WO2017016448A1 - 一种qr码特征检测方法及系统 - Google Patents

一种qr码特征检测方法及系统 Download PDF

Info

Publication number
WO2017016448A1
WO2017016448A1 PCT/CN2016/091006 CN2016091006W WO2017016448A1 WO 2017016448 A1 WO2017016448 A1 WO 2017016448A1 CN 2016091006 W CN2016091006 W CN 2016091006W WO 2017016448 A1 WO2017016448 A1 WO 2017016448A1
Authority
WO
WIPO (PCT)
Prior art keywords
line
image
rectangular image
column
threshold
Prior art date
Application number
PCT/CN2016/091006
Other languages
English (en)
French (fr)
Inventor
蒋声障
吴卫东
Original Assignee
福建联迪商用设备有限公司
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 福建联迪商用设备有限公司 filed Critical 福建联迪商用设备有限公司
Priority to EP16829814.9A priority Critical patent/EP3309705B1/en
Priority to BR112018001703-5A priority patent/BR112018001703A2/pt
Publication of WO2017016448A1 publication Critical patent/WO2017016448A1/zh
Priority to US15/871,912 priority patent/US10102408B2/en

Links

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
    • 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
    • G06K7/12Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using a selected wavelength, e.g. to sense red marks and ignore blue marks

Definitions

  • the present invention relates to the field of two-dimensional code identification, and in particular, to a QR code feature detection method and system.
  • Two-dimensional code feature detection is the most important first step in the processing of two-dimensional code data. Once the detection fails, the processing cannot be continued.
  • the amount of two-dimensional code data is generally large, and it is possible to quickly detect the characteristics of the two-dimensional code. Affecting the use performance, the rapid and accurate detection of the two-dimensional code features has become an important requirement for the processing of two-dimensional code data.
  • the QR code is a type of two-dimensional code, and there are three position detection patterns of the QR code, which are respectively located at the upper left corner, the upper right corner, and the lower left corner of the QR code.
  • Each position detection pattern can be thought of as consisting of three overlapping concentric squares, which are 7*7 dark blocks, 5*5 light blocks, and 3*3 dark blocks. Therefore, the black and white length ratio of the line segment passing through the position detection pattern center block satisfies 1:1:3:1:1.
  • the Chinese invention patent of CN104517090A discloses a method and system for detecting a QR code detection pattern. The steps are as follows: detecting the QR code along the first direction, detecting black: white: black: white: black The length ratio of the first line segment satisfying the preset ratio, determining the center point of the first line segment; detecting the straight line passing through the center point of the first line segment along the second direction, obtaining black: white: black : White: The black length ratio satisfies the preset second line segment, and the center point of the second line segment is the center point of the detection pattern.
  • the detection of the detection pattern of the QR code image can only find its center point, and the boundary line of the detection image of the complete QR code cannot be found. Especially for the case where the QR code is deformed and twisted, the boundary of the detected pattern cannot be accurately determined.
  • the technical problem to be solved by the present invention is to provide a QR code feature detecting method and system for improving the detection speed and reducing the amount of data calculation.
  • a QR code feature detecting method the method is:
  • a preset line number threshold dividing the received QR code image into a plurality of lines of images having the same number of lines as the line number threshold
  • the black and white demarcation points are connected to form a position detection pattern of the QR code.
  • the beneficial effects of the present invention are: processing the line data of the QR code, detecting the multi-line image of the received QR code image, processing the column data of the QR code, processing the middle column data of the rectangular image , reducing the amount of data calculation, improving the detection efficiency; initially finding the position of the position detection pattern of the QR code on the received QR image by using the line data and the preset proportional similarity in the line proportional threshold, by searching for the rectangle
  • a column line segment matching the preset scale on the middle column of the image is used to determine the position of the position detection pattern of the QR code on the received QR image, by panning a diagonal line of the square image and finding a black and white boundary point of the square image edge to determine
  • the position of the QR code detects the boundary line of the graphic. It is not affected by image distortion, torsion, and light. It can quickly and accurately find the area of the QR code position detection pattern.
  • a QR code feature detection system comprising:
  • a dividing module configured to divide the received QR code image into rows having the same pixel point as the line number threshold a number of lines of images
  • a calculation module configured to calculate an average gray value of each column of the plurality of lines of pixels to form row data
  • a first binarization module configured to binarize the row data
  • a statistics module configured to calculate a ratio of consecutive 1 and consecutive 0 in the row data after binarization to form a line ratio
  • an extraction module configured to extract a rectangular image corresponding to a row ratio of the similarity of the preset ratio in the row proportional threshold
  • a second binarization module that binarizes a rectangular image and a region range of the rectangular image up and down extending width thresholds
  • a first searching module configured to search for a column line segment that meets a preset ratio on a middle column of the extracted rectangular image
  • a panning module configured to translate a diagonal line of the square image including the rectangular image and the column line segment in two directions of upward and downward directions;
  • a second searching module configured to find the square image edge black and white boundary point on the diagonal of the translation
  • a recording module configured to record black and white boundary of the square image edge on the diagonal of the translation Point
  • a component module configured to connect the black and white boundary points to form a position detection pattern of the QR code.
  • the beneficial effects of the present invention are: dividing the received QR code image into at least one multi-line image having the same number of lines as the pixel of the preset line number threshold, and then passing the multi-line image Perform binarization, statistical line ratio, comparison of line scale and preset ratio similarity, record rectangular image with similarity to the requirement, compare the middle column of the rectangular image with the preset ratio, and then search for the square image through the diagonal
  • the black and white boundary point of the edge determines the position detection pattern of the received QR code image, and simultaneously processes the data of the multi-line image.
  • the processing of the column data is only one line, and the position of the boundary line of the position detection pattern can be determined without being deformed by the image.
  • the effects of torsion and lighting can effectively reduce the amount of data calculation in the system and improve the efficiency and accuracy of system testing.
  • DRAWINGS 1 is a flowchart of a QR code feature detecting method according to Embodiment 1 of the present invention
  • FIG. 2 is a partial image of a QR code including a position detection pattern according to Embodiment 1 of the present invention
  • FIG. 3 is a schematic diagram of dividing a plurality of lines of images in FIG. 2 according to Embodiment 1 of the present invention
  • FIG. 4 is a schematic diagram of a multi-line image divided in FIG. 2 according to Embodiment 1 of the present invention.
  • FIG. 5 is a schematic diagram of a middle column of a rectangular pattern in a multi-line image according to Embodiment 1 of the present invention.
  • FIG. 6 is a schematic diagram of a square image and a diagonal line thereof according to Embodiment 1 of the present invention. detailed description
  • the most critical idea of the present invention is to: simultaneously perform data on data of a plurality of lines of images of the received QR code image, reduce the amount of data calculation, and improve detection efficiency.
  • a specific embodiment of the invention is:
  • QR code feature detection method where the method is:
  • a preset line number threshold dividing the received QR code image into a plurality of lines of images having the same number of lines as the line number threshold
  • the line number threshold is greater than 1 and less than or equal to the number of pixel points corresponding to the position detection pattern of the QR code.
  • the threshold of the number of rows is greater than 1, the image of two or more rows can be processed at one time, and the detection efficiency is improved; the number of rows of pixels corresponding to the position detection pattern of the threshold of the QR code is less than or equal to the QR code. It can effectively ensure that the image of the position detection pattern can be accurately found in the multi-line image processed, and the detection accuracy is improved.
  • row data is binarized, specifically:
  • calculating the gray threshold of the line data by the maximum inter-class variance method can reduce the probability of foreground color and background color misclassification, and improve the accuracy of line data binarization.
  • the rectangular image and the rectangular image are binarized in the range of the width and threshold values of the extended width thresholds, specifically:
  • the data in the area range is binarized according to the gray threshold value, which facilitates processing of subsequent data, and is advantageous for maintaining consistency of image binarization processing, and improving The accuracy of the value.
  • a preset line proportional threshold comparing whether the similarity between the suspected line ratio and the preset ratio is within a line proportional threshold
  • the ratio of the suspected lines having the same starting data binarization value and the same number of data in the line proportion of the plurality of lines of images is first searched, and then the proportion of the suspected lines and the preset ratio are compared. Whether the similarity is in the range of the proportional threshold, and finally extracting the rectangular image corresponding to the proportion of the line satisfying the above conditions, A part of the position detection pattern in which the rectangular image is a QR code, the width of the rectangular image is the width of the position detection pattern of the QR code, and the data processing process is reasonable, and the position of the column of the position detection pattern can be quickly found.
  • comparing whether the similarity between the suspected line ratio and the preset ratio is within a proportional threshold specifically:
  • the sum of the width threshold of 2 times and the preset line number threshold is greater than the side length of the QR code position detection pattern.
  • the sum of the width threshold and the preset line number threshold is greater than the side length of the QR code position detection pattern, and the width of the area range is greater than the edge of the QR code position detection pattern. Long, effective to ensure fast, accurate and comprehensive search for position detection graphics.
  • the column line segment corresponding to the preset ratio on the middle column of the rectangular image is found to find the position of the column of the position detection pattern of the QR code, the amount of data processing information is small, and the detection rate is fast.
  • a diagonal line including the rectangular image and the square image of the column line segment is translated in two directions of upward and downward directions respectively; searching for and recording the square on the diagonal of the translation.
  • the coordinates of the recorded white point and black point are the black and white boundary points of the edge of the position detection pattern.
  • a square image can be formed according to the column position where the width of the rectangular image is located and the row position where the end point of the column line segment is located, and the diagonal direction of the direction image is moved up and down to find the moved image.
  • the black and white boundary point of the edge of the corresponding detection position on the diagonal line is used to determine the boundary line of the position detection pattern of the QR code, which is not affected by the image deformation, torsion, and light, and the data accuracy is high.
  • the first embodiment of the present invention is:
  • a QR code feature detecting method the method is:
  • the preset line number threshold is 10, and the received QR code image is divided into multiple lines of images having the same number of lines as the line number threshold; see FIG. 2
  • an image of a position detection pattern is included, wherein the pixel points are 28 lines in total
  • FIG. 3 is a division diagram of the position detection pattern in FIG. 2 divided into 10 lines of pixel point data
  • FIG. 4 is divided into A multi-line image containing 10 rows of pixel data:
  • a preset line proportional threshold and extracting a rectangular image corresponding to a line ratio of the similarity of the preset ratio to the preset ratio in the line proportional threshold, specifically:
  • the preset line proportional threshold is 0.3, and the ratio of the suspected lines having the same starting data binarization value and the same number of data in the line proportion is extracted, and the following are: (1) 6:5:17:5:6; (2) 17:5:6:22:7; (3) 6:22:7:5:11; (4) 7:5:11:11:11; 5) 11:11:11:5:6;
  • the rectangular image and the area range of the rectangular image up and down extending width thresholds are binarized, specifically:
  • [0114] preset a translation distance, and translate the diagonal lines in two directions of upward and downward directions, respectively;
  • the coordinates of the recorded white point and the black point are the edge black and white boundary points of the position detection pattern of the QR code; [0117] The black and white boundary points are connected to form a position detection pattern of the QR code.
  • a QR code feature detection system comprising:
  • a dividing module configured to divide the received QR code image into a plurality of lines of images having the same number of lines as the line number threshold;
  • a calculation module configured to calculate an average gray value of each column of pixels in the plurality of lines of images to form row data
  • a first binarization module configured to perform binarization on the row data
  • a statistics module configured to calculate a ratio of consecutive 1s and consecutive 0s in the row data after binarization to form a line ratio
  • an extraction module configured to extract a rectangular image corresponding to a row ratio of the similarity of the preset ratio in the row proportional threshold
  • a second binarization module that binarizes the rectangular image and the area range of the rectangular image up and down each extending width threshold
  • a first searching module configured to search for a column line segment that meets a preset ratio on a middle column of the extracted rectangular image
  • a panning module configured to translate a diagonal line of the square image including the rectangular image and the column line segment in two directions of upward and downward directions;
  • a second searching module configured to find the square image edge black and white boundary point on the diagonal of the translation
  • a recording module configured to record black and white boundary of the square image edge on the diagonal of the translation Point
  • the component module is configured to connect the black and white boundary points to form a position detection pattern of the QR code.
  • Embodiment 2 of the present invention is:
  • a QR code feature detection system comprising:
  • a dividing module configured to divide the received QR code image into a plurality of lines of images having the same number of lines as the line number threshold;
  • a calculation module configured to calculate an average gray value of each column of the plurality of lines of pixels to form row data
  • a first binarization module configured to binarize the row data
  • a statistics module configured to calculate a ratio of consecutive 1 and consecutive 0 in the row data after binarization to form a line ratio
  • an extraction module configured to extract a rectangular image corresponding to a row ratio of the similarity of the preset ratio in the row proportional threshold
  • a second binarization module that binarizes the rectangular image and the area range of the rectangular image up and down each extending width threshold
  • a first searching module configured to search for a column line segment that meets a preset ratio on a middle column of the extracted rectangular image
  • a panning module configured to translate a diagonal line of the square image including the rectangular image and the column line segment in upward and downward directions, respectively;
  • a second searching module configured to find the square image edge black and white boundary point on the diagonal of the translation
  • a recording module configured to record black and white boundary of the square image edge on the diagonal of the translation Point
  • a component module configured to connect the black and white boundary points to form a position detection pattern of the QR code.
  • the QR code feature detection method and system provided by the present invention first detects a plurality of lines of the received QR code image to find a partial rectangular image in the position detection pattern, and then Find the position of the column of the position detection pattern on the middle column, and then find the boundary line of the position detection image.

Landscapes

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

Abstract

一种QR码特征检测方法及系统,所述方法为:将接收到的QR码图像划分为与预设行数阀值具有相同像素点的行数的多行图像;计算多行图像中每列像素点的平均灰度值形成行数据,并进行二值化;统计二值化后的行数据中与预设比例的相似度在预设行比例阀值内的矩形图像;将所述矩形图像以及其向上和向下各延伸预设宽度阀值的区域范围进行二值化;寻找矩形图像的中间列上符合预设比例的列线段;将包含所述矩形图像与所述列线段的方形图像的一条对角线分别沿向上和向下两个方向进行平移;寻找并记录平移的对角线上的所述方形图像边缘黑白分界点组成QR码的位置探测图形。可快速找到接收到的QR码的位置探测图形,减少系统数据运算量。

Description

发明名称:一种 QR码特征检测方法及系统
技术领域
[0001] 本发明涉及二维码识别领域, 特别涉及一种 QR码特征检测方法及系统。
背景技术
[0002] 二维码特征检测是二维码数据处理中最关键的首要步骤, 一旦检测失败则无法 继续进行处理, 而二维码数据量一般都很大, 能否快速检测二维码特征直接影 响了使用性能, 因此二维码特征的快速、 精准检测成了二维码数据处理的重要 需求。
[0003] QR码是二维码的一种, QR码的位置探测图形有三个, 分别位于 QR码的左上角 、 右上角和左下角。 每个位置探测图形可以看作是由 3个重叠的同心的正方形组 成, 它们分别为 7*7个深色模块、 5*5个浅色模块和 3*3个深色模块。 故经过位置 探测图形中心分块的线段黑白长度比例满足 1:1:3:1:1。
[0004] 目前多数二维码特征检测方法采用逐行进行扫描、 检测, 这种方法的检测速度 比较慢。
[0005] 公幵号为 CN104517090A的中国发明专利公幵了一种 QR码探测图形的检测方法 及系统, 步骤为: 沿第一方向对 QR码进行检测, 检测黑: 白: 黑: 白: 黑的长 度比例满足预设比例的第一线段, 确定所述第一线段的中心点; 沿第二方向对 经过所述第一线段的中心点的直线进行检测, 得到黑: 白: 黑: 白: 黑的长度 比例满足预设比例的第二线段, 所述第二线段的中心点为探测图形的中心点。 虽然只需要沿第一方向检测吋进行逐行遍历检测, 提高了检测速度, 但是对 QR 码图像的探测图形的检测仅能找到其中心点, 不能找到完整的 QR码的检测图像 的边界线, 特别是对于 QR码发生变形和扭转的情况不能准确确定其探测图形的 边界。
技术问题
[0006] 本发明所要解决的技术问题是: 提供一种提高检测速度、 减少数据运算量的 Q R码特征检测方法及系统。 问题的解决方案
技术解决方案
[0007] 为了解决上述技术问题, 本发明采用的技术方案为:
[0008] 一种 QR码特征检测方法, 所述方法为:
[0009] 预设行数阀值, 将接收到的 QR码图像划分为与行数阀值具有相同像素点的行 数的多行图像;
[0010] 计算所述的多行图像中每列像素点的平均灰度值形成行数据;
[0011] 对所述行数据进行二值化;
[0012] 统计二值化后的行数据中连续的 1和连续的 0的个数组成行比例;
[0013] 预设行比例阀值, 提取所述行比例中与预设比例的相似度在行比例阀值内的行 比例所对应的矩形图像;
[0014] 预设宽度阀值, 将所述矩形图像以及矩形图像向上和向下各延伸宽度阀值的区 域范围进行二值化;
[0015] 寻找所述矩形图像的中间列上符合预设比例的列线段;
[0016] 将包含所述矩形图像与所述列线段的方形图像的一条对角线分别沿向上和向下 两个方向进行平移;
[0017] 寻找并记录平移的对角线上的所述方形图像边缘黑白分界点;
[0018] 将所述黑白分界点连接起来组成 QR码的位置探测图形。
[0019] 本发明的有益效果在于: 对 QR码的行数据处理吋, 同吋对接收到的 QR码图像 的多行图像进行检测, 对 QR码的列数据处理吋, 处理矩形图像中间列数据, 减 少数据计算量, 提高检测效率; 通过行数据与预设比例相似度在行比例阀值来 初步寻找到 QR码的位置探测图形在接收到的 QR图像上的列位置, 通过寻找所述 矩形图像的中间列上符合预设比例的列线段来确定 QR码的位置探测图形在接收 到的 QR图像上的行位置, 通过平移方形图像的一条对角线并寻找方形图像边缘 黑白分界点来确定 QR码的位置探测图形的边界线, 不受图像变形、 扭转、 灯光 的影响, 可以快速准确的找到 QR码的位置探测图形的区域。
[0020] 一种 QR码特征检测系统, 包括:
[0021] 划分模块, 用于将接收到的 QR码图像划分为与行数阀值具有相同像素点的行 数的多行图像;
[0022] 计算模块, 用于计算所述的多行图像中每列像素点的平均灰度值形成行数据; [0023] 第一二值化模块, 用于对所述行数据进行二值化;
[0024] 统计模块, 用于统计二值化后的行数据中连续的 1和连续的 0的个数组成行比例
[0025] 提取模块, 用于提取所述行比例中与预设比例的相似度在行比例阀值内的行比 例所对应的矩形图像;
[0026] 第二二值化模块, 将所述矩形图像以及矩形图像向上和向下各延伸宽度阀值的 区域范围进行二值化;
[0027] 第一寻找模块, 用于寻找提取的所述矩形图像的中间列上符合预设比例的列线 段;
[0028] 平移模块, 用于将包含所述矩形图像与所述列线段的方形图像的一条对角线分 别沿向上和向下两个方向进行平移;
[0029] 第二寻找模块, 用于寻找平移的对角线上的所述方形图像边缘黑白分界点; [0030] 记录模块, 用于记录平移的对角线上的所述方形图像边缘黑白分界点;
[0031 ] 组成模块, 用于将所述黑白分界点连接起来组成 QR码的位置探测图形。
发明的有益效果
有益效果
[0032] 本发明的有益效果在于: 将接收到的 QR码图像划分为至少一个的与预设的行 数阀值的像素点具有相同行数的多行图像后, 通过对所述多行图像进行二值化 、 统计行比例、 比较行比例与预设比例的相似度、 记录相似度符合要求的矩形 图像, 对矩形图像的中间列进一步与预设比例比较, 然后通过对角线寻找方形 图像边缘的黑白分界点来确定接收到的 QR码图像的位置探测图形, 同吋处理多 行图像的数据, 列数据的处理仅为一行, 且可确定出位置探测图形边界线位置 , 不受图像变形、 扭转、 灯光的影响, 可有效降低系统的数据计算量, 提高系 统检测的效率和准确性。
对附图的简要说明
附图说明 [0033] 图 1为本发明实施例一的 QR码特征检测方法流程图;
[0034] 图 2为本发明实施例一的包含位置探测图形的 QR码的部分图像;
[0035] 图 3为本发明实施例一的图 2划分多行图像的示意图;
[0036] 图 4为本发明实施例一的图 2中划分出的一个多行图像的示意图;
[0037] 图 5为本发明实施例一的多行图像中的矩形图形的中间列的示意图;
[0038] 图 6为本发明实施例一的方形图像及其一条对角线的示意图。 具体实施方式
[0039] 为详细说明本发明的技术内容、 所实现目的及效果, 以下结合实施方式并配合 附图予以说明。
[0040] 本发明最关键的构思在于: 同吋对接收到的 QR码图像的多行图像的数据进行 数据, 降低数据计算量, 提高检测效率。
[0041] 本发明的具体实施方式为:
[0042] 请参照图 1, 一种 QR码特征检测方法, 所述方法为:
[0043] 预设行数阀值, 将接收到的 QR码图像划分为与行数阀值具有相同像素点的行 数的多行图像;
[0044] 计算所述的多行图像中每列像素点的平均灰度值形成行数据;
[0045] 对所述行数据进行二值化;
[0046] 统计二值化后的行数据中连续的 1和连续的 0的个数组成行比例;
[0047] 预设行比例阀值, 提取所述行比例中与预设比例的相似度在行比例阀值内的行 比例所对应的矩形图像;
[0048] 预设宽度阀值, 将所述矩形图像以及矩形图像向上和向下各延伸宽度阀值的区 域范围进行二值化;
[0049] 寻找所述矩形图像的中间列上符合预设比例的列线段;
[0050] 将包含所述矩形图像与所述列线段的方形图像的一条对角线分别沿向上和向下 两个方向进行平移;
[0051] 寻找并记录平移的对角线上的所述方形图像边缘黑白分界点; [0053] 进一步的, 所述行数阀值大于 1且小于等于 QR码的位置探测图形对应的像素点 行数。
[0054] 由上述描述可知, 行数阀值大于 1, 则一次可处理两行或者两行以上的图像, 提高检测效率; 行数阀值小于等于 QR码的位置探测图形对应的像素点行数, 有 效保证所处理的多行图像中能较准确的找到位置探测图形的图像, 提高检测准 确性。
[0055] 进一步的, 对所述行数据进行二值化, 具体为:
[0056] 用最大类间方差法计算所述的行数据的灰度阀值;
[0057] 根据所述灰度阀值对所述的行数据进行二值化。
[0058] 由上述描述可知, 用最大类间方差法计算所述的行数据的灰度阀值可降低前景 色与背景色错分的概率, 提高行数据二值化的准确性。
[0059] 进一步的, 将所述矩形图像以及矩形图像向上和向下各延伸宽度阀值的区域范 围进行二值化, 具体为:
[0060] 根据所述灰度阀值对所述矩形图像以及矩形图像向上和向下各延伸宽度阀值的 区域范围进行二值化。
[0061] 由上述描述可知, 根据所述灰度阀值对所述区域范围内的数据进行二值化, 便 于对后续数据的处理, 且有利于保持图像二值化处理的一致性, 提高二值化的 准确性。
[0062] 进一步的, 提取所述行比例中与预设比例的相似度在行比例阀值内的行比例所 对应的矩形图像, 具体为:
[0063] 提取所述行比例中与预设比例的起始数据二值化值相同且数据个数相同的疑似 行比例;
[0064] 预设行比例阀值, 比较所述的疑似行比例与预设比例的相似度是否在行比例阀 值范围内;
[0065] 记录与预设比例相似度在行比例阀值范围内的疑似行比例所对应的矩形图像。
[0066] 由上述描述可知, 先寻找多行图像的行比例中与预设比例的起始数据二值化值 相同且数据个数相同的疑似行比例, 然后比较疑似行比例与预设比例的相似度 是否在行比例阀值范围, 最后提取满足上述条件的行比例对应的矩形图像, 得 到矩形图像为 QR码的位置探测图形的一部分, 矩形图像的宽度为 QR码的位置探 测图形的宽度, 数据处理过程合理, 可快速寻找到位置探测图形的列所在的位 置。
[0067] 进一步的, 比较所述的疑似行比例与预设比例的相似度是否在比例阀值范围内 , 具体为:
[0068] 将所述疑似行比例的起始数据转换为与预设比例起始数据相同的数据, 并将疑 似行比例的所有数据等比例转换得到转换后的疑似行比例;
[0069] 计算转换后的疑似行比例中除第一个数据外的其他数据与预设比例对应位上的 数据的差值并取绝对值;
[0070] 若每个位对应的绝对值均小于行比例阀值, 则所述的疑似行比例与预设比例的 相似度小于行比例阀值。
[0071] 由上述描述可知, 通过将所述疑似行比例的起始数据转换为与预设比例起始数 据相同的数据, 并将疑似行比例的所有数据等比例转换得到转换后的疑似行比 例, 然后比较转换后的疑似比例每个位与预设比例每个位的差异是否在比例阀 值范围内来确认相似度, 数据处理过程合理。
[0072] 进一步的, 所述宽度阀值的 2倍与预设行数阀值的和大于 QR码位置探测图形的 边长。
[0073] 由上述描述可知, 所述宽度阀值的 2倍与预设行数阀值的和大于 QR码位置探测 图形的边长, 保证所述区域范围的宽度大于 QR码位置探测图形的边长, 有效保 证快速准确全面的找到位置探测图形。
[0074] 进一步的, 寻找提取的所述矩形图像的中间列上符合预设比例的列线段, 具体 为:
[0075] 寻找所述矩形图像的中间列线段;
[0076] 将所述中间列线段分别向上和向下两个方向延伸宽度阀值组成新的中间列线段 [0077] 提取新的中间列线段中符合预设比例的列线段。
[0078] 由上述描述可知, 寻找矩形图像的中间列上的符合预设比例的列线段来找到 Q R码的位置探测图形的列所在的位置, 数据处理信息量小, 检测速率快。 [0079] 进一步的, 将包含所述矩形图像与所述列线段的方形图像的一条对角线分别沿 向上和向下两个方向进行平移; 寻找并记录平移的对角线上的所述方形图像边 缘黑白分界点, 具体为:
[0080] 形成包含所述矩形图像与所述列线段的方形图像;
[0081 ] 取所述方形图像的一条对角线;
[0082] 预设平移距离, 将所述对角线分别沿向上和向下两个方向进行平移;
[0083] 在平移的对角线上, 端点上的值若为黑点则向方形图形外侧移动该黑点直至遇 到白点并记录白点的坐标; 端点上的值若为白点则向方形图形内侧移动该白点 直至遇到黑点并记录黑点的坐标;
[0084] 记录的白点和黑点的坐标即为所述位置探测图形边缘的黑白分界点。
[0085] 由上述描述可知, 根据矩形图像的宽度所在的列位置与列线段的端点所在的行 位置可形成一个方形图像, 通过对方向图像的对角线向上和向下移动并寻找移 动后的对角线上对应的位置探测图形边缘的黑白分界点来确定 QR码的位置探测 图形的边界线, 不受图像变形、 扭转、 灯光的影响, 数据准确度高。
[0086]
[0087] 请参照图 1至图 6, 本发明的实施例一为:
[0088] 一种 QR码特征检测方法, 所述方法为:
[0089] 请参照图 1至图 4, 预设行数阀值为 10, 将接收到的 QR码图像划分为与行数阀 值具有相同像素点的行数的多行图像; 请参见图 2为接收到的 QR码图像中包含一 个位置探测图形的图像, 其中像素点共 28行, 图 3为图 2中的位置探测图形划分 为 10行像素点数据的划分示意图, 图 4为划分到的包含 10行像素点数据的一个多 行图像:
[0090] 计算所述的多行图像中每列像素点的平均灰度值形成行数据, 得到 251个灰度 值, 具体为:
Figure imgf000009_0001
,c
6,c6,c6,c6,cl,c2,c3,c3,c3,c3,ec,ff,ff,ff,ff,b7,39,39,39,39,39,39,94,c3,c3,c3,c3,b5,8e,89 ,dl,ff,f8,ff,ff,ff,ff, ff,ff,ff,ff,ff,ff,ff,ff,ff,ff,ff,ff,ff,ff,ff;
[0092] 对所述行数据进行二值化, 具体为:
[0093] 用最大类间方差法计算所述的行数据的灰度阀值为 0x9a;
[0094] 根据所述灰度阀值为 0x9a对所述的行数据进行二值化, 具体为:
Figure imgf000010_0001
[0096] 统计二值化后的行数据中连续的 1和连续的 0的个数组成行比例, 具体为:
[0097] 59:6:5:17:5:6:22:7:5:11:11:11:5:6:75;
[0098] 其中 59个连续的 0是指二维码图像中二维码之前的空白区域中 0的个数;
[0099] 预设行比例阀值, 提取所述行比例中与预设比例的相似度在行比例阀值内的行 比例所对应的矩形图像, 具体为:
[0100] 预设行比例阀值为 0.3, 提取所述行比例中与预设比例的起始数据二值化值相同 且数据个数相同的疑似行比例共 5组, 分别为: (1) 6:5:17:5:6; (2) 17:5:6:22: 7; (3) 6:22:7:5:11; (4) 7:5:11:11:11; (5) 11:11:11:5:6;
[0101] 比较所述的疑似行比例与预设比例的相似度是否在行比例阀值 0.3范围内, 前面 5组疑似行比例中很明显可以排除 (2) 、 (3) 、 (4) 、 (5) 的情况, 将第 ( 1) 组数据同吋除以 6得到 1:0.83:2.83:0.83:1;
[0102] 计算比例 1:0.83:2.83:0.83:1中后四位数据与标准行特征比例 1:1:3:1:1对应位上的 数据的差值并取绝对值, 分别为 0.17、 0.17、 0.17、 0, 这四个数据均小于行比例 阀值 0.3, 因此第二组疑似行比例与标准行特征比例相似度在行比例阀值范围内
[0103] 记录与预设比例相似度在行比例阀值范围内的疑似行比例所对应的矩形图像; 其中矩形图像所在的列即为位置探测图像所在的列; [0104] 请参照图 5, 预设宽度阀值为 n, n可取上述第 (1) 组数据中各像素点个数的和 , 即取 6+5+17+5+6=39, 将所述矩形图像以及矩形图像向上和向下各延伸宽度阀 值的区域范围进行二值化, 具体为:
[0105] 根据所述灰度阀值为 0x9a对所述矩形图像以及矩形图像向上和向下各延伸宽度 阀值的区域范围进行二值化;
[0106] 寻找所述矩形图像的中间列上符合预设比例的列线段, 具体为:
[0107] 寻找所述矩形图像的中间列线段;
[0108] 将所述中间列线段分别向上和向下两个方向延伸宽度阀值组成新的中间列线段
[0109] 提取新的中间列线段中符合预设比例的列线段; 其中列线段所在的行即为位置 探测图像所在的行;
[0110] 将包含所述矩形图像与所述列线段的方形图像的一条对角线分别沿向上和向下 两个方向进行平移;
[0111] 请参照图 6, 寻找并记录平移的对角线上的所述方形图像边缘黑白分界点, 具 体为:
[0112] 形成包含所述矩形图像与所述列线段的方形图像;
[0113] 取所述方形图像的一条对角线;
[0114] 预设平移距离, 将所述对角线分别沿向上和向下两个方向进行平移;
[0115] 在平移的对角线上, 端点上的值若为黑点则向方形图形外侧移动该黑点直至遇 到白点并记录白点的坐标; 端点上的值若为白点则向方形图形内侧移动该白点 直至遇到黑点并记录黑点的坐标;
[0116] 记录的白点和黑点的坐标即为所述 QR码的位置探测图形的边缘黑白分界点; [0117] 将所述黑白分界点连接起来组成 QR码的位置探测图形。
[0118]
[0119] 一种 QR码特征检测系统, 包括:
[0120] 划分模块, 用于将接收到的 QR码图像划分为与行数阀值具有相同像素点的行 数的多行图像;
[0121] 计算模块, 用于计算所述的多行图像中每列像素点的平均灰度值形成行数据; [0122] 第一二值化模块, 用于对所述行数据进行二值化;
[0123] 统计模块, 用于统计二值化后的行数据中连续的 1和连续的 0的个数组成行比例
[0124] 提取模块, 用于提取所述行比例中与预设比例的相似度在行比例阀值内的行比 例所对应的矩形图像;
[0125] 第二二值化模块, 将所述矩形图像以及矩形图像向上和向下各延伸宽度阀值的 区域范围进行二值化;
[0126] 第一寻找模块, 用于寻找提取的所述矩形图像的中间列上符合预设比例的列线 段;
[0127] 平移模块, 用于将包含所述矩形图像与所述列线段的方形图像的一条对角线分 别沿向上和向下两个方向进行平移;
[0128] 第二寻找模块, 用于寻找平移的对角线上的所述方形图像边缘黑白分界点; [0129] 记录模块, 用于记录平移的对角线上的所述方形图像边缘黑白分界点;
[0130] 组成模块, 用于将所述黑白分界点连接起来组成 QR码的位置探测图形。
[0131] 本发明的实施例二为:
[0132] 一种 QR码特征检测系统, 包括:
[0133] 划分模块, 用于将接收到的 QR码图像划分为与行数阀值具有相同像素点的行 数的多行图像;
[0134] 计算模块, 用于计算所述的多行图像中每列像素点的平均灰度值形成行数据; [0135] 第一二值化模块, 用于对所述行数据进行二值化;
[0136] 统计模块, 用于统计二值化后的行数据中连续的 1和连续的 0的个数组成行比例
[0137] 提取模块, 用于提取所述行比例中与预设比例的相似度在行比例阀值内的行比 例所对应的矩形图像;
[0138] 第二二值化模块, 将所述矩形图像以及矩形图像向上和向下各延伸宽度阀值的 区域范围进行二值化;
[0139] 第一寻找模块, 用于寻找提取的所述矩形图像的中间列上符合预设比例的列线 段; [0140] 平移模块, 用于将包含所述矩形图像与所述列线段的方形图像的一条对角线分 别沿向上和向下两个方向进行平移;
[0141] 第二寻找模块, 用于寻找平移的对角线上的所述方形图像边缘黑白分界点; [0142] 记录模块, 用于记录平移的对角线上的所述方形图像边缘黑白分界点;
[0143] 组成模块, 用于将所述黑白分界点连接起来组成 QR码的位置探测图形。
[0144]
[0145] 综上所述, 本发明提供的 QR码特征检测方法及系统, 先对接收到的 QR码图像 的多行图像进行检测找出位置探测图形中的部分矩形图像, 再对矩形图像的中 间列上寻找位置探测图形的列所在的位置, 然后寻找位置探测图像的边界线, 对多行图像和矩形图像的二值化采用同一灰度阀值, 采用最大类间方差法进行 二值化, 不受图像变形、 扭转、 灯光的影响, 可快速准确地找到接收到的 QR码 图像的位置探测图形, 有效降低系统的数据计算量, 提高系统检测的效率和准 确性。

Claims

权利要求书
一种 QR码特征检测方法, 其特征在于, 所述方法为:
预设行数阀值, 将接收到的 QR码图像划分为与行数阀值具有相同像 素点的行数的多行图像;
计算所述的多行图像中每列像素点的平均灰度值形成行数据; 对所述行数据进行二值化;
统计二值化后的行数据中连续的 1和连续的 0的个数组成行比例; 预设行比例阀值, 提取所述行比例中与预设比例的相似度在行比例阀 值内的行比例所对应的矩形图像;
预设宽度阀值, 将所述矩形图像以及矩形图像向上和向下各延伸宽度 阀值的区域范围进行二值化;
寻找所述矩形图像的中间列上符合预设比例的列线段;
将包含所述矩形图像与所述列线段的方形图像的一条对角线分别沿向 上和向下两个方向进行平移;
寻找并记录平移的对角线上的所述方形图像边缘黑白分界点; 将所述黑白分界点连接起来组成 QR码的位置探测图形。
根据权利要求 1所述的 QR码特征检测方法, 其特征在于, 所述行数阀 值大于 1且小于等于 QR码的位置探测图形对应的像素点行数。
根据权利要求 1所述的 QR码特征检测方法, 其特征在于, 对所述行数 据进行二值化, 具体为:
用最大类间方差法计算所述的行数据的灰度阀值;
根据所述灰度阀值对所述的行数据进行二值化。
根据权利要求 3所述的 QR码特征检测方法, 其特征在于, 将所述矩形 图像以及矩形图像向上和向下各延伸宽度阀值的区域范围进行二值化
, 具体为:
根据所述灰度阀值对所述矩形图像以及矩形图像向上和向下各延伸宽 度阀值的区域范围进行二值化。
根据权利要求 1所述的 QR码特征检测方法, 其特征在于, 提取所述行 比例中与预设比例的相似度在行比例阀值内的行比例所对应的矩形图 像, 具体为:
提取所述行比例中与预设比例的起始数据二值化值相同且数据个数相 同的疑似行比例;
预设行比例阀值, 比较所述的疑似行比例与预设比例的相似度是否在 行比例阀值范围内;
记录与预设比例相似度在行比例阀值范围内的疑似行比例所对应的矩 形图像。
[权利要求 6] 根据权利要求 5所述的 QR码特征检测方法, 其特征在于, 比较所述的 疑似行比例与预设比例的相似度是否在比例阀值范围内, 具体为: 将所述疑似行比例的起始数据转换为与预设比例起始数据相同的数 据, 并将疑似行比例的所有数据等比例转换得到转换后的疑似行比 例;
计算转换后的疑似行比例中除第一个数据外的其他数据与预设比例 对应位上的数据的差值并取绝对值;
若每个位对应的绝对值均小于行比例阀值, 则所述的疑似行比例与 预设比例的相似度小于行比例阀值。
[权利要求 7] 根据权利要求 1所述的 QR码特征检测方法, 其特征在于, 所述宽度阀 值的 2倍与预设行数阀值的和大于 QR码位置探测图形的边长。
[权利要求 8] 根据权利要求 1所述的 QR码特征检测方法, 其特征在于, 寻找所述矩 形图像的中间列上符合预设比例的列线段, 具体为:
寻找所述矩形图像的中间列线段;
将所述中间列线段分别向上和向下两个方向延伸宽度阀值组成新的中 间列线段;
提取新的中间列线段中符合预设比例的列线段。
[权利要求 9] 根据权利要求 1所述的 QR码特征检测方法, 其特征在于, 将包含所述 矩形图像与所述列线段的方形图像的一条对角线分别沿向上和向下两 个方向进行平移; 寻找并记录平移的对角线上的所述方形图像边缘黑 白分界点, 具体为:
形成包含所述矩形图像与所述列线段的方形图像; 取所述方形图像的一条对角线;
预设平移距离, 将所述对角线分别沿向上和向下两个方向进行平移; 在平移的对角线上, 端点上的值若为黑点则向方形图形外侧移动该黑 点直至遇到白点并记录白点的坐标; 端点上的值若为白点则向方形图 形内侧移动该白点直至遇到黑点并记录黑点的坐标;
记录的白点和黑点的坐标即为所述位置探测图形边缘的黑白分界点。
[权利要求 10] —种 QR码特征检测系统, 其特征在于, 包括:
划分模块, 用于将接收到的 QR码图像划分为与行数阀值具有相同像 素点的行数的多行图像;
计算模块, 用于计算所述的多行图像中每列像素点的平均灰度值形成 行数据;
第一二值化模块, 用于对所述行数据进行二值化; 统计模块, 用于统计二值化后的行数据中连续的 1和连续的 0的个数组 成行比例;
提取模块, 用于提取所述行比例中与预设比例的相似度在行比例阀值 内的行比例所对应的矩形图像;
第二二值化模块, 将所述矩形图像以及矩形图像向上和向下各延伸宽 度阀值的区域范围进行二值化;
第一寻找模块, 用于寻找所述矩形图像的中间列上符合预设比例的列 线段;
平移模块, 用于将包含所述矩形图像与所述列线段的方形图像的一条 对角线分别沿向上和向下两个方向进行平移;
第二寻找模块, 用于寻找平移的对角线上的所述方形图像边缘黑白分 界点;
记录模块, 用于记录平移的对角线上的所述方形图像边缘黑白分界点 组成模块, 用于将所述黑白分界点连接起来组成 QR码的位置探测图 形。
PCT/CN2016/091006 2015-07-30 2016-07-22 一种qr码特征检测方法及系统 WO2017016448A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP16829814.9A EP3309705B1 (en) 2015-07-30 2016-07-22 Qr code feature detection method and system
BR112018001703-5A BR112018001703A2 (pt) 2015-07-30 2016-07-22 método e sistema de detecção de recurso de código de resposta rápida
US15/871,912 US10102408B2 (en) 2015-07-30 2018-01-15 Quick response code feature detection method and system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510456884.6 2015-07-30
CN201510456884.6A CN105117677B (zh) 2015-07-30 2015-07-30 一种qr码特征检测方法及系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/871,912 Continuation US10102408B2 (en) 2015-07-30 2018-01-15 Quick response code feature detection method and system

Publications (1)

Publication Number Publication Date
WO2017016448A1 true WO2017016448A1 (zh) 2017-02-02

Family

ID=54665661

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/091006 WO2017016448A1 (zh) 2015-07-30 2016-07-22 一种qr码特征检测方法及系统

Country Status (5)

Country Link
US (1) US10102408B2 (zh)
EP (1) EP3309705B1 (zh)
CN (1) CN105117677B (zh)
BR (1) BR112018001703A2 (zh)
WO (1) WO2017016448A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110033466A (zh) * 2019-04-01 2019-07-19 数坤(北京)网络科技有限公司 一种基于多灰度级的冠脉拉直图像分割边界确定方法

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105117677B (zh) * 2015-07-30 2017-10-31 福建联迪商用设备有限公司 一种qr码特征检测方法及系统
CN106127276B (zh) * 2016-06-27 2018-12-21 福建联迪商用设备有限公司 一种量化污损QRCode的方法及系统
CN106503604B (zh) * 2016-10-18 2019-05-14 上海觉感视觉科技有限公司 一种qr码的位置探测图形分组方法
CA177629S (en) 2017-04-13 2019-02-06 Hamamatsu Photonics Kk Digital slide scanner
JP1585741S (zh) 2017-04-13 2017-09-11
USD848433S1 (en) * 2017-04-13 2019-05-14 Hamamatsu Photonics K.K. Management card for digital slides
JP1585743S (zh) 2017-04-13 2017-09-11
JP1585742S (zh) 2017-04-13 2017-09-11
JP1585739S (zh) * 2017-04-13 2017-09-11
JP1585744S (zh) 2017-04-13 2017-09-11
CN107729790B (zh) 2017-09-27 2020-12-29 创新先进技术有限公司 二维码定位方法及装置
CN110555498A (zh) * 2018-06-04 2019-12-10 北京三快在线科技有限公司 二维码的生成方法、装置、电子设备、存储介质
CN109993019A (zh) * 2019-04-15 2019-07-09 苏州国芯科技股份有限公司 基于连通域分析的二维码识别方法、系统及设备和介质
CN112329496B (zh) * 2020-12-02 2023-12-19 库卡机器人(广东)有限公司 矩形二维码识别方法、装置及电子设备

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120138694A1 (en) * 2008-03-27 2012-06-07 Denso Wave Incorporated Two-dimensional code having rectangular region provided with specific patterns for specify cell positions and distinction from background
US20130140356A1 (en) * 2011-12-05 2013-06-06 Xerox Corporation Camera positioning tool for symbology reading
CN104517089A (zh) * 2013-09-29 2015-04-15 北大方正集团有限公司 一种二维码解码系统及其方法
CN104517108A (zh) * 2013-09-29 2015-04-15 北大方正集团有限公司 一种确定qr码二值化图像边缘线的方法及系统
CN104517090A (zh) * 2013-09-29 2015-04-15 北大方正集团有限公司 一种qr码探测图形的检测方法及系统
CN104700062A (zh) * 2015-03-20 2015-06-10 中国联合网络通信集团有限公司 一种识别二维码的方法及设备
CN105117677A (zh) * 2015-07-30 2015-12-02 福建联迪商用设备有限公司 一种qr码特征检测方法及系统

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2938338B2 (ja) * 1994-03-14 1999-08-23 株式会社デンソー 二次元コード
CN104517092B (zh) * 2013-09-29 2018-04-03 北大方正集团有限公司 一种qr码校正图形的检测方法及系统
US9865027B2 (en) * 2014-05-09 2018-01-09 Graphiclead LLC System and method for embedding of a two dimensional code with an image

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120138694A1 (en) * 2008-03-27 2012-06-07 Denso Wave Incorporated Two-dimensional code having rectangular region provided with specific patterns for specify cell positions and distinction from background
US20130140356A1 (en) * 2011-12-05 2013-06-06 Xerox Corporation Camera positioning tool for symbology reading
CN104517089A (zh) * 2013-09-29 2015-04-15 北大方正集团有限公司 一种二维码解码系统及其方法
CN104517108A (zh) * 2013-09-29 2015-04-15 北大方正集团有限公司 一种确定qr码二值化图像边缘线的方法及系统
CN104517090A (zh) * 2013-09-29 2015-04-15 北大方正集团有限公司 一种qr码探测图形的检测方法及系统
CN104700062A (zh) * 2015-03-20 2015-06-10 中国联合网络通信集团有限公司 一种识别二维码的方法及设备
CN105117677A (zh) * 2015-07-30 2015-12-02 福建联迪商用设备有限公司 一种qr码特征检测方法及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3309705A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110033466A (zh) * 2019-04-01 2019-07-19 数坤(北京)网络科技有限公司 一种基于多灰度级的冠脉拉直图像分割边界确定方法

Also Published As

Publication number Publication date
BR112018001703A2 (pt) 2018-11-06
US10102408B2 (en) 2018-10-16
EP3309705A1 (en) 2018-04-18
EP3309705B1 (en) 2020-06-03
EP3309705A4 (en) 2018-07-11
US20180137323A1 (en) 2018-05-17
CN105117677A (zh) 2015-12-02
CN105117677B (zh) 2017-10-31

Similar Documents

Publication Publication Date Title
WO2017016448A1 (zh) 一种qr码特征检测方法及系统
CN109389121B (zh) 一种基于深度学习的铭牌识别方法及系统
CN106446894A (zh) 一种基于轮廓识别球型目标物位置的方法
WO2017041600A1 (zh) 一种汉信码特征图形检测方法及系统
CN101122953A (zh) 一种图片文字分割的方法
CN102663378A (zh) 连笔手写字符的识别方法
CN103310211A (zh) 一种基于图像处理的填注标记识别方法
US11074443B2 (en) Method and device for acquiring slant value of slant image, terminal and storage medium
WO2021253633A1 (zh) 一种批量二维码的识别方法及识别终端
CN114863492B (zh) 一种低质量指纹图像的修复方法及修复装置
Yadav et al. Text extraction in document images: highlight on using corner points
Liang et al. A new wavelet-Laplacian method for arbitrarily-oriented character segmentation in video text lines
CN107369162B (zh) 一种绝缘子候选目标区域的生成方法及系统
CN102831428B (zh) 图像内快速响应矩阵码区域的提取方法
Bera et al. Normalization of unconstrained handwritten words in terms of Slope and Slant Correction
CN115731257A (zh) 基于图像的叶片形态信息提取方法
Hung et al. Skyline localization for mountain images
CN109241975B (zh) 一种基于字符中心点定位的车牌字符分割方法
Chidiac et al. A robust algorithm for text extraction from images
CN112419225B (zh) 一种基于引脚分割的sop型芯片检测方法及系统
Gui et al. A fast caption detection method for low quality video images
CN104504385A (zh) 手写粘连数字串的识别方法
Tian et al. Table frame line detection in low quality document images based on hough transform
CN115063817A (zh) 一种基于形态学检测的表格识别方法、系统以及储存介质
Tang et al. Scene text detection via edge cue and multi-features

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: 16829814

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2016829814

Country of ref document: EP

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112018001703

Country of ref document: BR

REG Reference to national code

Ref country code: BR

Ref legal event code: B01E

Ref document number: 112018001703

Country of ref document: BR

Free format text: SOLICITA-SE APRESENTAR A TRADUCAO SIMPLES DA FOLHA DE ROSTO DA CERTIDAO DE DEPOSITO DA PRIORIDADE CN201510456884.6, DE 30/07/2015; OU DECLARACAO DE QUE OS DADOS DO PEDIDO INTERNACIONAL ESTAO FIELMENTE CONTIDOS NA PRIORIDADE REIVINDICADA, CONTENDO TODOS OS DADOS IDENTIFICADORES DESTA (TITULARES, NUMERO DE REGISTRO, DATA E TITULO), CONFORME ART. 25 DA RESOLUCAO 77/2013, UMA VEZ QUE O DOCUMENTO APRESENTADO NAO POSSUI OS DADOS DA MESMA.

ENP Entry into the national phase

Ref document number: 112018001703

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20180126