WO2013170663A1 - 一种纸类识别方法及相关装置 - Google Patents

一种纸类识别方法及相关装置 Download PDF

Info

Publication number
WO2013170663A1
WO2013170663A1 PCT/CN2013/073247 CN2013073247W WO2013170663A1 WO 2013170663 A1 WO2013170663 A1 WO 2013170663A1 CN 2013073247 W CN2013073247 W CN 2013073247W WO 2013170663 A1 WO2013170663 A1 WO 2013170663A1
Authority
WO
WIPO (PCT)
Prior art keywords
area
soiling
grayscale
value
pixel
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2013/073247
Other languages
English (en)
French (fr)
Inventor
梁添才
陈定喜
王卫锋
王锟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
GRG Banking Equipment Co Ltd
Original Assignee
GRG Banking Equipment Co Ltd
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 GRG Banking Equipment Co Ltd filed Critical GRG Banking Equipment Co Ltd
Priority to EP13791262.2A priority Critical patent/EP2851874A4/en
Priority to US14/352,302 priority patent/US9189842B2/en
Priority to AU2013262328A priority patent/AU2013262328B2/en
Priority to IN3754CHN2014 priority patent/IN2014CN03754A/en
Publication of WO2013170663A1 publication Critical patent/WO2013170663A1/zh
Priority to ZA2014/03713A priority patent/ZA201403713B/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • G06T7/0004Industrial image inspection
    • G06T7/0008Industrial image inspection checking presence/absence
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/90Determination of colour characteristics
    • GPHYSICS
    • G07CHECKING-DEVICES
    • G07DHANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
    • G07D7/00Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
    • G07D7/06Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency using wave or particle radiation
    • G07D7/12Visible light, infrared or ultraviolet radiation
    • GPHYSICS
    • G07CHECKING-DEVICES
    • G07DHANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
    • G07D7/00Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
    • G07D7/181Testing mechanical properties or condition, e.g. wear or tear
    • G07D7/187Detecting defacement or contamination, e.g. dirt

Definitions

  • the present invention relates to the field of image processing, and in particular, to a paper recognition method and related device.
  • banknotes and circulation is becoming more frequent. Just as the quality of the other items is destroyed after being used for a period of time, the banknotes will be recycled and destroyed if they are used for a period of time. Especially for the central branches of major banks, because of the huge flow of banknotes, banknotes need to be recycled and destroyed due to dirt and other conditions, and the amount of money that needs to be recovered and destroyed by banknotes is very large. If it is operated manually, it will take time and effort and cost. If the task is performed by the machine, the labor will be greatly liberated, and the cost of sorting the banknotes will be greatly reduced.
  • an area to be stain-detected on a sheet of paper is stored in advance as a detection target area, and the number of image pixels printed in the detection target area is stored in advance as the reference pixel number at the time of comparison.
  • Embodiments of the present invention provide a paper identification method and related device for accurately identifying paper stains based on the degree of oldness of paper.
  • the paper identification method provided by the present invention includes: acquiring a pixel gray value group of an image in an input paper type, wherein the pixel gray value group is a combination of gray value values of sampling pixel points of a specified area in the input paper type Obtaining the mean value of the gray levels of all the pixels in the pixel gray value group, as the first a gray level average; comparing the first gray average value and the old and new threshold values, determining the old and new levels of the input paper; respectively obtaining the dirty depth of the N areas in the input paper, wherein the N is greater than or equal to 1
  • An integer value of the input papers is determined according to the soiling depth of the N regions, the area and the soiling threshold, and the soiling threshold corresponds to the new and old levels.
  • the paper identification device includes: an image acquisition unit, configured to acquire a pixel gray value group of an image in an input paper, wherein the pixel gray value group is a sampling pixel point of a specified area in the input paper class a combination of grayscale values; a grayscale obtaining unit, configured to obtain an average of grayscales of all pixels in the set of grayscale values of the pixel, as a first grayscale mean; a new and old determining unit, configured to compare the first a gray level mean value and a new and old threshold value, determining a new and old level of the input paper type; a dirty depth obtaining unit, configured to respectively obtain a dirty depth of N areas in the input paper type, wherein the N is greater than or equal to 1
  • An integer determination unit is configured to determine a dirt level of the input paper according to a soil depth, an area, and a soiling threshold of the N regions, where the soiling threshold corresponds to the new and old levels.
  • the embodiment of the present invention samples the pixel gray value group of a specific area in the input paper, and determines the input paper type according to the pixel gray value group of the specific area.
  • New and old grades based on the new and old grades of the input papers, distinguish the different areas to judge the dirty grade of the input paper. Since different dirty grade templates are used for different old and new grades, the input and oldness of the input paper will not be input.
  • the level of soiling of the paper affects, making the determination of the contamination of the input paper more precise.
  • FIG. 1 is a schematic flow chart of a paper identification method according to an embodiment of the present invention.
  • FIG. 2 is another schematic flow chart of a paper identification method according to an embodiment of the present invention.
  • FIG. 3 is a schematic diagram showing the logical structure of a paper type identification device according to an embodiment of the present invention.
  • Embodiments of the present invention provide a paper identification method and related device for accurately identifying paper stains based on the degree of oldness of paper.
  • an embodiment of a paper identification method in an embodiment of the present invention includes: 101. Acquire a pixel gray value group of an image in the input paper class;
  • the paper type discriminating means acquires a pixel gradation value group of an image in the input paper type, the pixel gradation value group being a combination of gradation values of the sampling pixel points of the designated area in the input paper type.
  • the input paper type data may be a banknote
  • the designated area may be an entire banknote or any part of the banknote depending on the actual situation and the user's needs.
  • the designated area may select an area having a higher gray value.
  • the sampling pixel may be all the pixels in the designated area, or may be sampled in the specified area according to a certain ratio (for example, 1 to 10 sampling).
  • the input paper type image may be one side of the input paper type, or may be the front and back sides of the input paper type, which are determined according to actual needs, and are not limited herein.
  • the paper type discriminating means acquires the mean value of the gradations of all the pixel points in the set of pixel gradation values as the first gradation mean value.
  • the "first” in the first gray mean value has no meaning such as order or size, and is only used to distinguish different gray mean values.
  • the paper type identifying means compares the first gray average value with the old and new threshold values to determine the old and new levels of the input paper.
  • the new and old thresholds are preset grayscale thresholds, and may have multiple groups, respectively representing grayscale thresholds of different degrees of old and new.
  • the paper discriminating device respectively acquires the stain depths of the N regions in the input paper, and the N is an integer greater than or equal to 1.
  • the N regions may be regions with different gray levels (such as dark regions, light regions, etc.), or may be regions of different positions in the input paper, according to actual needs and determination accuracy. It is not limited here.
  • the dirty depth is a parameter indicating a degree of staining of the input paper, and may be determined according to a difference between the gray average value and the standard gray value.
  • the specific calculation process is described in the following embodiments. This is not a limitation.
  • the paper type identifying means determines the soiling level of the input paper based on the soiling depth, the area and the soiling threshold of the N areas, and the soiling threshold corresponds to the new and old levels.
  • the dirtyness threshold is a preset grayscale threshold, and may have multiple sets of grayscale thresholds representing different degrees of soiling.
  • the areas of the same gray level mean different dirt levels under different new and old levels, for example, a gray scale average of 180
  • the area may be judged to be dirty in new banknotes, but not in old coins.
  • a pixel gray value group of a specific area in the input paper type is sampled, and the old and new levels of the input paper type are determined according to the pixel gray value group of the specific area; and the different area pairs are distinguished based on the old and new levels of the input paper type. Enter the paper's dirt level to judge. Because different dirty grade templates are used for different old and new grades, the oldness of the input paper does not affect the dirty grade of the input paper, which makes the input paper dirty. Stain determination is more accurate.
  • FIG. 2 another embodiment of the paper identification method in the embodiment of the present invention is shown. Includes:
  • step 201 in this embodiment is the same as that of step 101 in the foregoing embodiment shown in FIG. 1, and details are not described herein again.
  • the paper type identifying means acquires the gradation value of the pixel located at the intermediate section among the sorted results as the sampled pixel gradation value group.
  • the middle section may be a segment in the middle of the ranking result of sixty percent, that is, the gray value of the pixel points respectively ranked by twenty percent at the head and the tail is removed as the sampled pixel gray value group;
  • the intermediate section may also be a section in the middle of the ranking result of 40 to 80 percent, which is not specifically limited herein.
  • the influence of the pixels whose gray scale is too large and the gray scale is too small is removed, so that the sampling of the pixel points for judging the newness degree is more reasonable, and the judgment of the newness and the oldness is more accurate.
  • the paper type identifying means compares the first gray average value with the old and new threshold values to determine the old and new levels of the input paper.
  • the new and old thresholds are preset grayscale thresholds, and may have multiple groups, respectively representing grayscale thresholds of different degrees of old and new.
  • the paper type identifying means acquires a second grayscale mean value of each of the N regions; the "first” and “second” are used only to distinguish between grayscale mean values for judging old and old.
  • the N regions may be regions with different gray levels; further, in addition to the regions of different gray levels (such as dark regions, light regions, etc.), the same gray level may be used.
  • the input paper type may include K gradation areas having different gradations, and each of the gradation areas may further include n sub-areas, where the K is multiplied by Said n is equal to said N.
  • the paper type identifying device respectively obtains the soiled depth of each of the regions according to the second gray level mean value of the respective regions and the corresponding standard gray value of the respective regions, and the standard gray value is the same region of the same old and new banknotes
  • the gray standard value can be obtained by statistically analyzing a large number of banknote images and synthesizing the information fed back by the customer.
  • the method for calculating the dirty depth of an area is: subtracting the second gray mean value of the X area from the standard gray value of the X area and taking an absolute value, and dividing the absolute value by the A standard gray value is obtained to obtain a dirty depth of the X region, and the X region is any one of the N regions. 208. Obtain a unit dirty value of each area.
  • the paper type identifying means acquires a unit dirty value of each of the areas, the unit dirty value being a product of the soiled depth of one area and the area of the one area.
  • the paper type discriminating means divides the sum of the unit dirty values of the respective areas by the total area of the N areas to obtain the stain depth of the input paper.
  • the paper type identifying means compares the soiling depth and the soiling threshold of the input paper to determine the level of soiling of the input paper.
  • the image sensor scans the input paper to be identified in the banknote passage, and obtains the gray value data of the image input into the input paper;
  • the old and new banknotes are a whole, global concept, reflected in the entire banknote area. Therefore, through the analysis of local areas, it is basically possible to basically judge the whole new and old, just like through the face, we can basically distinguish one's skin color. same.
  • the gray value is higher than that of other regions, and the higher gray value is more convenient for reflecting the old and new banknotes.
  • the acquisition area is a positive and negative watermark area.
  • Gray value group W i
  • V21 respectively, take the pixel gray value group of 60% of the middle of VI I, V 2 1 to obtain a new pixel gray value group VI 2 , V22 , and calculate the pixel gray mean values vl, v2 of VI 2 and ⁇ 22 respectively. , as the first grayscale mean value of the front and back of the renminbi.
  • the front-old new and old grade A is 50% new and below.
  • the front-old new and old grade L1 is 6%
  • P 2 ⁇ vl ⁇ P 3 the front-old new and old grade L1 is 7 Chengxin
  • the front and the new level L1 is 80% new.
  • the front and the new level L1 is 90% new.
  • 5 ⁇ vl ⁇ 255 the front and the new level are brand new.
  • the new and old grade L 2 is 5% new and below.
  • the front and back sides of the input paper image are divided into pixel bright color area, light color area and dark color area c, wherein "are divided into” 1, "2, ..., a total of sub-areas, b is divided into W, bl, ... , by a total of y sub-regions, C is divided into Ci, Cl, ..., "a total of two sub-areas, these sub-areas are basically the same area. Calculate the grayscale mean of each subregion.
  • the dirt level is obtained according to the old and new grades L and the corresponding template data.
  • the depth of the stain is defined as the absolute value of the difference between the pixel gray value ⁇ ⁇ ⁇ and the standard gray value ⁇ -W of the pixel corresponding to the input paper type under the same new and old level.
  • the percentage of the standard gray value, ie ⁇ — ' ⁇ 100%*
  • the image's dirty depth 0 ⁇ _ w takes the larger of the dirty depth of the front and back of the image. If the dirty is divided into 4 levels, then each new and old level has three dirty templates. Stain threshold Zl, Z2, Z3, 0 ⁇ Z1 ⁇ Z2 ⁇ Z3 ⁇ 1. When o ⁇ —— ' ⁇ " ⁇ 1, The input paper type dirt level is 1, when ⁇ 1 ⁇ ⁇ ⁇ - ⁇ Z 2 , the input paper type dirt level is 2, and when Z2 ⁇ ⁇ - blood ⁇ ⁇ z3 , the input paper type dirt level is 3, when
  • An embodiment of the paper type discriminating device of the present invention for performing the above-described paper type identifying method will be described.
  • An embodiment of the paper type identifying device in the embodiment of the present invention includes:
  • the image obtaining unit 301 is configured to acquire a pixel gray value group of the image in the input paper, wherein the pixel gray value group is a combination of gray value values of the sampling pixel points of the designated area in the input paper class;
  • the unit 302 is configured to obtain an average value of gray levels of all the pixel points in the pixel gray value group, as a first gray level mean value;
  • a new and old determining unit 303 configured to compare the first gray mean value and the old and new threshold values, and determine an old and new level of the input paper type
  • a dirty depth obtaining unit 304 configured to respectively obtain a dirty depth of N regions in the input paper, wherein the N is an integer greater than or equal to 1;
  • the dirtyness determining unit 305 is configured to determine a dirtyness level of the input paper according to a dirty depth, an area of the area, and a dirtyness threshold of the N areas, where the dirtyness threshold corresponds to the new and old levels.
  • the paper identification device in the embodiment of the present invention may further include:
  • a gray sorting unit 306 configured to sort gray values of all pixel points in the pixel gray value group according to the size of the gray value
  • a sampling unit 307 configured to obtain a gray value of a pixel point located in the middle segment of the sorted result as a sampled pixel gray value group
  • the grayscale obtaining unit 302 is further configured to acquire an average value of gray levels of all the pixels in the sampled pixel gray value group as the first grayscale mean value.
  • the pollution determining unit 305 in the embodiment of the present invention may include:
  • the dirty value obtaining module 3051 is configured to obtain a unit dirty value of each area, where the unit dirty value is a product of a dirty depth of one area and an area of the one area; a dirty depth obtaining module 3052, configured to obtain a dirty depth of the input paper by dividing a sum of unit dirty values of the respective regions by a total area of the N regions;
  • the dirtyness determination module 3053 is configured to compare the dirty depth and the dirtyness threshold of the input paper, and determine a dirtyness level of the input paper.
  • the specific interaction process of each unit in the paper identification device of the embodiment of the present invention is as follows:
  • the image acquisition unit 301 acquires a pixel gray value group of an image in the input paper, and the pixel gray value group is a designated area in the input paper class.
  • the input paper type data may be a banknote; the designated area may be an entire banknote according to an actual situation and a user demand, or may be any part of the banknote, preferably, for identification, a designated area You can select an area with a higher gray value.
  • the sampling pixel may be all pixels in the specified area, or may be sampled in the specified area according to a certain ratio (for example, 1 to 10 samples).
  • the input paper type image may be one side of the input paper type, or may be the front and back sides of the input paper type, which are determined according to actual needs, and are not limited herein.
  • the gray leveling unit 306 may first sort the gray values of all the pixel points in the pixel gray value group according to the size of the gray value; Obtaining, in the result of the sorting, a gray value of a pixel located in the middle segment as a sampled pixel gray value group; preferably, the middle segment may be a segment in the middle of the sorting result of sixty percent, that is, The gray value of the pixel points of 20% of the head and tail respectively is removed as the sampled pixel gray value group; optionally, the middle section may also be in the middle of the sorting result in the middle row of 40 to 80 The section is not limited here.
  • the grayscale obtaining unit 302 acquires an average value of the gradations of all the pixels in the pixel gray value group as the first grayscale mean.
  • the grayscale obtaining unit 302 is further configured to acquire an average value of gray levels of all pixel points in the sampled pixel gray value group as the first grayscale mean value.
  • the old and new decision unit 303 compares the first gray mean value with the old and new threshold values to determine the old and new levels of the input paper.
  • the new and old thresholds are preset grayscale thresholds, and may have multiple groups, respectively representing grayscale thresholds of different degrees of old and new.
  • the dirty depth obtaining unit 304 acquires the dirty depths of the N regions in the input paper, respectively, and the N is an integer greater than or equal to 1.
  • the N regions may be regions with different gray levels (such as dark regions, light regions, etc.), or may be regions of different positions in the input paper, according to actual needs and determination accuracy. It is not limited here.
  • the smear depth is a parameter indicating the degree of the smear of the input paper, and can be determined according to the difference between the gradation mean value and the standard gradation value. The specific calculation process is described in the following embodiments, which is not limited herein.
  • the product and soiling thresholds determine the level of soiling of the input paper, and the soiling threshold corresponds to the new and old levels.
  • the dirtyness threshold is a preset grayscale threshold, and may have multiple sets of grayscale thresholds representing different degrees of soiling.
  • the areas of the same gray level mean different dirt levels under different new and old levels, for example, a gray scale average of 180
  • the area may be judged to be dirty in new banknotes, but not in old coins.
  • the dirty value obtaining module 3051 of the dirtyness determining unit 305 acquires a unit dirty value of the respective regions, and the unit dirty value is a product of a dirty depth of one region and an area of the one region;
  • the dirty depth obtaining module 3052 divides the sum of the unit dirty values of the respective regions by the total area of the N regions to obtain the dirty depth of the input paper;
  • the dirtyness determining module 3053 compares the Enter the paper's soiling depth and soiling threshold to determine the level of soiling of the input paper.
  • the disclosed apparatus and method can be implemented in other ways.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. You can choose which one according to your actual needs. Some or all of the units implement the objectives of the solution of the embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
  • the integrated unit if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in a computer readable storage medium.
  • the technical solution of the present invention may contribute to the prior art or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium.
  • a number of instructions are included to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and the like, which can store program codes. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Toxicology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Inspection Of Paper Currency And Valuable Securities (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Description

一种纸类识别方法及相关装置 本申请要求于 2012 年 5 月 17 日提交中国专利局、 申请号为 201210154819.4、 发明名称为"一种纸类识别方法及相关装置"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
本发明涉及图像处理领域, 尤其涉及一种纸类识别方法及相关装置。
背景技术
随着经济和社会的发展, 钞票越来越多, 流通也越来越频繁。 与其他 物品使用一段时间后质量下降就被回收销毁一样, 钞票使用一段时间后如 果质量下降也将被回收销毁。 特别是对各大银行的中心分行, 由于钞票流 通量巨大, 钞票由于脏污等情况需回收销毁, 而对钞票分门别类挑出需回 收销毁的脏污钞票的工作量很大。 如果由人工操作, 将费时费力, 成本很 大。 而若由机器执行此项任务, 则会大大解放人力, 钞票脏污分类成本也 将大大降低。
在现有技术中, 会预先存储纸张上成为污损检测对象的区域作为检测 对象区域, 预先存储所述检测对象区域中印刷的图像像素数作为比较时的 基准像素数。
在实际应用中, 由于待检测纸类的新旧程度会对脏污程度的判断产生 影响, 若现有技术使用新纸类的图像像素作为基准像素进行脏污判定, 则 在待检测纸类较为陈旧的情况下, 脏污判定容易产生误差。 发明内容
本发明实施例提供了一种纸类识别方法及相关装置, 用于基于纸类的 新旧程度对纸类的脏污情况进行准确的识别。
本发明提供的纸类识别方法, 包括: 获取输入纸类中图像的像素灰度 值组, 所述像素灰度值组为所述输入纸类中指定区域的采样像素点的灰度 值的组合; 获取所述像素灰度值组中所有像素点的灰度的均值, 作为第一 灰度均值; 比较所述第一灰度均值及新旧阈值, 确定所述输入纸类的新旧 等级; 分别获取所述输入纸类中 N个区域的脏污深度, 所述 N为大于或等 于 1的整数; 根据所述 N个区域的脏污深度, 区域面积和脏污阈值确定所 述输入纸类的脏污等级, 所述脏污阈值与所述新旧等级相对应。
本发明提供的纸类识别装置, 包括: 图像获取单元, 用于获取输入纸 类中图像的像素灰度值组, 所述像素灰度值组为所述输入纸类中指定区域 的采样像素点的灰度值的组合; 灰度获取单元, 用于获取所述像素灰度值 组中所有像素点的灰度的均值, 作为第一灰度均值; 新旧判定单元, 用于 比较所述第一灰度均值及新旧阈值, 确定所述输入纸类的新旧等级; 脏污 深度获取单元, 用于分别获取所述输入纸类中 N个区域的脏污深度, 所述 N为大于或等于 1的整数; 脏污判定单元, 用于根据所述 N个区域的脏污 深度, 区域面积和脏污阈值确定所述输入纸类的脏污等级, 所述脏污阈值 与所述新旧等级相对应。
从以上技术方案可以看出, 本发明实施例具有以下优点: 本发明实施 例会采样输入纸类中特定区域的像素灰度值组, 并根据该特定区域的像素 灰度值组确定输入纸类的新旧等级; 再基于输入纸类的新旧等级, 区分不 同区域对输入纸类的脏污等级进行判断, 由于不同新旧等级的采用不同的 脏污等级模板, 因此输入纸类的新旧程度不会对输入纸类的脏污等级造成 影响, 从而使得输入纸类的脏污判定更加精确。
附图说明
图 1是本发明实施例纸类识别方法的一个流程示意图;
图 2是本发明实施例纸类识别方法的另一个流程示意图;
图 3是本发明实施例纸类识别装置的逻辑结构示意图。
具体实施方式
本发明实施例提供了一种纸类识别方法及相关装置, 用于基于纸类的 新旧程度对纸类的脏污情况进行准确的识别。
请参阅图 1 , 本发明实施例中纸类识别方法的一个实施例包括: 101、 获取输入纸类中图像的像素灰度值组;
纸类识别装置获取输入纸类中图像的像素灰度值组, 所述像素灰度值 组为所述输入纸类中指定区域的采样像素点的灰度值的组合。
具体的, 所述输入纸类数据可以为纸币;
所述指定区域 居实际情况和用户需求而定, 可以为整张纸币, 也可 以为纸币中任意一部分区域, 优选的, 为了便于识别, 指定区域可以选择 灰度值较高的区域。
可选的, 所述采样像素点可以为所述指定区域内所有的像素点, 也可 以为按一定比例在所述指定区域内进行采样(如, 1比 10采样)。
可选的, 输入纸类的图像可以为所述输入纸类的一面, 也可以为输入 纸类的正反两面, 具体根据实际需求而定, 此处不作限定。
102、 获取所述像素灰度值组中所有像素点的灰度的均值;
纸类识别装置获取所述像素灰度值组中所有像素点的灰度的均值, 作 为第一灰度均值。
需要注意的是,所述第一灰度均值中的"第一"并无次序或大小等含义, 仅用于区分不同的灰度均值。
103、 比较所述第一灰度均值及新旧阈值;
纸类识别装置比较所述第一灰度均值及新旧阈值, 确定所述输入纸类 的新旧等级。
具体的, 所述新旧阈值为预置的灰度阈值, 可以有多组, 分别代表不 同新旧程度的灰度阈值。
104、 分别获取所述输入纸类中 N个区域的脏污深度;
纸类识别装置分别获取所述输入纸类中 N个区域的脏污深度, 所述 N 为大于或等于 1的整数。
具体的, 所述 N个区域可以为灰度等级不同的区域(如深色区域、 浅 色区域等),也可以为所述输入纸类中不同位置的区域,具体根据实际需要 以及判定精度而定, 此处不作限定。
具体的, 所述脏污深度为表示输入纸类污损程度的参数, 可以根据灰 度均值和标准灰度值的差值进行判定,具体计算过程在后续实施例中描述, 此处不作限定。
105、 确定所述输入纸类的脏污等级。
纸类识别装置根据所述 N个区域的脏污深度, 区域面积和脏污阈值确 定所述输入纸类的脏污等级, 所述脏污阈值与所述新旧等级相对应。
具体的, 所述脏污阈值为预置的灰度阈值, 可以有多组, 分别代表不 同脏污程度的灰度阈值。
在实际应用中, 不同的新旧等级在同一区域设置有不同的脏污阈值, 即同一灰度均值的区域在不同的新旧等级下会有不同的脏污等级, 如, 一 个灰度均值为 180的区域在新纸币中可能会被判定为脏污, 而在旧币中则 不然。
本发明实施例会采样输入纸类中特定区域的像素灰度值组, 并根据该 特定区域的像素灰度值组确定输入纸类的新旧等级; 再基于输入纸类的新 旧等级, 区分不同区域对输入纸类的脏污等级进行判断, 由于不同新旧等 级的采用不同的脏污等级模板, 因此输入纸类的新旧程度不会对输入纸类 的脏污等级造成影响, 从而使得输入纸类的脏污判定更加精确。
在实际应用中, 某些小区域发生严重脏污可能会影响新旧程度的判定 准确度, 本发明提供了相应的解决方法, 请参阅图 2, 本发明实施例中纸 类识别方法的另一个实施例包括:
201、 获取输入纸类中图像的像素灰度值组;
本实施例中的步骤 201的内容与前述图 1所示的实施例中步骤 101的 内容相同, 此处不再赘述。
202、 对所述像素灰度值组中所有像素点的灰度值进行排序; 纸类识别装置根据灰度值的大小对所述像素灰度值组中所有像素点的 灰度值进行排序。
203、 获取取样像素灰度值组;
纸类识别装置获取所述排序的结果中位于中间区段的像素点的灰度值 作为取样像素灰度值组。
优选的, 中间区段可以为排序结果中中间排在百分之六十的区段, 即 去掉分别排在头尾百分之二十的像素点的灰度值作为取样像素灰度值组; 可选的,中间区段也可以为排序结果中中间排在百分之四十至八十的区段, 此处具体不作限定。
在本发明实施例中, 去掉灰度过大以及灰度过小的像素点的影响, 使 得判定新旧程度的像素点的取样更加合理, 新旧程度的判定更加精确。
204、 获取所述取样像素灰度值组中所有像素点的灰度的均值; 纸类识别装置获取所述取样像素灰度值组中所有像素点的灰度的均 值, 作为第一灰度均值。
205、 比较所述第一灰度均值及新旧阈值;
纸类识别装置比较所述第一灰度均值及新旧阈值, 确定所述输入纸类 的新旧等级。
具体的, 所述新旧阈值为预置的灰度阈值, 可以有多组, 分别代表不 同新旧程度的灰度阈值。
206、 获取所述 N个区域中各个区域的第二灰度均值;
纸类识别装置获取所述 N个区域中各个区域的第二灰度均值;所述"第 一"和"第二"仅用于区分判断新旧和脏污的灰度均值。
可选的, 所述 N个区域可以为灰度等级不同的区域; 进一步的, 除了 不同灰度等级的区域(如深色区域、 浅色区域等) 划分之外, 还可以在同 一灰度等级的区域内进行位置上的区域划分, 因此, 所述输入纸类可以包 括 K个灰度等级不同的灰度区域, 各个所述灰度区域中还可以包括 n个子 区域, 所述 K乘以所述 n等于所述N。
207、 获取所述各个区域的脏污深度;
纸类识别装置根据所述各个区域的第二灰度均值和所述各个区域相应 的标准灰度值分别获取所述各个区域的脏污深度, 所述标准灰度值为同一 新旧等级钞票同一区域的灰度标准值, 具体可以对大量钞票图像进行统计 并综合客户反馈的信息而得到。
具体的, 一个区域的脏污深度的计算方法为: 将 X区域的第二灰度均 值与所述 X区域的标准灰度值相减并取绝对值, 再将所述绝对值除以所述 标准灰度值, 得到所述 X区域的脏污深度, 所述 X区域为所述 N个区域 中的任一区域。 208、 获取所述各个区域的单位脏污值;
纸类识别装置获取所述各个区域的单位脏污值, 所述单位脏污值为一 个区域的脏污深度与所述一个区域的面积的乘积。
209、 获取所述输入纸类的脏污深度;
纸类识别装置将所述各个区域的单位脏污值的和除以所述 N个区域的 总面积, 得到所述输入纸类的脏污深度。
210、 确定所述输入纸类的脏污等级。
纸类识别装置比较所述输入纸类的脏污深度和脏污阈值, 确定所述输 入纸类的脏污等级。
为了便于理解, 下面以一具体应用场景对上述的实施例中描述的纸类 识别方法再进行详细描述, 具体为:
输入待识别的纸类, 进入走钞通道, 流程开始;
( 2 )图像传感器扫描走钞通道中的待鉴别输入纸类,获取投入输入纸 类的图像的灰度值数据;
( 3 )获取输入纸类的第一灰度均值,获取数据区域为输入纸类正反面 的灰度值较高区域;
钞票的新旧是一个整体、 全局的概念, 反映在整张钞票区域上, 所以, 通过局部区域的分析基本就可以基本判断出整体的新旧, 就好像通过脸, 我们就可以基本分辨一个人的肤色一样。 另外, 钞票使用一段时间后, 灰 度值较高区域比其他区域像素灰度值变化更大, 灰度值较高区域更能方便 反映钞票的新旧程度。
( 4 )对第一灰度均值与新旧阈值的储存数据进行对比分析,确定纸页 类新旧等级;
以人民币为例, 获取区域为正反水印区域。 以人民币为例, 分别获取 正反水印区域像素灰度值组 W、 V2 ( yi , V2分别代表正面和反面的像素 灰度值), 对 W、 中灰度值进行大小排序得到新的像素灰度值组 W i、
V21 , 分别取 VI I、 V21中正中 60%区段的像素灰度值组得到新的像素灰度 值组 VI2、 V22 , 分别计算 VI2、 ^22的像素灰度均值 vl、 v2 , 作为人民币 正反面第一灰度均值。 如果把输入纸类正反面都分成全新、 9成新、 8成新、 7成新、 6成新、 5成新及以下六个等级, 则正面有五个新旧阈值 ^、 P2、 P3、 P45, 其中 0≤ ^< 2< P3<JP4<JP5<255,类似,则反面也有五个新旧阈值" 1、 "2、 «3、 《4、 η5 , 其中 0≤ wl< ^〈/^〈/^〈/^≤255。
当 0≤νΐ< ^时正面新旧等级 A为 5成新及以下, 当 ^≤^1</^时正面 新旧等级 L1为 6 成新, 当 P2≤ vl<P3时正面新旧等级 L1为 7 成新, 当 P3≤vl< Ρ4时正面新旧等级 L1为 8成新,当 ≤ vl < 时正面新旧等级 L1为 9成新, 当 5≤vl≤255时正面新旧等级 为全新。 同理, 当 0≤ν2<"1时反 面新旧等级 L2为 5成新及以下,当" 1≤ν22时反面新旧等级 L2为 6成新, 当 "2≤ν2<"3时反面新旧等级 L2为 7成新,当 "3≤v2<"4时反面新旧等级 L2 为 8成新, 当" 4≤v2<"5时反面新旧等级 L2为 9成新, 当" 5≤v2≤255时反 面新旧等级 L2为全新, 则输入纸类的新旧等级 L取 与 的较小者。
(5)获取输入纸类的第二灰度均值。
把输入纸类图像正反面各分为像素亮色区"、 浅色区 与深色区 c , 其 中"又分为 "1、 "2、 … 、 共 个子区域, b又分为 W、 bl、 …、 by共 y个 子区域, C又分为 Ci、 Cl、 …、 《共2个子区域, 这些子区域面积基本相同。 计算每个子区域的灰度均值。
本地存储有每一级新旧等级下输入纸类每个子区域的标准脏污数据, 即每一级新旧等级下输入纸类每个子区域的灰度均值。
(6)对第二灰度均值与脏污阈值的储存数据进行对比分析,确定纸页 类脏污等级;
根据新旧等级 L与相应的模版数据获取脏污等级。 先定义脏污深度, 某个区域的脏污深度 deeP - stain定义为像素灰度值 Ρίχ与相同新旧等级下输 入纸类所对应像素的标准灰度值 ^-W 的差值的绝对值比标准灰度值 的百分比, 即 ^— '·" = 100%*|/^— 分别令正反 面每个子区域的面积与其脏污深度的乘积的和除以图像正面或反面区域面 积, 即得图像正反面的脏污深度。 图像的脏污深度0^ _ w取图像正反面 的脏污深度的较大者。 如果把脏污分为 4个等级, 则每个新旧等级下模版 有三个脏污阈值 Zl、 Z2、 Z3, 0≤Z1<Z2<Z3≤1。 当 o≤ — '·"<Ζ1时, 则输入纸类脏污等级为 1 , 当 Ζ1≤^φ— < Z 2时, 则输入纸类脏污等级 为 2 , 当 Z2 ≤ ^ —血 ^ < z3时, 则输入纸类脏污等级为 3 , 当
Z3 < deep _ stain ≤ 1时, 则输入纸类脏污等级为 4。
上面仅以一些例子对本发明实施例中的应用场景进行了说明, 可以理 解的是, 在实际应用中, 还可以有更多的应用场景, 具体此处不作限定。
下面对用于执行上述纸类识别方法的本发明纸类识别装置的实施例进 行说明, 其逻辑结构请参考图 3 , 本发明实施例中纸类识别装置的一个实 施例包括:
图像获取单元 301 , 用于获取输入纸类中图像的像素灰度值组, 所述 像素灰度值组为所述输入纸类中指定区域的采样像素点的灰度值的组合; 灰度获取单元 302, 用于获取所述像素灰度值组中所有像素点的灰度 的均值, 作为第一灰度均值;
新旧判定单元 303 , 用于比较所述第一灰度均值及新旧阈值, 确定所 述输入纸类的新旧等级;
脏污深度获取单元 304, 用于分别获取所述输入纸类中 N个区域的脏 污深度, 所述 N为大于或等于 1的整数;
脏污判定单元 305 , 用于根据所述 N个区域的脏污深度, 区域面积和 脏污阈值确定所述输入纸类的脏污等级, 所述脏污阈值与所述新旧等级相 对应。
可选的, 本发明实施例中的纸类识别装置还可以包括:
灰度排序单元 306, 用于根据灰度值的大小对所述像素灰度值组中所 有像素点的灰度值进行排序;
取样单元 307 , 用于获取所述排序的结果中位于中间区段的像素点的 灰度值作为取样像素灰度值组;
所述灰度获取单元 302还用于获取所述取样像素灰度值组中所有像素 点的灰度的均值, 作为第一灰度均值。
可选的, 本发明实施例中的脏污判定单元 305可以包括:
脏污值获取模块 3051 , 用于获取所述各个区域的单位脏污值, 所述单 位脏污值为一个区域的脏污深度与所述一个区域的面积的乘积; 脏污深度获取模块 3052, 用于所述各个区域的单位脏污值的和除以所 述 N个区域的总面积, 得到所述输入纸类的脏污深度;
脏污判定模块 3053 , 用于比较所述输入纸类的脏污深度和脏污阈值, 确定所述输入纸类的脏污等级。
本发明实施例纸类识别装置中各个单元具体的交互过程如下: 图像获取单元 301获取输入纸类中图像的像素灰度值组, 所述像素灰 度值组为所述输入纸类中指定区域的采样像素点的灰度值的组合。
具体的, 所述输入纸类数据可以为纸币; 所述指定区域根据实际情况 和用户需求而定, 可以为整张纸币, 也可以为纸币中任意一部分区域, 优 选的, 为了便于识别, 指定区域可以选择灰度值较高的区域。 可选的, 所 述采样像素点可以为所述指定区域内所有的像素点, 也可以为按一定比例 在所述指定区域内进行采样(如, 1 比 10采样)。 可选的, 输入纸类的图 像可以为所述输入纸类的一面, 也可以为输入纸类的正反两面, 具体根据 实际需求而定, 此处不作限定。
可选的, 确定第一灰度均值之前, 可以由灰度排序单元 306先根据灰 度值的大小对所述像素灰度值组中所有像素点的灰度值进行排序; 再由取 样单元 307获取所述排序的结果中位于中间区段的像素点的灰度值作为取 样像素灰度值组; 优选的, 中间区段可以为排序结果中中间排在百分之六 十的区段, 即去掉分别排在头尾百分之二十的像素点的灰度值作为取样像 素灰度值组; 可选的, 中间区段也可以为排序结果中中间排在百分之四十 至八十的区段, 此处具体不作限定。
灰度获取单元 302 获取所述像素灰度值组中所有像素点的灰度的均 值, 作为第一灰度均值。 所述灰度获取单元 302还用于获取所述取样像素 灰度值组中所有像素点的灰度的均值, 作为第一灰度均值。
在获取了所述第一灰度均值之后, 新旧判定单元 303比较所述第一灰 度均值及新旧阈值, 确定所述输入纸类的新旧等级。 具体的, 所述新旧阈 值为预置的灰度阈值, 可以有多组, 分别代表不同新旧程度的灰度阈值。
在确定了输入纸类的新旧程度之后, 脏污深度获取单元 304分别获取 所述输入纸类中 N个区域的脏污深度, 所述 N为大于或等于 1的整数。 具体的, 所述 N个区域可以为灰度等级不同的区域(如深色区域、 浅 色区域等),也可以为所述输入纸类中不同位置的区域,具体根据实际需要 以及判定精度而定, 此处不作限定。 具体的, 所述脏污深度为表示输入纸 类污损程度的参数, 可以根据灰度均值和标准灰度值的差值进行判定, 具 体计算过程在后续实施例中描述, 此处不作限定。 积和脏污阈值确定所述输入纸类的脏污等级, 所述脏污阈值与所述新旧等 级相对应。 具体的, 所述脏污阈值为预置的灰度阈值, 可以有多组, 分别 代表不同脏污程度的灰度阈值。
在实际应用中, 不同的新旧等级在同一区域设置有不同的脏污阈值, 即同一灰度均值的区域在不同的新旧等级下会有不同的脏污等级, 如, 一 个灰度均值为 180的区域在新纸币中可能会被判定为脏污, 而在旧币中则 不然。
具体的,脏污判定单元 305的脏污值获取模块 3051获取所述各个区域 的单位脏污值, 所述单位脏污值为一个区域的脏污深度与所述一个区域的 面积的乘积;再由脏污深度获取模块 3052将所述各个区域的单位脏污值的 和除以所述 N个区域的总面积, 得到所述输入纸类的脏污深度; 脏污判定 模块 3053再比较所述输入纸类的脏污深度和脏污阈值,确定所述输入纸类 的脏污等级。
在本申请所提供的几个实施例中, 应该理解到, 所揭露的装置和方法 可以通过其它的方式实现。 例如, 以上所描述的装置实施例仅仅是示意性 的, 例如, 所述单元的划分, 仅仅为一种逻辑功能划分, 实际实现时可以 有另外的划分方式, 例如多个单元或组件可以结合或者可以集成到另一个 系统, 或一些特征可以忽略, 或不执行。 另一点, 所显示或讨论的相互之 间的耦合或直接耦合或通信连接可以是通过一些接口, 装置或单元的间接 耦合或通信连接, 可以是电性, 机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的, 作为单元显示的部件可以是或者也可以不是物理单元, 即可以位于一个地 方, 或者也可以分布到多个网络单元上。 可以根据实际的需要选择其中的 部分或者全部单元来实现本实施例方案的目的。
另外, 在本发明各个实施例中的各功能单元可以集成在一个处理单元 中, 也可以是各个单元单独物理存在, 也可以两个或两个以上单元集成在 一个单元中。 上述集成的单元既可以采用硬件的形式实现, 也可以采用软 件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销 售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解, 本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方 案的全部或部分可以以软件产品的形式体现出来, 该计算机软件产品存储 在一个存储介质中, 包括若干指令用以使得一台计算机设备(可以是个人 计算机, 服务器, 或者网络设备等)执行本发明各个实施例所述方法的全 部或部分步骤。 而前述的存储介质包括: U盘、 移动硬盘、 只读存储器 ( ROM, Read-Only Memory ), 随机存取存储器(RAM, Random Access Memory )、 磁碟或者光盘等各种可以存储程序代码的介质。
以上所述, 仅为本发明的具体实施方式, 但本发明的保护范围并不局 限于此, 任何熟悉本技术领域的技术人员在本发明揭露的技术范围内, 可 轻易想到变化或替换, 都应涵盖在本发明的保护范围之内。 因此, 本发明 的保护范围应所述以权利要求的保护范围为准。

Claims

权 利 要 求
1、 一种纸类识别方法, 其特征在于, 包括:
获取输入纸类中图像的像素灰度值组, 所述像素灰度值组为所述输入 纸类中指定区域的采样像素点的灰度值的组合;
获取所述像素灰度值组中所有像素点的灰度的均值, 作为第一灰度均 值;
比较所述第一灰度均值及新旧阈值, 确定所述输入纸类的新旧等级; 分别获取所述输入纸类中 N个区域的脏污深度,所述 N为大于或等于 1的整数;
根据所述 N个区域的脏污深度, 区域面积和脏污阈值确定所述输入纸 类的脏污等级, 所述脏污阈值与所述新旧等级相对应。
2、根据权利要求 1所述的方法, 其特征在于, 所述获取输入纸类中图 像的像素灰度值组之后, 包括:
根据灰度值的大小对所述像素灰度值组中所有像素点的灰度值进行排 序;
获取所述排序的结果中位于中间区段的像素点的灰度值作为取样像素 灰度值组;
所述获取像素灰度值组中所有像素点的灰度的均值, 具体为: 获取所述取样像素灰度值组中所有像素点的灰度的均值。
3、根据权利要求 1或 2所述的方法, 其特征在于, 所述分别获取所述 输入纸类中 N个区域的脏污深度, 包括:
获取所述 N个区域中各个区域的第二灰度均值;
根据所述各个区域的第二灰度均值和所述各个区域相应的标准灰度值 分别获取所述各个区域的脏污深度。
4、根据权利要求 3所述的方法, 其特征在于, 根据所述各个区域的第 污深度, 具体为: 再将所述绝对值除以所述标准灰度值, 得到所述 X区域的脏污深度, 所述 X区域为所述 N个区域中的任一区域。
5、 根据权利要求 1或 2所述的方法, 其特征在于, 所述根据所述 N 个区域的脏污深度, 区域面积和脏污阈值确定所述输入纸类的脏污等级, 包括:
获取所述各个区域的单位脏污值, 所述单位脏污值为一个区域的脏污 深度与所述一个区域的面积的乘积;
所述各个区域的单位脏污值的和除以所述 N个区域的总面积, 得到所 述输入纸类的脏污深度;
比较所述输入纸类的脏污深度和脏污阈值, 确定所述输入纸类的脏污 等级。
6、 根据权利要求 1或 2所述的方法, 其特征在于, 所述 N个区域为 灰度等级不同的区域。
7、根据权利要求 1或 2所述的方法, 其特征在于, 所述输入纸类包括 K个灰度等级不同的灰度区域, 各个所述灰度区域中还包括 n个子区域, 所述 K乘以所述 n等于所述N。
8、 一种纸类识别装置, 其特征在于, 包括:
图像获取单元, 用于获取输入纸类中图像的像素灰度值组, 所述像素 灰度值组为所述输入纸类中指定区域的采样像素点的灰度值的组合;
灰度获取单元, 用于获取所述像素灰度值组中所有像素点的灰度的均 值, 作为第一灰度均值;
新旧判定单元, 用于比较所述第一灰度均值及新旧阈值, 确定所述输 入纸类的新旧等级;
脏污深度获取单元, 用于分别获取所述输入纸类中 N个区域的脏污深 度, 所述 N为大于或等于 1的整数;
脏污判定单元, 用于根据所述 N个区域的脏污深度, 区域面积和脏污 阈值确定所述输入纸类的脏污等级,所述脏污阈值与所述新旧等级相对应。
9、 根据权利要求 8所述的装置, 其特征在于, 所述装置还包括: 灰度排序单元, 用于根据灰度值的大小对所述像素灰度值组中所有像 素点的灰度值进行排序; 取样单元, 用于获取所述排序的结果中位于中间区段的像素点的灰度 值作为取样像素灰度值组;
所述灰度获取单元还用于获取所述取样像素灰度值组中所有像素点的 灰度的均值, 作为第一灰度均值。
10、 根据权利要求 8或 9所述的装置, 其特征在于, 所述脏污判定单 元包括:
脏污值获取模块, 用于获取所述各个区域的单位脏污值, 所述单位脏 污值为一个区域的脏污深度与所述一个区域的面积的乘积;
脏污深度获取模块, 用于将所述各个区域的单位脏污值的和除以所述 Ν个区域的总面积, 得到所述输入纸类的脏污深度;
脏污判定模块, 用于比较所述输入纸类的脏污深度和脏污阈值, 确定 所述输入纸类的脏污等级。
PCT/CN2013/073247 2012-05-17 2013-03-27 一种纸类识别方法及相关装置 Ceased WO2013170663A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
EP13791262.2A EP2851874A4 (en) 2012-05-17 2013-03-27 Method for paper identification and corresponding device
US14/352,302 US9189842B2 (en) 2012-05-17 2013-03-27 Paper identifying method and related device
AU2013262328A AU2013262328B2 (en) 2012-05-17 2013-03-27 Paper identifying method and related device
IN3754CHN2014 IN2014CN03754A (zh) 2012-05-17 2013-03-27
ZA2014/03713A ZA201403713B (en) 2012-05-17 2014-05-21 Paper identifying method and related device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201210154819.4A CN102682514B (zh) 2012-05-17 2012-05-17 一种纸类识别方法及相关装置
CN201210154819.4 2012-05-17

Publications (1)

Publication Number Publication Date
WO2013170663A1 true WO2013170663A1 (zh) 2013-11-21

Family

ID=46814366

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/073247 Ceased WO2013170663A1 (zh) 2012-05-17 2013-03-27 一种纸类识别方法及相关装置

Country Status (7)

Country Link
US (1) US9189842B2 (zh)
EP (1) EP2851874A4 (zh)
CN (1) CN102682514B (zh)
AU (1) AU2013262328B2 (zh)
IN (1) IN2014CN03754A (zh)
WO (1) WO2013170663A1 (zh)
ZA (1) ZA201403713B (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020052263A1 (zh) * 2018-09-14 2020-03-19 威海华菱光电股份有限公司 纸张厚度的检测方法、装置、存储介质和处理器

Families Citing this family (35)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102682514B (zh) 2012-05-17 2014-07-02 广州广电运通金融电子股份有限公司 一种纸类识别方法及相关装置
CN103413375B (zh) * 2013-07-05 2017-05-17 深圳市国科微半导体股份有限公司 基于图像统计特征的纸币新旧判别系统及其判别方法
CN103324946B (zh) 2013-07-11 2016-08-17 广州广电运通金融电子股份有限公司 一种纸币识别分类的方法及系统
CN103489255B (zh) * 2013-08-23 2016-04-06 广州智萃电子科技有限公司 一种钞票污损检测装置和方法
CN103456075B (zh) * 2013-09-06 2015-11-25 广州广电运通金融电子股份有限公司 一种纸币处理方法及装置
CN104504802B (zh) * 2014-10-23 2017-02-15 深圳怡化电脑股份有限公司 一种纸币拼接信号的检测方法
CN104867225B (zh) * 2015-05-15 2018-12-11 深圳怡化电脑股份有限公司 一种纸币的面向识别方法及装置
WO2016183831A1 (zh) * 2015-05-20 2016-11-24 深圳怡化电脑股份有限公司 纸币缺损识别方法及装置
CN105184950A (zh) * 2015-06-03 2015-12-23 深圳怡化电脑股份有限公司 一种分析纸币新旧的方法及装置
CN105139510B (zh) * 2015-08-25 2018-07-17 深圳怡化电脑股份有限公司 纸币识别方法及系统
CN105844780B (zh) * 2016-03-18 2019-06-28 光荣电子工业(苏州)有限公司 纸张判别装置、校正信息设定方法、以及纸张判别方法
CN105957061A (zh) * 2016-04-22 2016-09-21 武汉理工大学 一种用于旋冠盖的阈值图制作和图片检测方法
DE102016011417A1 (de) * 2016-09-22 2018-03-22 Giesecke+Devrient Currency Technology Gmbh Verfahren und Vorrichtung zur Erkennung von Farbabnutzungen an einem Wertdokument, insbesondere einer Banknote, sowie Wertdokumentbearbeitungssystem
CN106815923A (zh) * 2016-12-29 2017-06-09 深圳怡化电脑股份有限公司 一种纸币版本的识别方法及装置
CN106910276B (zh) * 2017-02-24 2019-04-26 深圳怡化电脑股份有限公司 检测纸币新旧的方法及装置
CN107067390B (zh) * 2017-03-09 2020-01-14 深圳怡化电脑股份有限公司 一种检测图像传感器的灰度分辨率的方法及装置
CN106898082B (zh) * 2017-03-15 2019-08-27 深圳怡化电脑股份有限公司 一种图像污损判定的方法及装置
CN108665604B (zh) * 2017-03-30 2020-09-15 深圳怡化电脑股份有限公司 一种纸币新旧等级的检测方法及装置
CN107012717B (zh) * 2017-04-06 2018-09-14 南京三宝弘正视觉科技有限公司 一种纸张生产控制设备和方法
CN107481391B (zh) * 2017-07-03 2019-11-19 广州广电运通金融电子股份有限公司 纸币涂鸦的检测方法及装置
CN107610317B (zh) * 2017-08-21 2019-09-20 深圳怡化电脑股份有限公司 假币鉴伪方法及装置
CN107705415A (zh) * 2017-10-25 2018-02-16 深圳怡化电脑股份有限公司 一种纸币字迹污损的检测方法及检测设备
CN108257292A (zh) * 2018-02-06 2018-07-06 深圳怡化电脑股份有限公司 纸币分类回收方法、装置和系统以及存储介质
CN108460900A (zh) * 2018-03-02 2018-08-28 深圳怡化电脑股份有限公司 一种计算纸币污损等级的方法、装置、设备及存储介质
CN108492447B (zh) * 2018-03-20 2020-09-15 深圳怡化电脑股份有限公司 美元面向识别方法、电子装置及计算机可读存储介质
CN110867015A (zh) * 2019-09-29 2020-03-06 广州国瀚计算机通讯科技有限公司 一种人民币鉴伪方法
CN110791942B (zh) * 2019-11-13 2021-06-25 东华大学 在图像分析系统中对浅色衣物脏污程度的评估方法
CN113269708B (zh) * 2020-04-17 2024-07-12 深圳市怡化时代科技有限公司 介质新旧的确定方法、装置、计算机设备和存储介质
CN113255674B (zh) * 2020-09-14 2024-09-03 深圳怡化时代智能自动化系统有限公司 字符识别方法、装置、电子设备及计算机可读存储介质
US11311192B1 (en) * 2020-11-05 2022-04-26 Hand Held Products, Inc. Methods and systems for predicting printed label's life
CN115131910B (zh) * 2022-05-30 2024-02-13 华中科技大学同济医学院附属协和医院 一种基于大数据的票据检验系统
CN115393630A (zh) * 2022-08-03 2022-11-25 中国电力科学研究院有限公司 一种基于机器图像识别的电气设备表面污秽程度检测方法
CN116721067B (zh) * 2023-05-29 2024-04-12 宿迁凯达环保设备制造有限公司 一种基于机器视觉的浸渍纸浸渍质量检测方法
CN118038366B (zh) * 2024-02-20 2024-11-22 青岛法牧机械有限公司 一种猪场养殖智能监控系统及方法
CN118348023B (zh) * 2024-03-05 2024-12-20 江苏梅曼新能源科技有限公司 一种基于数据分析的电子纸母片aoi瑕疵检测系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2009072211A1 (ja) * 2007-12-07 2009-06-11 Glory Ltd. 紙幣識別装置および紙幣識別方法
US20110246278A1 (en) * 2010-03-31 2011-10-06 Glory Ltd. Paper-sheet processing apparatus and paper-sheet processing method
CN102236897A (zh) * 2011-05-09 2011-11-09 哈尔滨工业大学 基于纸币退化能量函数的清分机纸币新旧和残缺检测方法
CN102456247A (zh) * 2010-10-20 2012-05-16 上海古鳌电子科技股份有限公司 具有双面图像扫描功能的纸币清分机
CN102682514A (zh) * 2012-05-17 2012-09-19 广州广电运通金融电子股份有限公司 一种纸类识别方法及相关装置

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE3276200D1 (en) * 1981-08-11 1987-06-04 De La Rue Syst Apparatus for scanning a sheet
CN1064154C (zh) * 1995-09-19 2001-04-04 顾临怡 新旧纸币分检机及分检方法
GB9714083D0 (en) * 1997-07-04 1997-09-10 Ncr Int Inc Document recognition apparatus
GB0106817D0 (en) * 2001-03-19 2001-05-09 Rue De Int Ltd Monitoring method
JP2004265036A (ja) * 2003-02-28 2004-09-24 Nippon Conlux Co Ltd 紙葉類識別装置および方法
US8577117B2 (en) * 2008-06-30 2013-11-05 Ncr Corporation Evaluating soiling of a media item
RU2421818C1 (ru) * 2010-04-08 2011-06-20 Общество С Ограниченной Ответственностью "Конструкторское Бюро "Дорс" (Ооо "Кб "Дорс") Способ классификации банкнот (варианты)
CN101908241B (zh) * 2010-08-03 2012-05-16 广州广电运通金融电子股份有限公司 有价文件识别方法及其识别系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2009072211A1 (ja) * 2007-12-07 2009-06-11 Glory Ltd. 紙幣識別装置および紙幣識別方法
US20110246278A1 (en) * 2010-03-31 2011-10-06 Glory Ltd. Paper-sheet processing apparatus and paper-sheet processing method
CN102456247A (zh) * 2010-10-20 2012-05-16 上海古鳌电子科技股份有限公司 具有双面图像扫描功能的纸币清分机
CN102236897A (zh) * 2011-05-09 2011-11-09 哈尔滨工业大学 基于纸币退化能量函数的清分机纸币新旧和残缺检测方法
CN102682514A (zh) * 2012-05-17 2012-09-19 广州广电运通金融电子股份有限公司 一种纸类识别方法及相关装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2851874A4 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020052263A1 (zh) * 2018-09-14 2020-03-19 威海华菱光电股份有限公司 纸张厚度的检测方法、装置、存储介质和处理器
JP2022500645A (ja) * 2018-09-14 2022-01-04 ウェイハイ ファーリング オプト−エレクトロニクス カンパニー リミテッド 用紙厚さの検知方法、装置、記憶媒体およびプロセッサ
US11741591B2 (en) 2018-09-14 2023-08-29 Weihai Hualing Opto-Electronics Co., Ltd. Method and apparatus for detecting thickness of paper and storage medium

Also Published As

Publication number Publication date
CN102682514A (zh) 2012-09-19
US9189842B2 (en) 2015-11-17
ZA201403713B (en) 2015-07-29
AU2013262328B2 (en) 2015-09-24
EP2851874A1 (en) 2015-03-25
EP2851874A4 (en) 2015-05-13
AU2013262328A1 (en) 2014-05-01
US20140270460A1 (en) 2014-09-18
CN102682514B (zh) 2014-07-02
IN2014CN03754A (zh) 2015-07-03

Similar Documents

Publication Publication Date Title
WO2013170663A1 (zh) 一种纸类识别方法及相关装置
JP6667541B2 (ja) スライドサムネイル画像を使用する関心範囲検出のためのシステム及び方法
CN106952393B (zh) 纸币识别方法和装置、电子设备和存储介质
CN104867225B (zh) 一种纸币的面向识别方法及装置
CN106056751B (zh) 冠字号码的识别方法及系统
CN105989659B (zh) 一种相似字符识别方法及纸币冠字码识别方法
US10546213B2 (en) Method for evaluating the authenticity of a painting as well as a corresponding use
CN104992496A (zh) 一种纸币面向识别方法及装置
CN104966349A (zh) 一种基于atm平台的纸币新旧检测方法
CN113920434A (zh) 一种基于目标的图像翻拍检测方法、装置及介质
WO2018082540A1 (zh) 一种票据一维信号的检测方法及装置
CN107134047B (zh) 白水印检测方法及装置
CN106898078B (zh) 一种港币版本的识别方法及装置
Berger Objective ink color comparison through image processing and machine learning
CN107705418B (zh) 一种纸币面向的识别方法、装置、设备及可读存储介质
CN105139508B (zh) 一种检测纸币的方法及装置
Wang et al. CSSRS: Pore segmentation method of sandstone cast thin section images based on weak supervised learning
CN108806058A (zh) 一种纸币检测方法及装置
CN105957237B (zh) 一种纸币的版本识别方法和装置
CN106204616B (zh) 一种伊朗纸币币值的识别方法及装置
CN107680246B (zh) 一种纸币图案中的曲线边界定位方法及设备
CN106934922A (zh) 一种纸币检测方法及装置
CN111563845A (zh) 票据表面涂刮检测方法、装置、设备及存储介质
CN107481391B (zh) 纸币涂鸦的检测方法及装置
CN116092230B (zh) 一种票证鉴伪方法及系统

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: 13791262

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2013791262

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 14352302

Country of ref document: US

ENP Entry into the national phase

Ref document number: 2013262328

Country of ref document: AU

Date of ref document: 20130327

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 2014001403

Country of ref document: CL

NENP Non-entry into the national phase

Ref country code: DE