准,并设置pattern Y=0,计算B2到B4的长度,记为B2B4,若
Figure PCTCN2021081732-appb-000007
成立,记pattern Y=pattern Y+1,并继续计算后续点列,直到不满足上述关系式,停止遍历,最后得到垂直方向的边界点坐标EngPointY,其中,D3为接近于0的值,具体数值大小根据Micro QR二维码的模糊程度确定。
在获取水平方向的边界点和垂直方向的边界点后,分别计算垂直方向的边界点与定位块中心的距离以及水平方向的边界点与定位块中心的距离,图8为本发明实施例中码区图像三个定位点,只需根据水平边界点与垂直边界点的坐标,利用距离公式可求出相应的距离。为了简化和提高的算法计算速度,利用下式,若pattern X=pattern Y,则表示水平方向的边界点和垂直方向的边界点是正确的,继续执行下一步骤;否则继续遍历下一个矩形框。
本发明实施例中还包括后处理,即在步骤4之后对图像进行校正,与二值化处理,得到精确定位后的Micro QR二维码图像,在后处理的过程中若图像为倾斜状态,则还需要采用仿射变换的方法进行计算。具体计算过程如下:
利用仿射变换对二维码图像进行校正,将码区校正为正方形区域并切割出码区图像。图9为本发明实施例中码区图像,对它进行仿射变换,主要包括以下步骤:
步骤5.1、将Micro QR二维码的定位块中心坐标(X0,Y0)和水平方向 的边界点EngPointX坐标以及垂直方向的边界点EngPointY坐标映射到标准QR二维码的三个定位块的中心,其中,(X0,Y0)对应的点为(3.5,3.5),EngPointX和EngPointY对应的点分别为(0.5,7+pattern X×2),(7+pattern X×2,0.5),标准二维码的模块尺寸取值为1,它们满足以下关系式:
Figure PCTCN2021081732-appb-000008
通过求解得到图像旋转校正系数a11、a12、a21、a22、b1、b2,将原始的Micro QR二维码按照仿射变换旋转后,利用双线性插值法得到标准的Micro QR二维码图像;
步骤5.2、最大类间方差阈值法是一种自适应的阈值确定方法。算法假设图像像素能够根据阈值,被分成背景和目标两部分。然后,计算该最佳阈值来区分这两类像素,使得两类像素区分度最大。鉴于码区图像为黑白分明的编码图像,采用最大类间方差阈值法最为合适,具体过程为,按照0到255个灰度值遍历,计算并找到前景和背景最大类间方差variance=w0×w1×(u0-u1)×(u0-u1),其中,w0为分开后前景像素点数占图像的比例;u0为分开后前景像素点的平均灰度;w1为分开后背景像素点数占图像的比例;u1为分开后背景像素点的平均灰度,当找到最大方差时停止遍历,此时的T为分割阈值,根据分隔阈值进行二值化处理,得到精确定位后的Micro QR二维码图像。其中,在处理码区图像时,为避免边缘对于最大类间方差阈值法阈值产生影响,获取的码区图像特意选择超出码区边缘上下左右各2pixel,即满足国际标准的MicroQR二维码区域的尺寸。
上述说明示出并描述了本发明的若干优选实施方式,但如前所述,应当理解本发明并非局限于本文所披露的形式,不应看作是对其他实施方式的排除,而可用于各种其他组合、修改和环境,并能够在本文所述发明构想范围内,通过上述教导或相关领域的技术或知识进行改动。而本领域人员所进行的改动和变化不脱离本发明的精神和范围,则都应在本发明所附权利要求的保护范围内。

Claims (10)

  1. 一种Micro QR二维码的定位方法,其特征在于,包括以下步骤:
    步骤1、对Micro QR二维码图像进行边缘检测,获取连通域后,再根据连通域提取所有轮廓;
    步骤2、从步骤1中得到的所有轮廓中挑选出矩形连通域,确定矩形定位块中心;
    步骤3、根据二维码定位块的比例关系,计算出定位块中心的精确坐标;
    步骤4、以定位块的中心点为起点,分别沿时钟区的水平和垂直方向获取码区的水平方向的边界点和垂直方向的边界点,并根据这两个边界点和定位块的中心点,确定Micro QR二维码的码区边界。
  2. 根据权利要求1中所述的Micro QR二维码的定位方法,其特征在于,所述步骤1包括以下步骤:
    步骤1.1、滤除噪声;
    步骤1.2、计算图像中每个像素点的梯度强度和方向;
    步骤1.3、使用非极大值抑制,消除边缘检测带来的杂散响应;
    步骤1.4、使用双阈值检测确定真实的和潜在的边缘;
    步骤1.5、通过抑制孤立的弱边缘完成边缘检测并提取所有轮廓,其中,轮廓的集合为{L1,L2,...,Ln},每个轮廓Li由点列{A0,A1,...,An}构成。
  3. 根据权利要求1中所述的Micro QR二维码的定位方法,其特征在于,所述步骤2具体为在获取的全部轮廓中筛选出矩形封闭的轮廓,具体包括以下步骤:
    步骤2.1、遍历轮廓集合{L1,L2,...,Ln},计算每个轮廓Li的始点A0与终点An的距离,记为Distance,若Distance<D1 pixel,则该轮廓为封闭的轮廓,其中,D1为矩形判断参数,其大小由Micro QR二维码的尺寸确定;
    步骤2.2、根据矩形的性质,计算提取出的轮廓的面积与轮廓长度,再去除不满足比例关系式
    Figure PCTCN2021081732-appb-100001
    的轮廓,其中,Area为轮 廓的面积,Length为轮廓的长度,D2为接近于0的值,具体数值大小根据Micro QR二维码的模糊程度确定。
  4. 根据权利要求3中所述的Micro QR二维码的定位方法,其特征在于,所述步骤3具体包括以下步骤:
    步骤3.1、遍历所述步骤2中获取的所有矩形封闭轮廓;
    步骤3.2、提取矩形封闭轮廓的质心(X,Y);
    步骤3.3、计算矩形封闭轮廓与水平方向的夹角,记为Angle;
    步骤3.4、以质心(X,Y)的横坐标X为起点,根据定位块1:1:3:1:1的关系沿横向遍历图像,得到质心校正点的横坐标X0;
    步骤3.5、以质心校正点的横坐标X0为起点,纵向遍历图像,得到中心的校正点纵坐标Y0;
    步骤3.6、矩形框中心的精确位置为StartPoint(X0,Y0)。
  5. 根据权利要求4中所述的Micro QR二维码的定位方法,其特征在于,所述步骤4中水平方向的边界点获取具体包括以下步骤:
    步骤4.1a、从矩形框中心的精确位置StartPoint(X0,Y0)出发,沿夹角Angle的-90°方向出发,即沿逆时针方向旋转90度,与码区的边缘相交的三点分别记为A1,A2,A3;
    步骤4.2a、取新的起点,记为NewStartPoint(X,Y),其坐标设置为NewStartPoint=0.5×(A2+A3),再从NewStartPoint(X,Y)出发,转向90°,沿夹角Angle遍历整幅图像,与码区边缘相交,交点依次记为B1,B2,B3...Bn;
    步骤4.3a、将B1和B2的长度B1B2设置为基准,并设置pattern X=0,计算B2到B4的长度,记为B2B4,若
    Figure PCTCN2021081732-appb-100002
    成立,记pattern X=pattern X+1,并继续计算后续点列,直到不满足上述关系式,停止遍历,得到水平方向的边界点坐标EngPointX,其中,D3为接近于0的值,具体数值大小根据Micro QR二维码的模糊程度确定。
  6. 根据权利要求5中所述的Micro QR二维码的定位方法,其特征在于,所述步骤4中垂直方向的边界点获取具体包括以下步骤:
    步骤4.1b、从矩形框中心的精确位置StartPoint(X0,Y0)出发,沿夹角Angle的180°方向出发,与码区的边缘相交的三点分别记为A1,A2,A3;
    步骤4.2b、取新的起点,记为NewStartPoint(X,Y),其坐标设置为NewStartPoint=0.5×(A2+A3),再从NewStartPoint(X,Y)出发,转向90°,沿夹角Angle的90°方向遍历整幅图像,与码区边缘相交,交点依次记为B1,B2,B3...Bn;
    步骤4.3b、将B1和B2的长度B1B2设置为基准,并设置pattern Y=0,计算B2到B4的长度,记为B2B4,若
    Figure PCTCN2021081732-appb-100003
    成立,记pattern Y=pattern Y+1,并继续计算后续点列,直到不满足上述关系式,停止遍历,最后得到垂直方向的边界点坐标EngPointY,其中,D3为接近于0的值,具体数值大小根据Micro QR二维码的模糊程度确定。
  7. 根据权利要求5中所述的Micro QR二维码的定位方法,其特征在于:在获取水平方向的边界点和垂直方向的边界点后,分别计算垂直方向的边界点与定位块中心的距离以及水平方向的边界点与定位块中心的距离,若pattern X=pattern Y,则表示水平方向的边界点和垂直方向的边界点是正确的,继续执行下一步骤;否则继续遍历下一个矩形框。
  8. 根据权利要求1中所述的Micro QR二维码的定位方法,其特征在于:所述步骤1在对Micro QR二维码图像进行边缘检测前,还包括对图像的预处理,所述预处理包括以下步骤:
    步骤0.1、读取Micro QR二维码的灰度图像;
    步骤0.2、对输入的图像采用非线性滤波器终止滤波进行图像平滑处理,其二维中值滤波输出为:
    g(x,y)=med{f(x-k,y-l),(k,l∈W)},
    其中,f(x,y),g(x,y)分别为原始图像和处理后的图像,W为二维模板。
  9. 根据权利要求1中所述的Micro QR二维码的定位方法,其特征在于, 所述步骤4之后还包括对二维码图像的后处理,所述后处理包括对二维码图像进行校正,再经过二值化处理后得到精确定位后的Micro QR二维码图像。
  10. 根据权利要求9中所述的Micro QR二维码的定位方法,其特征在于,所述后处理中若图像为倾斜状态,则按照以下步骤进行校正:
    步骤5.1、将Micro QR二维码的定位块中心坐标和水平方向的边界点EngPointX坐标以及垂直方向的边界点EngPointY坐标映射到标准QR二维码的三个定位块的中心,它们满足以下关系式:
    Figure PCTCN2021081732-appb-100004
    通过求解得到图像旋转校正系数a11、a12、a21、a22,将原始的Micro QR二维码按照仿射变换旋转后,利用双线性插值法得到标准的Micro QR二维码图像;
    步骤5.2、将Micro QR二维码图像像素分为背景和目标两部分,按照0到255个灰度值遍历,计算并找到前景和背景最大类间方差variance=w0×w1×(u0-u1)×(u0-u1),其中,w0为分开后前景像素点数占图像的比例;u0为分开后前景像素点的平均灰度;w1为分开后背景像素点数占图像的比例;u1为分开后背景像素点的平均灰度,当找到最大方差时停止遍历,此时的T为分割阈值,根据分隔阈值进行二值化处理,得到精确定位后的Micro QR二维码图像。
PCT/CN2021/081732 2020-07-07 2021-03-19 一种Micro QR二维码的定位方法 WO2022007431A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010643510.6A CN111931538B (zh) 2020-07-07 2020-07-07 一种Micro QR二维码的定位方法
CN202010643510.6 2020-07-07

Publications (1)

Publication Number Publication Date
WO2022007431A1 true WO2022007431A1 (zh) 2022-01-13

Family

ID=73313988

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/081732 WO2022007431A1 (zh) 2020-07-07 2021-03-19 一种Micro QR二维码的定位方法

Country Status (2)

Country Link
CN (1) CN111931538B (zh)
WO (1) WO2022007431A1 (zh)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114492496A (zh) * 2022-01-20 2022-05-13 三维码(厦门)网络科技有限公司 一种点阵码快速识读方法、设备和存储介质
CN114549569A (zh) * 2022-03-01 2022-05-27 中国电建集团中南勘测设计研究院有限公司 一种基于二维正射影像的房屋轮廓提取方法
CN114564978A (zh) * 2022-04-27 2022-05-31 北京紫光青藤微系统有限公司 用于二维码解码的方法及装置、电子设备、存储介质
CN114638849A (zh) * 2022-05-19 2022-06-17 山东亚历山大智能科技有限公司 Dm二维码识别方法、agv定位方法、系统及agv
CN115320258A (zh) * 2022-08-09 2022-11-11 深圳市国人光速科技有限公司 线路板标识字符和可变动态二维码同时喷印的方法及系统
CN116086896A (zh) * 2023-04-06 2023-05-09 国网湖北省电力有限公司电力科学研究院 一种适用于变压器取油机器人的标准化取油装置及方法
CN117237442A (zh) * 2023-11-16 2023-12-15 宁德时代新能源科技股份有限公司 连通域定位方法、图形处理器、设备和生产线
CN117694107A (zh) * 2023-12-28 2024-03-15 广东若铂智能机器人有限公司 一种扰动下串型水果用机器人智能避障采摘方法

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111931538B (zh) * 2020-07-07 2024-03-29 广东奥普特科技股份有限公司 一种Micro QR二维码的定位方法
CN112800798B (zh) * 2021-01-15 2023-11-03 广东奥普特科技股份有限公司 一种Aztec码的定位方法
CN112926350B (zh) * 2021-02-25 2023-02-28 中国人民解放军32181部队 一种批量识别二维码的方法及终端
CN113688649B (zh) * 2021-08-16 2024-09-13 江苏博赛孚医疗科技有限公司 一种快速的qr码定位方法
CN113984761B (zh) * 2021-10-14 2023-07-21 上海原能细胞生物低温设备有限公司 一种二维码批扫设备的二维码快速筛选方法
CN117314951B (zh) * 2023-11-20 2024-01-26 四川数盾科技有限公司 一种二维码识别预处理方法及系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103294980A (zh) * 2013-06-18 2013-09-11 四川大学 基于图像处理的微型qr码识别方法
CN103413129A (zh) * 2012-06-29 2013-11-27 信实数码股份有限公司 图样取像范围的设定结构
US20150090793A1 (en) * 2013-09-29 2015-04-02 Founder Mobile Media Technology (Beijing) Co., Ltd. Method and system for determining edge line in qr code binary image
WO2017036264A1 (zh) * 2015-09-02 2017-03-09 福建联迪商用设备有限公司 一种二维码初步定位方法及系统
CN109902527A (zh) * 2019-02-26 2019-06-18 凌云光技术集团有限责任公司 一种qr码自动识读方法和装置
CN111931538A (zh) * 2020-07-07 2020-11-13 广东奥普特科技股份有限公司 一种Micro QR二维码的定位方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI506563B (zh) * 2013-01-28 2015-11-01 Tencent Tech Shenzhen Co Ltd A method and apparatus for enhancing reality of two - dimensional code
CN105260693B (zh) * 2015-12-01 2017-12-08 浙江工业大学 一种激光二维码定位方法
CN107066914A (zh) * 2017-03-13 2017-08-18 广州卓德信息科技有限公司 一种复杂背景下的二维条码图像定位方法及系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103413129A (zh) * 2012-06-29 2013-11-27 信实数码股份有限公司 图样取像范围的设定结构
CN103294980A (zh) * 2013-06-18 2013-09-11 四川大学 基于图像处理的微型qr码识别方法
US20150090793A1 (en) * 2013-09-29 2015-04-02 Founder Mobile Media Technology (Beijing) Co., Ltd. Method and system for determining edge line in qr code binary image
WO2017036264A1 (zh) * 2015-09-02 2017-03-09 福建联迪商用设备有限公司 一种二维码初步定位方法及系统
CN109902527A (zh) * 2019-02-26 2019-06-18 凌云光技术集团有限责任公司 一种qr码自动识读方法和装置
CN111931538A (zh) * 2020-07-07 2020-11-13 广东奥普特科技股份有限公司 一种Micro QR二维码的定位方法

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114492496A (zh) * 2022-01-20 2022-05-13 三维码(厦门)网络科技有限公司 一种点阵码快速识读方法、设备和存储介质
CN114492496B (zh) * 2022-01-20 2023-04-14 三维码(厦门)网络科技有限公司 一种点阵码快速识读方法、设备和存储介质
CN114549569A (zh) * 2022-03-01 2022-05-27 中国电建集团中南勘测设计研究院有限公司 一种基于二维正射影像的房屋轮廓提取方法
CN114549569B (zh) * 2022-03-01 2024-04-09 中国电建集团中南勘测设计研究院有限公司 一种基于二维正射影像的房屋轮廓提取方法
CN114564978A (zh) * 2022-04-27 2022-05-31 北京紫光青藤微系统有限公司 用于二维码解码的方法及装置、电子设备、存储介质
CN114564978B (zh) * 2022-04-27 2022-07-15 北京紫光青藤微系统有限公司 用于二维码解码的方法及装置、电子设备、存储介质
CN114638849A (zh) * 2022-05-19 2022-06-17 山东亚历山大智能科技有限公司 Dm二维码识别方法、agv定位方法、系统及agv
CN115320258A (zh) * 2022-08-09 2022-11-11 深圳市国人光速科技有限公司 线路板标识字符和可变动态二维码同时喷印的方法及系统
CN116086896A (zh) * 2023-04-06 2023-05-09 国网湖北省电力有限公司电力科学研究院 一种适用于变压器取油机器人的标准化取油装置及方法
CN117237442A (zh) * 2023-11-16 2023-12-15 宁德时代新能源科技股份有限公司 连通域定位方法、图形处理器、设备和生产线
CN117237442B (zh) * 2023-11-16 2024-04-09 宁德时代新能源科技股份有限公司 连通域定位方法、图形处理器、设备和生产线
CN117694107A (zh) * 2023-12-28 2024-03-15 广东若铂智能机器人有限公司 一种扰动下串型水果用机器人智能避障采摘方法

Also Published As

Publication number Publication date
CN111931538B (zh) 2024-03-29
CN111931538A (zh) 2020-11-13

Similar Documents

Publication Publication Date Title
WO2022007431A1 (zh) 一种Micro QR二维码的定位方法
CN110349207B (zh) 一种复杂环境下的视觉定位方法
TWI655586B (zh) 用於檢測預定區域中特定標識圖像的方法及裝置
CN110569857B (zh) 一种基于质心距离计算的图像轮廓角点检测方法
CN107633253B (zh) 基于含噪图像中矩形包围框的精确提取和定位方法
CN112233116B (zh) 基于邻域决策与灰度共生矩阵描述的凹凸痕视觉检测方法
Lo et al. Joint trilateral filtering for depth map super-resolution
WO2021109697A1 (zh) 字符分割方法、装置以及计算机可读存储介质
CN104933434A (zh) 一种结合LBP特征提取和surf特征提取方法的图像匹配方法
CN116993731B (zh) 基于图像的盾构机刀头缺陷检测方法
CN111369570B (zh) 一种视频图像的多目标检测跟踪方法
CN112734729B (zh) 适用于夜间补光条件的水尺水位线图像检测方法、装置及存储介质
WO2017193414A1 (zh) 一种基于转弯半径的图像角点检测方法
CN110472521B (zh) 一种瞳孔定位校准方法及系统
CN112801141B (zh) 基于模板匹配和孪生神经网络优化的异源图像匹配方法
CN105787912B (zh) 一种基于分类的阶跃型边缘亚像素定位方法
CN113888456B (zh) 一种基于轮廓的角点检测方法
CN111524139B (zh) 一种基于双边滤波器的角点检测方法和检测系统
CN107808165B (zh) 一种基于susan角点检测的红外图像匹配方法
CN106778766B (zh) 一种基于定位点的旋转数字识别方法及系统
CN103914829A (zh) 一种含噪图像边缘检测方法
CN106815851A (zh) 一种基于视觉测量的栅格圆形油位计自动读数方法
CN109389595B (zh) 一种表格线交点检测方法、电子设备及可读存储介质
CN110276260B (zh) 一种基于深度摄像头的商品检测方法
CN116563582A (zh) 基于国产CPU和opencv的图像模板匹配方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21838294

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 13.06.2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21838294

Country of ref document: EP

Kind code of ref document: A1