WO2014056311A1 - 一种有价文件鉴别方法及装置 - Google Patents

一种有价文件鉴别方法及装置 Download PDF

Info

Publication number
WO2014056311A1
WO2014056311A1 PCT/CN2013/073647 CN2013073647W WO2014056311A1 WO 2014056311 A1 WO2014056311 A1 WO 2014056311A1 CN 2013073647 W CN2013073647 W CN 2013073647W WO 2014056311 A1 WO2014056311 A1 WO 2014056311A1
Authority
WO
WIPO (PCT)
Prior art keywords
sensor
signal
image
value document
magnetic
Prior art date
Application number
PCT/CN2013/073647
Other languages
English (en)
French (fr)
Inventor
梁添才
余元超
王卫锋
王锟
Original Assignee
广州广电运通金融电子股份有限公司
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 广州广电运通金融电子股份有限公司 filed Critical 广州广电运通金融电子股份有限公司
Priority to AU2013330103A priority Critical patent/AU2013330103B2/en
Priority to EP13844647.1A priority patent/EP2905754B1/en
Priority to US14/426,969 priority patent/US9406183B2/en
Publication of WO2014056311A1 publication Critical patent/WO2014056311A1/zh
Priority to ZA2015/02610A priority patent/ZA201502610B/en

Links

Classifications

    • 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
    • G07D7/16Testing the dimensions
    • G07D7/164Thickness
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/018Certifying business or products
    • G06Q30/0185Product, service or business identity fraud
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/80Recognising image objects characterised by unique random patterns
    • 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
    • G07D7/04Testing magnetic properties of the materials thereof, e.g. by detection of magnetic imprint
    • 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
    • G07D7/20Testing patterns thereon
    • G07D7/2008Testing patterns thereon using pre-processing, e.g. de-blurring, averaging, normalisation or rotation
    • 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
    • G07D7/20Testing patterns thereon
    • G07D7/202Testing patterns thereon using pattern matching
    • G07D7/206Matching template patterns
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10Services
    • G06Q50/18Legal services
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/95Pattern authentication; Markers therefor; Forgery detection

Definitions

  • the present invention relates to the field of value document authentication, and more particularly to a value document identification method and apparatus. Background technique
  • An automatic rejecting machine (ATM) of a financial institution such as a bank is equipped with a valuable document discriminating device for identifying the current value document type and authenticity by identifying characteristics of current value documents (such as banknotes).
  • ATM automatic rejecting machine
  • template information of various value documents is pre-stored, and the template information includes an image signal, a magnetic signal, and a thickness signal of the value document.
  • the discriminating device uses the image sensor to collect the image signal of the current value document, uses the magnetic sensor to collect the magnetic signal of the current value document, and uses the thickness sensor to collect the thickness signal of the current value document; then, Comparing the collected image signal, the magnetic signal and the thickness signal with the image signal, the magnetic signal and the thickness signal in the template information, and determining whether the collected plurality of signals are identical to the plurality of signals in the template information, and if they are the same, Means that the current value document is a real currency, otherwise, it indicates the current The value document is a counterfeit currency, thereby identifying the type and authenticity of the current value document by identifying the image signal, the magnetic signal and the thickness signal.
  • the image signal, magnetic signal, and thickness signal stored in the template information of the automatic rejector are collected while the plurality of sensors are always parallel to the longest side of the value document.
  • the value document is transmitted in the automatic rejection machine, often with a tilt or offset, so that the multiple sensors of the authentication device collect the value document at the tilt or The signal in the offset state, so when the discriminating device compares the tilted signal collected by the plurality of sensors with the signal in the template information, even if the current value document is a genuine currency, the discriminating device will have a corresponding position The signal is different and it is determined that the current value document is a counterfeit currency, or the authentication device cannot determine the authenticity.
  • the automatic rejection machine's identification device will be unable to judge or misreport the counterfeit currency, and the user is required to put the value document newly.
  • a value document may need to be inserted multiple times, and the identification device can correctly identify and process, so the existing identification device has a low recognition accuracy for the value document, thereby reducing the working efficiency of the authentication device, and bringing the user with Inconvenience, affecting the user experience of self-service devices.
  • the present invention is designed to provide a method for authenticating value documents and loading Set to improve the recognition accuracy of the authentication device.
  • a method for identifying valuable documents including:
  • the detection object Acquiring an image signal and a first signal of the detection object by the image sensor and the first sensor, the detection object comprising the value document and the reference unit;
  • Determining an image signal of the value document in an image signal of the detection object when an image signal in a preset area of the value document matches an image signal in the preset area of template information, And determining a positional relationship between the image sensor and the first sensor according to a positional relationship between the image sensor and the reference unit and a positional relationship between the first sensor and the reference unit, where the template information includes an image. a signal and a second signal; determining a second signal of the template information at a preset position;
  • Determining a positional relationship between the preset position of the detection object and the image sensor determining a positional relationship between the preset position of the detection object and the first sensor according to a positional relationship between the image sensor and the first sensor ;
  • the first sensor is specifically a magnetic sensor, and the first signal and the second signal are both magnetic signals.
  • the first sensor is specifically a thickness sensor, and the first signal and the second signal are both thickness signals.
  • the first sensor comprises a magnetic sensor and a thickness sensor
  • the second signal includes both a magnetic signal and a thickness signal.
  • a value document authentication device comprising:
  • An image sensor configured to collect an image signal of the detection object, where the detection object includes the value document and a reference unit;
  • a first sensor configured to collect a first signal of the detection object
  • a first determining module configured to determine an image signal of the value document in an image signal of the detection object
  • a second determining module configured to: when the image signal in the preset area of the value document matches the image signal in the preset area of the template information, according to the image sensor and the reference unit a positional relationship and a positional relationship between the first sensor and the reference unit to determine a positional relationship between the image sensor and the first sensor, the template information comprising an image signal and a second signal;
  • a third determining module configured to determine a second signal of the template information at a preset position
  • a fourth determining module configured to determine a positional relationship between the preset position of the detecting object and the image sensor
  • a fifth determining module determining a positional relationship between the preset position of the detecting object and the first sensor according to a positional relationship between the image sensor and the first sensor;
  • a determining module configured to determine whether the first signal of the first sensor to collect the detection object at the preset position is the same as the second signal of the template information at the preset position
  • a display module configured to display that the value document is a real coin or a counterfeit currency.
  • the first sensor is specifically a magnetic sensor, and the first signal and the second signal are both magnetic signals.
  • the first sensor is specifically a thickness sensor, and the first signal and the second signal are both thickness signals.
  • the first sensor comprises a magnetic sensor and a thickness sensor
  • the first signal and the second signal each comprise a magnetic signal and a thickness signal.
  • the technical solution provided by the embodiment has the following advantages and features:
  • the identification does not affect the identification.
  • the effect of the price file because before using the value document, the reference unit is needed to determine the positional relationship between the image sensor and the first sensor, thereby avoiding the error of identifying the value document due to the offset of each sensor;
  • the solution provided by the present invention improves the recognition accuracy and work efficiency of the authentication device, thereby providing convenience to the user.
  • FIG. 1 is a flow chart of a method for authenticating a value document according to the present invention
  • FIG. 2 is a block diagram of a value document discriminating device provided by the present invention.
  • FIG. 3 is a schematic diagram of an image of a value document provided by the present invention.
  • FIG. 4 is a schematic diagram of an image collected by an image sensor provided by the present invention
  • Figure 5 is a schematic diagram of an image on a xoy coordinate axis provided by the present invention
  • Figure 6 is a schematic diagram of an image on a coordinate axis provided by the present invention.
  • FIG. 7 is a schematic diagram of a coordinate axis image and a template image provided by the present invention
  • FIG. 8 is a schematic diagram of an image on a xoy coordinate axis provided by the present invention
  • Figure 9 shows xoy, XOY and ⁇ ' provided by the present invention. '" image diagram;
  • FIG. 10 is a schematic diagram of an image of a value document provided by the present invention.
  • Figure 11 is a schematic diagram of an image on a xoy coordinate axis provided by the present invention.
  • FIG. 12 shows xoy, 0 ⁇ and ⁇ ' provided by the present invention. ";" A schematic view of an image; FIG. 13 there is a further provided by the present invention, the discriminating means of the value document module schematic diagram; FIG. 14 is a side view of the delivery system provided by the invention;
  • Figure 15 is a plan view of the transport system provided by the present invention.
  • Figure 16 is a schematic view showing the distribution of sensors provided by the present invention.
  • Figure 17 is a schematic view of a reference unit provided by the present invention.
  • FIG. 18 is a schematic diagram of a module of a processor provided by the present invention.
  • the embodiment of the invention provides a method and device for identifying a value document, so as to improve identification
  • the identification accuracy and work efficiency of the device provide convenience for the user.
  • FIG. 1 is a flowchart of a value document identification method, and the method includes:
  • Step S11 Acquire an image signal of the detection object and a first signal by using an image sensor and a first sensor, where the detection object includes the value document and the reference unit;
  • the first sensor may be a magnetic sensor or a thickness sensor, or the first sensor includes a magnetic sensor and a thickness sensor, but the design idea of the present invention is not limited to the two sensors or a combination thereof, and may be other Type of sensor.
  • the magnetic sensor collects the magnetic signal from the value document
  • the thickness sensor collects the thickness signal from the value document.
  • Step S12 determining an image signal of the value document in an image signal of the detection object
  • the detection object since the detection object includes the value document and the reference unit, and the subsequent step requires the image signal of the value document, the image signal of the value document needs to be cut out from the image signal of the detection object.
  • Step S13 when the image signal of the preset area of the value document matches the image signal of the preset area of the template information, according to the positional relationship between the image sensor and the reference unit Determining a positional relationship between the first sensor and the reference unit to determine a positional relationship between the image sensor and the first sensor, the template information comprising an image signal and a second signal;
  • the preset area may be large or small, and the staff may pre-set the preset area.
  • the preset area may be a RMB avatar area, if the image signal of the value document matches the image signal of the template information.
  • you can pass the current denomination of the template information for example, the denomination of the current template information is 50 yuan, then the denomination of the value document is also 50 yuan. In this step, only the denomination value input by the user can be judged, and the authenticity of the value document cannot be determined.
  • Step S14 Determine a second signal that the template information is at a preset position
  • the second signal at the preset position may be a magnetic signal, a thickness signal, or a magnetic signal and a thickness signal.
  • Step S15 determining a relationship between a preset position of the detection object and a position of the image sensor
  • Step S16 determining a positional relationship between the preset position of the detection object and the first sensor according to a positional relationship between the image sensor and the first sensor;
  • the preset position of the detection object and the positional relationship of the first sensor are further determined by the above two steps.
  • Step S17 determining whether the first signal of the first sensor to collect the detection object at the preset position is the same as the second signal of the template information at the preset position; if yes, proceeding to step S18; Step S19.
  • the determination is an error correction determination, that is, determining whether the first signal of the detection object is the same as the second signal of the template information, and the second signal of the template information is a signal acquired in an ideal state, and the first signal of the detection object is even
  • the acquisition in the shift state does not affect the correctness of the judgment, because the positional relationship is changed inside.
  • Step S18 Display that the value document is a real coin
  • Step S19 Display the value document as a counterfeit currency.
  • the solution provided by the present invention improves the recognition accuracy and work efficiency of the authentication device, thereby providing convenience to the user.
  • FIG. 2 is a block diagram of a value document discriminating device, the device comprising: an image sensor 1 for collecting an image signal of a detection object, the detection object including the price a first sensor 2, configured to acquire a first signal of the detection object; a first determining module 3, configured to determine an image signal of the value document in the image signal of the detection object; The module 4 is configured to: when the image signal in the preset area of the value document matches the image signal in the preset area of the template information, according to the position of the image sensor 1 and the reference unit Relationship between the relationship between the first sensor 2 and the reference unit to determine a positional relationship between the image sensor 1 and the first sensor 2, the template information including an image signal and a second signal; a second signal for determining that the template information is at a preset position; a fourth determining module 6 configured to determine a preset position of the detection object and a position of the image sensor 1
  • the fifth determining module 7 determines a positional relationship between the preset position of the detecting object and the
  • the above tube is intended to summarize the design of the present invention, and the solution of the present invention will be described below with reference to specific examples. Since the design ideas of magnetic signal detection and thickness signal detection are substantially the same in the specific implementation, the following examples only introduce the part of the magnetic signal detection. According to the principle of magnetic signal detection, the same implementation can be implemented by using the same implementation method. The thickness signal collected by the sensor is accurately positioned and detected to complete the final identification of the current value document, and the content repetition is not repeated here.
  • This embodiment is a specific introduction of the value document authentication method provided by the present invention.
  • a method for identifying valuable documents including:
  • Step 1 Detect the current value document
  • FIG. 3 is a physical shape diagram of a current value document, through which a value document to be identified (eg, a banknote) is transmitted, and a detection sensor is disposed in the transmission channel for detecting the arrival of the current value document, according to the device setting.
  • the data gets the transfer speed V (inches per second) of the current value document.
  • the cooperative control unit controls the closed loop conveyor operation and controls the linear velocity of the closed loop transmission to coincide with the value document transmission speed V.
  • the closed-loop conveyor transfer reference unit passes through the image sensor, magnetic sensor and thickness sensor in sequence from the starting position.
  • the image sensor is a reflection sensor that detects a transmitted image and a reflective sensor that detects a reflected image
  • the image sensor starts to operate when the detection end sensor reaches the detection sensor at the beginning of the current value document, and the image data acquisition unit acquires the projection image and the reflection image of the reference unit and the current value document in a progressive scan manner.
  • the magnetic signal acquisition unit is composed of a magnetic sensor for detecting the magnetic signal data of the reference unit and the current value document.
  • the magnetic sensor starts to work, and the magnetic signal acquisition unit acquires the reference by progressive scanning. The magnetic signal data of the unit and the current value document.
  • the thickness sensor starts to work, and the thickness signal obtaining unit acquires the thickness signal data of the reference unit and the current value document in a progressive scan manner.
  • Step 6 Stop signal acquisition
  • the elapsed time ⁇ (seconds) ends the data acquisition of the image sensor, magnetic sensor and thickness sensor.
  • Figure 4 is an image of the reference unit and the current value document collected by the image sensor.
  • the number of image data lines is:
  • the number of image data columns is:
  • the image sensor has a horizontal width of ⁇ inches, a horizontal resolution of 7 ⁇ dpi, and a vertical resolution of dpi. Since the magnetic sensor is a line scan, the magnetic signal data acquired at one time is w discrete data, where W is the number of magnetic sensor heads.
  • the complete magnetic signal data of the current value document is a numerical matrix ⁇ ⁇ of M rows and N columns, as shown in (Equation 6.3). d 0 d 0 d 0 d 0 0 0 0
  • the magnetic signal data acquired at one time is a discrete data, where is the number of thickness sensors.
  • the complete magnetic signal data of the current value document is the value matrix %' of the ⁇ : row, as shown in (Formula 6.4).
  • Step 7.1 Image Cutting The image data collected by the image sensor is used to initially identify the current value document. Because there is uncertainty in identifying the useful foreground region in the entire image, the foreground region should be cut first by boundary detection and image segmentation.
  • the linear direction of the left boundary of the entire image captured by the image sensor is the y-axis
  • the linear direction of the lower boundary of the entire image captured by the image sensor is the axis
  • the lower left vertex of the entire image acquired by the image sensor is the coordinate origin. (ie pixel coordinate system, each point of the coordinate system represents one pixel), as shown in Figure 5.
  • the four vertices of the foreground area are A, B, (and 0.
  • the upper boundary line equation (that is, the straight line equation where the edge AD is located;):
  • the image is cut and the cut foreground area is cut.
  • the domain uses existing image rotation techniques for rotational correction.
  • the line in which the left boundary of the foreground area of the current value document is located is the ordinate direction
  • the lower left vertex of the foreground area of the current value document is used as the coordinate origin to establish the coordinate system y.
  • Figure 6 shows.
  • Step 7.2 Image Feature Template Matching Recognition
  • the corresponding position of the foreground area is selected for template matching identification, as shown in FIG. 7.
  • the result of image recognition is obtained by template matching, and if the image recognition fails, the current value document is exited.
  • the image recognition result is RMB100, and the front is erect.
  • the identification scheme has completed the image recognition of the current value document, and the recognition result is CNY100.
  • M can be magnetized on the banknote (the magnetic signal fluctuates sharply).
  • it is necessary to first determine the value document M. The position of the magnetic signal at the magnetic signal collected by the magnetic sensor.
  • Step 8.1 Determine the physical position relationship between the image sensor and the reference unit
  • Drx and Dry are length units, Drx is the distance from W to Y, and Dry is W. The distance to the X axis.
  • the physical size coordinate system is established (the left end of the image sensor 13 is taken as the coordinate origin, the horizontal direction of the image sensor 13 is the horizontal coordinate direction, and the paper feed direction is the ordinate direction), as shown in FIG.
  • the line number of the line_y contains the number of pixel lines:
  • VTL -D is based on the vertical resolution of the image sensor as dpi, so the physical size distance between the line where the horizontal direction of the image sensor is located and the center line Line_y of the reference unit is:
  • the coordinates of the reference unit center point W in the physical size coordinate system are
  • the physical dimension coordinate system is established (the left vertex of the reference unit 12 is taken as the coordinate origin, the horizontal direction of the reference unit 12 is the abscissa direction, and the paper feed direction is the ordinate direction), as shown in FIG.
  • the coordinates of the center point W of the reference unit 12 in the coordinate system are D,
  • the physical dimension distance of the reference unit center point w to the coordinate axis of the coordinate system is
  • W data Mag(D wy , D ⁇
  • the positional relationship between the physical size coordinate system " ⁇ "y'' and the physical size coordinate system (i.e., the positional relationship between the magnetic sensor 14 and the reference unit 12) may be assumed to be > ⁇ ' ⁇ , that is, the left end of the reference unit 12 It is more to the left than the left end of the magnetic sensor 14, as shown in FIG.
  • the physical size coordinate system " ⁇ "y" satisfies the following relationship:
  • Equation 8.3 As can be seen from Equation 8.3, the positional relationship between the magnetic sensor 14 and the image sensor 13 is independent of the setting of the reference unit 12, that is, the reference unit 12 merely serves as a transition bridge for locating the position of the magnetic sensor 14 by the image sensor 13 in the identification system. The system does not create any additional errors.
  • Image signal is fused with magnetic signal, and image signal is used to locate magnetic signal
  • the value file is in physical size
  • the position in the coordinate system is shown in Figure 12. According to the horizontal resolution of the image sensor is 7 ⁇ dpi, the vertical resolution is dpi, and the coordinates of the vertices A, B, and D in the physical size coordinate system are:
  • Equation 8.4 Use the upper left vertex A (JCA , ) and the lower left vertex B (JCB , ⁇ ) of the current value document to find the left edge of the current value document in the physical size coordinate system.
  • the equation of the line in y is:
  • the magnetic signal data of M is located in the entire magnetic signal data matrix.
  • the ⁇ column data is the magnetic signal data at M on the current value document.
  • the recognition result is RMB100, the positive erecting hypothesis, detection
  • This embodiment is a specific introduction of the value document discriminating device provided by the present invention.
  • FIG. 13 is a value document discriminating device, including: a control unit 16: controlling the workflow of the entire device;
  • Transfer System 20 See Figures 14 and 15 for specific details on the transfer system.
  • At least two sensors for acquiring the machine-readable signal of the value document are mounted on the upper wall 23 of the transmission system channel, and the two ends of the two closed-loop conveyor belts 22 are fixed on the side wall 25 of the channel by the drive shaft, when the value document is in the transmission channel When transmitting, the drive shaft 21 drives the conveyor belt 22 to operate.
  • the present invention preferably has the following sensor types:
  • the image sensor 13 obtains image data of the input value document in a transmissive and reflective manner. It is assumed that the horizontal width of the image sensor 13 in the present embodiment is ⁇ inches, the horizontal resolution is dpi, and the vertical resolution is ⁇ dpi, as shown in FIG. 16 . Shown
  • the magnetic sensor 14 is a magnetic signal data acquisition unit that acquires magnetic signal data of a value document, and assumes that the magnetic sensor 14 has w heads in the horizontal direction, and is distributed in a horizontal direction with a width of ⁇ inches, and the magnetic sensor has a vertical resolution. Is ⁇ (1 as shown in Figure 16;
  • the thickness sensor 15 is a thickness data acquisition unit that acquires thickness data of the input value document, and assumes that the thickness sensor 15 has r thickness sensing heads in the horizontal direction, and is distributed in a width of ⁇ inch.
  • the vertical resolution of the thickness sensor is Lt dpi, as shown in FIG. 16;
  • reference unit 12 two rectangular reference units for determining the spatial position constraint relationship between different sensors, and the physical size length is D , Let the center line of the reference unit be Line_x, the center line of the vertical direction be Line_y, and the intersection point of the two lines be W, then W is the center point of the reference unit, as shown in Figure 17.
  • the two ends of each reference unit 12 are respectively overlapped on two closed-loop conveyor belts, and when the conveyor belt moves, the reference unit moves together with the conveyor belt, as shown in FIGS. 14 and 15.
  • the reference unit 12 has features:
  • the reference unit is centered on the boundary line, the left half is black (opaque), and the right half is transparent;
  • the reference unit is magnetized in a black portion
  • the thickness of the reference unit in the black portion (left half) is significantly greater than the thickness of the transparent portion (right half);
  • the reference unit When the value document is transmitted on the transport path, the reference unit is attached to the conveyor belt and passes through the image sensor, the magnetic sensor and the thickness sensor at the same speed; that is, each time the image sensor, the magnetic sensor and the thickness sensor collect signals And including corresponding signals of the reference unit and corresponding signals of the current value document;
  • Collaborative Control Unit 21 As shown in Figure 13, the coordinated control of the closed-loop conveyor line speed and the value file transfer speed are the same, so that the reference unit's transfer speed is the same as the value file transfer speed.
  • the cooperative control unit 21 also controls the reset of the reference unit, that is, each time a signal acquisition of a value document is ended, the cooperative control unit controls the closed loop conveyor to transmit the reference unit to the designated starting position, as shown in FIGS. 14 and 15. Shown.
  • the starting position is between the detection sensor and the image sensor.
  • the storage unit 18 stores a standard template required to identify the current value document image signal; the processor 17: identifies the image signal, the magnetic signal, and the thickness signal to complete the identification of the current value document;
  • Display part 19 If the current value document (such as banknote) is recognized as a legal document (such as a genuine banknote), the currency, denomination, version, etc. will be displayed; if the current value document is identified as untrue, "Not” will be displayed. Accept!.
  • a legal document such as a genuine banknote
  • the value document discriminating device of the present invention further includes
  • the detecting sensor 11 is arranged at the beginning of the transfer wall to detect the arrival of the value document and trigger other sensors to enter the working mode, as shown in Figs. 14 and 15;
  • the processor 17 in the present invention further includes the following modules (soft modules under the processor), as shown in FIG. 18:
  • the image cutting unit 31 cuts out the foreground area useful for recognition from the background area;
  • the image correcting unit 32 determines whether there is tilt and offset of the foreground area, and performs image rectification using the existing rectifying technique;
  • the image recognition unit 33 performs matching authentication based on the image data of the current value document and the standard template stored in the storage unit;
  • Image sensor and magnetic sensor spatial position relationship modifying unit 34 correcting the image by using the distribution characteristics of the reference unit in the entire image acquired by the image sensor and the distribution characteristics of the magnetic signal of the reference unit in the entire magnetic signal collected by the magnetic sensor The spatial positional relationship between the sensor and the magnetic sensor, the real position constraint of the current position of the two sensors is obtained;
  • Image information and magnetic information are different types of information descriptions collected by different types of data acquisition units for the same to-be-identified medium. Although it is not two The same type of information, but there is some correlation between the two, because the printing of the value document is very precise, that is, the image information (especially the image feature information) and the magnetic signal (especially the magnetic characteristic signal) are in the same kind.
  • the physical location distribution on the price file is very stable and reliable, and the transmission speed of the current value document and the distribution position and resolution of the image sensor and the magnetic sensor are fixed. With this fixed relationship, the effective magnetic signal distribution of the current value document can be calculated based on the image information and the correspondence between the image sensor and the magnetic sensor at the physical position.
  • the magnetic signal recognition unit 36 based on the multi-source information fusion described above, discriminates the magnetic signal of the current value document to obtain a magnetic signal recognition result for the current value document;
  • Image sensor and thickness sensor spatial position relationship modifying unit 37 Correcting the distribution characteristics of the distribution signal in the entire image acquired by the image sensor and the thickness signal of the reference unit in the entire thickness signal collected by the thickness sensor by using the identification system reference unit The spatial positional relationship between the image sensor and the thickness sensor obtains the true position constraint of the two sensors in the current state; the image information and thickness information fusion unit 38: the image information and the thickness information are also collected by different types of data acquisition units for the same Identify different types of information descriptions for the media. Both of them also have a similar relationship between image signals and magnetic signals. Using this relationship, the effective thickness signal distribution of the current value document can be calculated based on the image information and the correspondence between the image sensor and the thickness sensor at the physical position.
  • the thickness signal identifying unit 39 based on the multi-source information fusion described above, identifying the thickness signal of the current value document, and obtaining a thickness signal recognition result for the current value document;
  • Result output unit 40 Outputs the final recognition result of the current value document.
  • an image sensor, a magnetic sensor and a thick The degree sensor obtains the observation data of the reference unit and the current value document in time series. Because in the value document discriminating device proposed by the present invention, the transmission speed of the current value document is constant, and the spatial distribution positions of the image sensor, the magnetic sensor and the thickness sensor are constant, so the present invention first utilizes the image signal of the reference unit. The magnetic signal and the thickness signal determine the true spatial positional relationship of the image sensor, the magnetic sensor and the thickness sensor, and then use the constraint relationship between the respective sensor positions to fuse the signals acquired by the image sensor, the magnetic sensor and the thickness sensor. Identify, to obtain a description and identification of the consistency of the current value documents.
  • the lower cylinder introduces the workflow between the components, step S21, inputs the value documents to be identified (such as banknotes), and the process begins;
  • Step S22 image sensor processing, scanning the reference unit and the current value document, and acquiring image information of the reference unit and the current value document;
  • Step S23 magnetic sensor processing, acquiring magnetic signal data of the reference structural unit and the current value document;
  • Step S24 processing the thickness sensor to obtain the thickness signal data of the reference structural unit and the current value document;
  • Step S25 image cutting, first determining the positional relationship of the foreground area in the entire background area, and cutting out the useful image foreground area from the background area;
  • Step S26 Correcting the image, correcting the image with the tilt and the offset, and facilitating subsequent image matching and recognition;
  • Step S27 Image recognition, placing a foreground area of the value document obtained by cutting the image into a coordinate system, and selecting image data of the specific position and matching the standard template data stored in the storage unit to obtain an image recognition result. If the image recognition fails, the current value document is exited; Step S28, correcting the physical position relationship between the image sensor and the magnetic sensor, and using the image signal and the magnetic signal of the reference unit to obtain displacements of the left vertex of the image sensor and the left vertex of the magnetic sensor in the horizontal direction and the vertical direction; thereby obtaining the image sensor and the magnetic sensor a true constraint relationship;
  • Step S29 the image information is merged with the magnetic information, according to the current value of the file transfer speed, the image sensor and the magnetic sensor physical position constraint relationship, and according to the result of the image recognition, the position of the current valuable file effective area in the background image is utilized. (Distribution), calculate the position (distribution) of the magnetic signal valid for the current value document in the entire magnetic signal collected by the magnetic sensor.
  • Step S30 magnetic signal recognition, according to the situation that the image information and the magnetic information are fused, it is determined whether the magnetic signal law of the specific position of the magnetic signal data collected by the magnetic sensor satisfies the distribution law of the magnetic signal required by the current document image recognition result. If the magnetic signal recognition fails, the current value file is exited;
  • Step S31 correcting the physical position relationship between the image sensor and the thickness sensor, and determining the displacement of the left vertex of the image sensor and the left vertex of the thickness sensor in the horizontal direction and the vertical direction by using the image signal and the thickness signal of the reference unit; thereby obtaining the image sensor and the thickness sensor a true constraint relationship;
  • Step S32 the image information is merged with the thickness information, according to the rule that the current value file transmission speed is constant, the physical position of the image sensor and the thickness sensor are unchanged, and according to the result of the image recognition, the current valuable file effective area is used in the background image.
  • the position (distribution) calculates the position (distribution) of the thickness signal effective for the current value document in the entire thickness signal acquired by the thickness sensor.
  • Step S33 the thickness signal recognition is determined according to the fusion of the image information and the thickness information. Whether the thickness signal law of the specific position of the thickness signal data collected by the fixed thickness sensor satisfies the thickness signal distribution law required by the current value document image recognition result. If the thickness signal recognition fails, the current value file is exited;
  • Step S34 If the thickness recognition is successful, the identification of the current value document is completed, and the recognition result (such as currency, denomination, version, etc.) is output.
  • the recognition result such as currency, denomination, version, etc.
  • FIG. 1 to FIG. 18 are only preferred embodiments of the present invention, and those skilled in the art can design more embodiments based on this, and therefore are not described herein.

Landscapes

  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Finance (AREA)
  • Development Economics (AREA)
  • Accounting & Taxation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Technology Law (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Primary Health Care (AREA)
  • Multimedia (AREA)
  • Inspection Of Paper Currency And Valuable Securities (AREA)
  • Image Analysis (AREA)

Abstract

一种有价文件鉴别方法及装置,无论有价文件、图像传感器或第一传感器是否发生偏移,都不会影响鉴别有价文件的效果,因为在鉴别有价文件之前,需要利用参考单元来确定图像传感器和第一传感器的位置关系,从而避免了因各个传感器发生偏移而导致鉴别有价文件发生错误;而且,在鉴别有价文件时,即使有价文件发生偏移,通过检测对象和传感器之间位置关系的转化,也可保证鉴别有价文件的正确性。因此,本发明提供的方案提高了鉴别装置的识别精度和工作效率,从而给用户提供了便利。

Description

一种有价文件鉴别方法及装置
本申请要求 2012 年 10 月 08 日提交中国专利局、 申请号为 201210378191.6、 发明名称为"一种有价文件鉴别方法及装置"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
本发明涉及有价文件鉴别领域, 更具体的说, 涉及有价文件鉴别方法 及装置。 背景技术
随着经济和科技的日益发展, 自助设备在金融领域(如银行 ATM )和 泛金融领域(如轨道交通行业的自助售票机、 自助缴费机等 )越来越普及。 可以预见, 在未来社会金融自助设备将与人们的日常生活更加密切相关, 同时也为金融自助设备的性能提出了更高的要求。
在银行等金融机构的自动拒员机(ATM ) 中装有通过鉴别当前有价文 件(如纸币)的特征来鉴别当前有价文件种类和真假的有价文件鉴别装置。 在自动拒员机中, 预先存储有各种有价文件的模板信息, 模板信息包括有 价文件的图像信号、 磁信号和厚度信号。 当投入有价文件时, 鉴别装置利 用其图像传感器采集当前有价文件的图像信号, 利用其磁传感器采集当前 有价文件的磁信号, 利用厚度传感器采集当前有价文件的厚度信号; 然后, 再将采集到的图像信号、 磁信号和厚度信号与模板信息中的图像信号、 磁 信号和厚度信号进行对比, 判断采集到的多个信号与模板信息中的多个信 号是否相同, 如果相同, 则意味着当前有价文件是真币, 否则, 说明当前 有价文件是假币, 从而通过对图像信号、 磁信号和厚度信号的识别, 来完 成对当前有价文件的种类和真假的鉴别。
在对现有技术的研究和实践过程中, 本发明的发明人发现现有技术存 在以下问题:
存储在自动拒员机的模板信息中的图像信号、 磁信号和厚度信号, 是 在多个传感器与有价文件的最长边始终平行的情况下所采集到的。 但是, 在实际操作的情况下, 有价文件在自动拒员机中, 常常是以倾斜或偏移的 情况下进行传输的, 从而鉴别装置的多个传感器采集到的就是有价文件在 倾斜或偏移状态下的信号, 所以在鉴别装置将多个传感器采集到的倾斜的 信号与模板信息中的信号进行对比时, 即使当前的有价文件是真币, 鉴别 装置也会因为两者对应位置的信号不同而判定当前的有价文件是假币, 或 者鉴别装置出现无法判断真伪的情况。
在用户使用自动拒员机的过程中, 一旦有价文件没有摆正时, 那么自 动拒员机的鉴别装置就会出现无法判断或误报假币的情况, 要求用户从新 放入有价文件, 同一个有价文件有可能需要多次插入, 鉴别装置才能正确 识别和处理, 所以现有的鉴别装置对于有价文件的有效识别精度 4艮低, 从 而降低了鉴别装置的工作效率, 给使用者带来不便, 影响自助设备使用者 的使用体验。
因此, 如何提高鉴别装置的识别精度, 成为目前最需要解决的问题。
发明内容
有鉴于此, 本发明的设计目的在于, 提供一种有价文件鉴别方法及装 置, 以提高鉴别装置的识别精度。
本发明实施例是这样实现的:
一种有价文件鉴别方法, 包括:
通过图像传感器和第一传感器采集检测对象的图像信号和第一信号, 所述检测对象包括所述有价文件和参考单元;
在所述检测对象的图像信号中确定所述有价文件的图像信号; 当在所述有价文件的预设区域的图像信号与在模板信息的所述预设区 域的图像信号相匹配时, 则根据所述图像传感器与所述参考单元的位置关 系和所述第一传感器与所述参考单元的位置关系来确定所述图像传感器和 所述第一传感器的位置关系, 所述模板信息包括图像信号和第二信号; 确定所述模板信息在预设位置上的第二信号;
确定所述检测对象的预设位置与所述图像传感器的位置关系; 根据所述图像传感器与所述第一传感器的位置关系确定所述检测对象 的预设位置与所述第一传感器的位置关系;
判断所述第一传感器采集所述检测对象在预设位置上的第一信号与所 述模板信息在预设位置上的第二信号是否相同, 若是, 则显示所述有价文 件是真币; 否则, 显示所述有价文件为假币。
优选地, 所述第一传感器具体为磁传感器, 所述第一信号和第二信号 均为磁信号。
优选地, 所述第一传感器具体为厚度传感器, 所述第一信号和第二信 号均为厚度信号。
优选地, 所述第一传感器包括磁传感器和厚度传感器, 所述第一信号 和第二信号均包括磁信号和厚度信号。
一种有价文件鉴别装置, 包括:
图像传感器, 用于采集检测对象的图像信号, 所述检测对象包括所述 有价文件和参考单元;
第一传感器, 用于采集检测对象的第一信号;
第一确定模块, 用于在所述检测对象的图像信号中确定所述有价文件 的图像信号;
第二确定模块, 用于当在所述有价文件的预设区域的图像信号与在模 板信息的所述预设区域的图像信号相匹配时, 则根据所述图像传感器与所 述参考单元的位置关系和所述第一传感器与所述参考单元的位置关系来确 定所述图像传感器和所述第一传感器的位置关系, 所述模板信息包括图像 信号和第二信号;
第三确定模块, 用于确定所述模板信息在预设位置上的第二信号; 第四确定模块, 用于确定所述检测对象的预设位置与所述图像传感器 的位置关系;
第五确定模块, 根据所述图像传感器与所述第一传感器的位置关系确 定所述检测对象的预设位置与所述第一传感器的位置关系;
判断模块, 用于判断所述第一传感器采集所述检测对象在预设位置上 的第一信号与所述模板信息在预设位置上的第二信号是否相同;
显示模块, 用于显示所述有价文件是真币或假币。
优选地, 所述第一传感器具体为磁传感器, 所述第一信号和第二信号 均为磁信号。 优选地, 所述第一传感器具体为厚度传感器, 所述第一信号和第二信 号均为厚度信号。
优选地, 所述第一传感器包括磁传感器和厚度传感器, 所述第一信号 和第二信号均包括磁信号和厚度信号。
与现有技术相比, 本实施例提供的技术方案具有以下优点和特点: 在本发明提供的方案中, 无论有价文件、 图像传感器或第一传感器是 否发生偏移,都不会影响鉴别有价文件的效果, 因为在鉴别有价文件之前, 需要利用参考单元来确定图像传感器和第一传感器的位置关系, 从而避免 了因各个传感器发生偏移而导致鉴别有价文件发生错误; 而且, 在鉴别有 价文件时, 即使有价文件发生偏移, 通过检测对象和传感器之间位置关系 的转化, 也可保证鉴别有价文件的正确性。 因此, 本发明提供的方案提高 了鉴别装置的识别精度和工作效率, 从而给用户提供了便利。 附图说明
为了更清楚地说明本发明或现有技术中的技术方案, 下面将对实施例 或现有技术描述中所需要使用的附图作筒单地介绍, 显而易见地, 下面描 述中的附图仅仅是本发明的一些实施例, 对于本领域普通技术人员来讲, 在不付出创造性劳动的前提下, 还可以根据这些附图获得其他的附图。
图 1为本发明所提供的一种有价文件鉴别方法的流程图;
图 2为本发明所提供的一种有价文件鉴别装置的模块图;
图 3为本发明所提供的有价文件的图像示意图;
图 4为本发明所提供的图像传感器采集到的图像示意图; 图 5为本发明所提供的 xoy坐标轴上的图像示意图;
图 6为本发明所提供的 坐标轴上的图像示意图;
图 7为本发明所提供的 坐标轴图像与模板图像对照的示意图; 图 8为本发明所提供的 xoy坐标轴上的图像示意图;
图 9为本发明所提供的 xoy、 XOY和^'。' "的图像示意图;
图 10为本发明所提供的有价文件的图像示意图;
图 11为本发明所提供的 xoy坐标轴上的图像示意图;
图 12为本发明所提供的 xoy、 0丫和^'。" ;"的图像示意图; 图 13为本发明所提供的另一种有价文件鉴别装置的模块示意图; 图 14为本发明所提供的传送系统的侧视图;
图 15为本发明所提供的传送系统的俯视图;
图 16为本发明所提供的传感器分布的示意图;
图 17为本发明所提供的参考单元的示意图;
图 18为本发明所提供的处理器的模块示意图。
具体实施方式
下面将结合本发明实施例中的附图, 对本发明实施例中的技术方案进 行清楚、 完整地描述, 显然, 所描述的实施例仅仅是本发明一部分实施例, 而不是全部的实施例。 基于本发明中的实施例, 本领域普通技术人员在没 有做出创造性劳动前提下所获得的所有其他实施例, 都属于本发明保护的 范围。
本发明实施例提供了一种有价文件鉴别方法及装置, 以实现提高鉴别 装置的识别精度和工作效率, 从而给用户提供了便利。
由于上述有价文件鉴别方法及装置的具体实现存在多种方式, 下面通 过具体实施例进行详细说明:
请参见图 1所示, 图 1所示的为一种有价文件鉴别方法的流程图, 该 方法包括:
步骤 Sll、 通过图像传感器和第一传感器采集检测对象的图像信号和 第一信号, 所述检测对象包括所述有价文件和参考单元;
其中, 第一传感器既可以为磁传感器, 也可以为厚度传感器, 或者第 一传感器包括磁传感器和厚度传感器, 但是本发明的设计思想并不局限于 这两种传感器或其结合, 还可以为其他类型的传感器。 对于磁传感器从有 价文件上采集到的为磁信号, 厚度传感器从有价文件上采集到的为厚度信 号。
步骤 S12、 在所述检测对象的图像信号中确定所述有价文件的图像信 号;
其中, 由于检测对象包括有价文件和参考单元, 而且后续步骤需要用 到有价文件的图像信号, 所以需要从检测对象的图像信号中切割出有价文 件的图像信号。
步骤 S13、 当在所述有价文件的预设区域的图像信号与在模板信息的 所述预设区域的图像信号相匹配时, 则根据所述图像传感器与所述参考单 元的位置关系和所述第一传感器与所述参考单元的位置关系来确定所述图 像传感器和所述第一传感器的位置关系, 所述模板信息包括图像信号和第 二信号; 其中,预设区域可大可小, 工作人员可以对预设区域进行预先设定好, 例如, 可以预设区域可以为人民币头像区域, 如果有价文件的图像信号和 模板信息的图像信号相匹配, 则可以通过当前这个模板信息的面额, 例如 当前模板信息的面额为 50元, 那么说明有价文件的面额也为 50元。 在这 个步骤中,仅仅能够判断用户投入的面额值, 而不能确定有价文件的真伪。
步骤 S14、 确定所述模板信息在预设位置上的第二信号;
其中, 预设位置上的第二信号可以为磁信号, 也可以为厚度信号, 或 者为磁信号和厚度信号。
步骤 S15、 确定所述检测对象的预设位置与所述图像传感器的位置关 系;
步骤 S16、 根据所述图像传感器与所述第一传感器的位置关系确定所 述检测对象的预设位置与所述第一传感器的位置关系;
其中, 通过上述两步骤, 进一步确定检测对象的预设位置和第一传感 器的位置关系。
步骤 S17、 判断所述第一传感器采集所述检测对象在预设位置上的第 一信号与所述模板信息在预设位置上的第二信号是否相同; 若是, 则进入 步骤 S18; 否则, 进入步骤 S19。
其中, 这个判断为纠错判断, 即判断检测对象的第一信号与模板信息 的第二信号是否相同, 模板信息的第二信号是理想状态下获取的信号, 检 测对象的第一信号即使为偏移状态下获取的也不会影响判断的正确性, 因 为里面进行位置关系的转换。
步骤 S18、 显示所述有价文件是真币; 步骤 S19、 显示所述有价文件为假币。
在图 1所示的实施例中, 无论有价文件、 图像传感器或第一传感器是 否发生偏移,都不会影响鉴别有价文件的效果, 因为在鉴别有价文件之前, 需要利用参考单元来确定图像传感器和第一传感器的位置关系, 从而避免 了因各个传感器发生偏移而导致鉴别有价文件发生错误; 而且, 在鉴别有 价文件时, 即使有价文件发生偏移, 通过检测对象和传感器之间位置关系 的转化, 也可保证鉴别有价文件的正确性。 因此, 本发明提供的方案提高 了鉴别装置的识别精度和工作效率, 从而给用户提供了便利。
请参见图 2所示, 图 2所示的为一种有价文件鉴别装置的模块图, 该 装置包括: 图像传感器 1 , 用于采集检测对象的图像信号, 所述检测对象 包括所述有价文件和参考单元; 第一传感器 2, 用于采集检测对象的第一 信号; 第一确定模块 3 , 用于在所述检测对象的图像信号中确定所述有价 文件的图像信号; 第二确定模块 4, 用于当在所述有价文件的预设区域的 图像信号与在模板信息的所述预设区域的图像信号相匹配时, 则根据所述 图像传感器 1与所述参考单元的位置关系和所述第一传感器 2与所述参考 单元的位置关系来确定所述图像传感器 1和所述第一传感器 2的位置关系, 所述模板信息包括图像信号和第二信号; 第三确定模块 5 , 用于确定所述 模板信息在预设位置上的第二信号; 第四确定模块 6, 用于确定所述检测 对象的预设位置与所述图像传感器 1的位置关系; 第五确定模块 7, 根据 所述图像传感器 1与所述第一传感器 2的位置关系确定所述检测对象的预 设位置与所述第一传感器 2的位置关系; 判断模块 8, 用于判断所述第一 传感器 2采集所述检测对象在预设位置上的第一信号与所述模板信息在预 设位置上的第二信号是否相同; 显示模块 9, 用于显示所述有价文件是真 币或 币。
上面筒要概括了本发明的设计思想, 下面结合具体实例来介绍本发明 的方案。 由于在具体的实现中, 磁信号检测和厚度信号检测的设计思路大 体相同,所以下面实例仅介绍磁信号检测的部分,根据磁信号检测的原理, 完全可以采用相同的实现方式来实施, 对厚度传感器采集到的厚度信号进 行精确定位和检测, 完成对当前有价文件的最终识别, 避免内容重复在此 不再赘述。
实施例一
本实施例为本发明提供的有价文件鉴别方法的具体介绍。
一种有价文件鉴别方法, 包括:
步骤 1.侦测当前有价文件
图 3为当前有价文件的物理形状图, 通过传输通道来传送待鉴别的有 价文件(例如纸币), 探测传感器设置在传送通道内, 用于侦测当前有价文 件到来, 根据设备设定数据获取当前有价文件的传送速度 V (英寸 /秒)。
步骤 2.协同控制
在当前有价文件的始端到达探测传感器时, 协同控制部控制闭环式传 送带运行, 并且控制闭环式传送的线速度与有价文件传送速度 V—致。 闭 环式传送带传送参考单元从起始位置开始、 以相同的速度依次经过图像传 感器、 磁传感器和厚度传感器。
步骤 3. 图像采集
图像传感器由检测透过图像的投射传感器和检测反射图像的反射传感 器构成,在当前有价文件的始端到达探测传感器时, 图像传感器开始工作, 图像数据取得部以逐行线扫描的方式获取参考单元和当前有价文件的投射 图像和反射图像。
步骤 4. 磁信号采集
磁信号取得部由检测参考单元和当前有价文件的磁信号数据的磁传感 器组成, 当前有价文件的始端到达探测传感器时, 磁传感器开始工作, 磁 信号取得部以逐行扫描的方式获取参考单元和当前有价文件的磁信号数 据。
步骤 5. 厚度信号采集
当前有价文件的始端到达探测传感器时, 厚度传感器开始工作, 厚度 信号取得部以逐行扫描的方式获取参考单元和当前有价文件的厚度信号数 据。
步骤 6. 停止信号采集
从当前有价文件始端到达传感器 Detction开始, 经过时间 Γ (秒), 结 束图像传感器、 磁传感器和厚度传感器数据采集工作。
图 4是图像传感器采集到的参考单元和当前有价文件的图像。
图像数据行数为:
Figure imgf000013_0001
(式 6.1)
图像数据列数为:
C = T,
(式 6.2) 图像传感器水平宽度为 Φ英寸,水平分辨率7 ^dpi,垂直分辨率为 dpi。 由于磁传感器为线扫描, 一次采集的磁信号数据为 w个离散的数据, 其中 W为磁传感器磁头的个数。 当前有价文件完整的磁信号数据为 M行 N 列的数值矩阵 Μ^ , 如(式 6.3 )所示。 d0 d0 d0 d0
d、, d, d, d,
Mag
Figure imgf000014_0001
(式 6.3)
其中
M=VT 同样的, 由于厚度传感器为线扫描, 一次采集的磁信号数据为 个离 散的数据, 其中 为厚度传感器的个数。 当前有价文件完整的磁信号数据 为^:行 Γ列的数值矩阵 %', 如(式 6.4)所示。
^00 ^01 ^OJ-l
^10
Thi
t K—1 ^κ-ι,τ-ι
tK,. ^κ,τ-ι
(式 6.4)
其中
K=VTL 步骤 7. 图像识别
步骤 7.1图像切割 利用图像传感器所采集到的图像数据对当前有价文件进行初步识别, 因对于识别有用的前景区域在整个图像中位置存在不确定性, 所以首先应 通过边界探测和图像分割将前景区域切割出来。
以图像传感器采集到的整个图像的左边界所在直线方向为 y轴、 图像 传感器采集到的整个图像的下边界所在直线方向为 轴、 图像传感器采集 到的整个图像的左下顶点为坐标原点建立坐标系 (即像素坐标系, 坐标 系的每一个点表示一个像素点 ), 如图 5所示。
殳设前景区域的四个顶点分别为 A, B, ( 和0。 通过探测前景区域的 边界点, 再利用边界点进行线性拟合, 可得到
左边界直线方程(即边 AB所在直线方程;):
y = kxx + ^
(式 7.1)
右边界直线方程(即边 CD所在直线方程;):
y = k2x -\- b2
(式 7.2 )
上边界直线方程(即边 AD所在直线方程;):
y = k x + b3
(式 7.3 )
下边界直线方程(即边 BC所在直线方程;):
y = k4x + b4
(式 7.4 )
根据前景区域四个边界直线, 完成图像切割, 并将切割出来的前景区 域利用现有图像旋转技术进行旋转纠偏。 以当前有价文件前景区域左边界 所在直线为纵坐标方向、 以当前有价文件前景区域下边界所在直线为横坐 标方向、 以当前有价文件前景区域左下顶点为坐标原点建立坐标系 y中 如图 6所示。
步骤 7.2 图像特征模板匹配识别
完成上述操作后, 根据存储部保存的标准模板数据, 选择前景区域的 相应位置进行模板匹配识别, 如图 7所示。 通过模板匹配识别得到图像识 别的结果, 如果图像识别失败, 则退出当前有价文件。 为了叙述方便, 假 设图像识别结果为 RMB100元, 正面正立。
步骤 8. 磁信号识别
根据上述假设, 识别方案已经完成了对当前有价文件的图像识别, 识 别结果为 CNY100元。 根据对 CNY100元的先验知识, 纸币上 M能被磁 化(磁信号波动剧烈), 为了检验当前有价文件磁信号是否满足 CNY100 元对 M处的磁信号特征要求, 首先需要确定有价文件 M处的磁信号在磁 传感器采集到的整个磁信号中的位置。
步骤 8.1 确定图像传感器和参考单元物理位置关系
( 1 )确定参考单元中心点 w在像素坐标系 中的坐标
利用现有技术, 结合对该识别系统参考单元成像位置的先验知识, 很 容易在搜素区域8_∑^3中求得 W在坐标系" 中的坐标, 如图 8中所示。 设
Figure imgf000016_0001
Drx和 Dry均为长度单位, Drx是 W点到 Y轴的距离, Dry是 W点 到 X轴的距离。
( 2 )确定参考单元中心点 W在物理坐标系 ^。 中的坐标
建立物理尺寸坐标系 (以图像传感器 13左端点为坐标原点, 图像 传感器 13水平方向为横坐标方向,走纸方向为纵坐标方向),如图 9所示。
设想参考单元静止, 而图像传感器以相反于走纸方向的速度扫描, 则 从图像传感器固定位置到参考单元垂直方向中心线 Line_y位置包含的像 素行数为:
VTL -D 根据图像传感器的垂直分辨率为 dpi , 故图像传感器水平方向所在直 线和参考单元垂直方向中心线 Line_y间的物理尺寸距离为:
VTh - D^
根据图像传感器的水平分辨率为 , 可得参考单元中心点 W在物理尺 寸坐标系 中的坐标为
Xw ―
即参考单元中心点 W 到坐标系 "^^的纵坐标轴的物理尺寸距离为
£^
( 3 )确定参考单元中心点 W的在物理坐标系 中的坐标
建立物理尺寸坐标系 (以参考单元 12 的左顶点为坐标原点, 参 考单元 12水平方向为横坐标方向,走纸方向为纵坐标方向),如图 9所示。 可得参考单元 12中心点 W在坐标系 中的坐标为 D,
即参考单元中心点 w 到坐标系 的纵坐标轴的物理尺寸距离为
( 4 )确定物理坐标系 与物理坐标系 的位置关系(即图像传感 器 13物理与参考单元 12物理位置关系), 不妨假设 2 , 即参考单元 12左端点比图像传感器 13左端点更靠左边, 如图 9中所示。 由此, 物理 尺寸坐标系 ^ y和 满足如下关系:
D D
Χ (; ■)
2 Ί
VTL. - D
Y = y+ (:
(式 8.1 )
8.2确定磁传感器和参考单元物理位置关系
( 1 )确定参考单元中心点 W磁信号数据的在整个磁信号数据矩阵 中的位置在
Figure imgf000018_0001
Mag d,,„ d,. d 中, 根据该识别系统参考单元磁信号先验知识, 利用现有搜索技术很 容易确定 W处的磁信号数据位于矩阵 M^中位置, 假设求得 W处的磁信 号数据位于矩阵 M^中第 DD^列, 即
Wdata = Mag(Dwy ,D ^
( 2 )确定参考单元中心点 W在物理尺寸坐标系 ^"。" 中的坐标 建立坐标系 ^。" " (以磁传感器 14左边第一个磁头所在位置为坐标 原点, 磁传感器 14水平方向为横坐标方向, 走纸方向为纵坐标方向), 如 图 9中所示。 同样假设参考单元静止, 由 (1 ) 中所得, 磁传感器 14以相 反于走纸方向的速度扫描至参考单元 12中心点 W时, 共采集磁信号矩阵 行数为 β 。 根据磁传感器 14磁头个数 Ν, 垂直分辨率 Lmdpi, 有 W在物 理尺寸坐标系 ywy"的坐标为
Figure imgf000019_0001
( 3 )物理尺寸坐标系 "^^"y''与物理尺寸坐标系 的位置关系(即磁 传感器 14与参考单元 12之间的位置关系) 不妨假设 >Ι'Φ , 即参考单元 12左端点比磁传感器 14左端点更 靠左边, 如图 9中所示。 由此, 物理尺寸坐标系 "^^"y"和 满足如下关 系:
Figure imgf000019_0002
(式 8.2 )
8.3 确定磁传感器和图像传感器的位置关系
一般情况下, 均假设图像传感器 13和磁传感器 14是理想对准的, 即 T(xoy,x"o''y")满足:
X = X
'=y+Ay 其中 表示图像传感器 13和磁传感器 14在传送通道上的间隔距离。 但根据背景技术中所述, 由于制作工艺限制以及有价文件识别系统长时间 使用等情况, 各个传感器之间并不是理想对准的。 因此联立式(8.1 )和式
(8.2),
D D
X=x+( .)
2 Ί\
VTL -D
Y = y+(:
D D
X =Λ;"+(· )
2 N
D,
容易得到物理尺寸坐标系 JC"。" ;"和物理尺寸坐标系 之间的真实的 位置关系1 ^xoy^'oH (即是磁传感器 14左端点与图像传感器 13左端点 的空间物理位置对应关系)
Figure imgf000020_0001
,, VTL -D^ D
y =y+(.——;—— -) 化简得 χ"=χ+ (^·Φ-¾
Ν
,, VTL^D^ Dw
y =y+(.— - ~~ -) -—
L,' L
(式 8.3 ) 从式 8.3中可知, 磁传感器 14和图像传感器 13的位置关系与参考单 元 12的设定无关, 即参考单元 12只是充当识别系统内利用图像传感器 13 位置定位磁传感器 14位置的过渡桥梁,对识别系统不会产生任何的附加误 差。
8.4 图像信号与磁信号融合, 利用图像信号定位磁信号
根据图像识别的结果(即假设当前有价文件为 CNY100元,正面正立), 根据对 CNY100元正面正立的先验知识,假设 M距离当前有价文件上边缘 物理距萬为 dMQ , M距离当前有价文件左边缘物理距萬为 dwp , 如图 10所 示。
在像素坐标系中^ 如图 11所示, 联立有价文件前景区域左边界所 在直线方程和上边界所在直线方程
y = ^ + ^
y = k x + b 求得左上顶点 A在像素坐标系中的坐标, 假设为(^,)^;
联立前景区域左边界所在直线方程和下边界所在直线方程
y = / x +
y = k4x + b4 求得左下顶点 B在像素坐标系中的坐标, 假设为( , ^;
联立前景区域右边界所在直线方程和上边界所在直线方程
y = k2x + b2
y = k x -\- b 求得右上顶点 D在像素坐标系中的坐标, 假设为( ,>^;
在图像传感器停止扫描时刻 (即经过时间 Γ秒, 参考单元和当前有价 文件全部通过图像传感器、 磁传感器和厚度传感器)有价文件在物理尺寸 坐标系 中的位置如图 12所示。 ^据图像传感器的水平分辨率7^ dpi, 垂 直分辨率为 dpi可得有价文件顶点 A, B, D在物理尺寸坐标系 中的 坐标分别为:
XA
_ yA
Jig
_ yB
yB
Figure imgf000022_0001
用当前有价文件的左上顶点 Α(^, ^和右上顶点 D^^y^可求得当 前有价文件上边界在物理尺寸坐标系 中的直线方程为:
Figure imgf000022_0002
化筒为: y = Ti(yD- yA) χ XA yD-xD yA
L{ L{ 由有价文件上 M距离有价文件上边界的物理距离为 dMQ , 在物理尺寸 坐标系 中, 经过 M且平行于当前有价文件上边界所在直线 linel (如图 12所示) 的直线方程为: y =
Figure imgf000023_0001
T yD-yA)
Figure imgf000023_0002
则 linel的直线方程化为:
y = kM1x+bA
(式 8.4 ) 用当前有价文件的左上顶点 A(JCA, )和左下顶点 B(JCB,^)可求得当前 有价文件左边界在物理尺寸坐标系 。y中的直线方程为:
Figure imgf000023_0003
化筒为: y = Ti(yB- yA χ ] XA yB-xB yt
L{ (XB— A ) L{ (XB— A ) 故在物理尺寸坐标系 。y中, 经过 M且平行于当前有价文件左边界所 在直线 line2 (如图 12所示) 的直线方程为: y
Figure imgf000023_0004
T yB-yA)
Figure imgf000024_0001
XA yB-xB yA
Figure imgf000024_0002
则 line2的直线方程化为: y =kM2x+bk
(式 8.5 )
联立直线方程(式 8.4 )和(式 8.5 )
Figure imgf000024_0003
可求得当前有价文件上 M处在物理尺寸坐标系 。y中的坐标为
Figure imgf000024_0004
才艮据式 8.3 中所描述的物理尺寸坐标系 'v'y"和物理尺寸坐标系 之间的位置关系 (即是磁传感器左端点与图像传感器左端点的空间物理位 置对应关系 ), 可求得当前有价文件上 M处在物理尺寸坐标系 "^^"y"中的 坐标为
Figure imgf000024_0005
, VT -D^ Dwy
yM =yM + ( : )-
Figure imgf000025_0001
(式 8.6)
设:
XM r b -b ,Dwx ^ Drx ^ N
N T Φ D、
Figure imgf000025_0002
则 M的磁信号数据位于整个磁信号数据矩阵中行 ^列 满足
行位置: =[ ]或^ =[ ] + 1. 歹 ij位置: CM =[dN]^CM =[^] + 1_ 其中 ["]表示对 "进行取整操作。 至此, 完成了从图像传感器信号、 利 考单元为过渡桥梁对磁传感器信号的定位。
8.5 磁信号精确鉴别
根据上述定位, 在磁信号矩阵
d0 d0 d0 d0
d、, d、 d、 d、
Mag
Figure imgf000025_0003
中, 第 行<^列数据即是当前有价文件上 M处的磁信号数据。 根据图像识别中对识别结果为 RMB100 元, 正面正立的假设, 检测
Mflg(?M,CM)邻域磁信号数据是否满足真实 RMB100元, 正面正立时 M处 的磁信号规律。 如果满足, 则判定磁信号识别结果也是 RMB100元, 与图像识别结果 一致。
如果不满足, 则判定磁信号识别结果不是 RMB100元, 与图像识别结 果不一致, 则当前有价文件判定为非真实的 RMB100元, 拒收。 实施例二
本实施例为本发明提供的有价文件鉴别装置的具体介绍。
请参见图 13所示, 图 13所示的为一种有价文件鉴别装置, 包括: 控制部 16: 控制整个装置的工作流程;
传送系统 20: 关于传送系统的具体细节部分请参见图 14和 15所示。 在传送系统通道上壁 23 上至少安装有两个获取有价文件机读信号的传感 器, 两条闭环式传送带 22的两端依靠传动轴固定在通道侧壁 25上, 当有 价文件在传送通道传送时, 传动轴 21带动传送带 22运行。 本发明优选如 下传感器类型:
图像传感器 13: 以透射和反射的方式取得投入的有价文件的图像数 据,假设本实施方式中图像传感器 13水平宽度为 Φ英寸,水平分辨率 dpi, 垂直分辨率为 ^ dpi, 如图 16中所示;
磁传感器 14:即磁信号数据取得部,取得投入有价文件的磁信号数据, 假设磁传感器 14水平方向有 w个磁头,均有分布在宽度为 Φ英寸的水平方 向上, 磁传感器垂直分辨率为^(1 如图 16中所示;
厚度传感器 15: 即厚度数据取得部, 取得投入有价文件的厚度数据, 假设厚度传感器 15水平方向有 r个厚度感应头, 均有分布在宽度为 Γ英寸 的水平方向上, 厚度传感器垂直分辨率为 Ltdpi, 如图 16中所示; 参考单元 12: 两个用于确定不同传感器间空间位置约束关系的矩形参 考单元,设其物理尺寸长度为 D ,设参考单元的水平方向中心线为 Line_x, 垂直方向中心线为 Line_y, 两直线交点为 W, 则 W为参考单元的中心点, 如图 17所示。 每个参考单元 12的两端分别搭接在两条闭环式传送带上, 当传送带运动时, 参考单元随着传送带一同运动, 如图 14和图 15所示。
该参考单元 12具有特征:
①该参考单元以正中为分界线, 左半部分成黑色(不透光), 右半部分 成透明;
②该参考单元成黑色部分能磁化;
③该参考单元成黑色部分(左半部分)厚度明显大于透明部分(右半 部分) 的厚度;
当有价文件在传送通道上传送时, 该参考单元以附着在传送带上、 以 相同的速度通过图像传感器、磁传感器和厚度传感器; 即每次图像传感器、 磁传感器和厚度传感器采集到的信号都同时包括该参考单元的相应信号和 当前有价文件的相应信号;
协同控制部 21 : 如图 13 中所示, 协同控制闭环式传送带线速度和有 价文件传送速度相同,让参考单元的传送速度和有价文件的传送速度一样。
同时,协同控制部 21还控制参考单元的复位, 即每结束一张有价文件 的信号采集, 协同控制部控制闭环式传送带将参考单元传送到指定的起始 位置, 如图 14和图 15中所示。 该起始位置位于探测传感器与图像传感器 之间。 存储部 18: 存储识别当前有价文件图像信号时需要的标准模板; 处理器 17: 识别图像信号、 磁信号和厚度信号, 完成对当前有价文件 的鉴别;
显示部 19: 若当前有价文件(如纸币 )识别为合法文件时(如真钞), 将显示币种、 面额、 版本等信息; 若当前有价文件识别为不真实时, 将显 示 "Not Accept!"。
本发明所述的有价文件鉴别装置还进一步包括
探测传感器 11 : 布置在传送壁的开始位置, 侦测有价文件的到来, 并 触发其他传感器进入工作模式, 如图 14和图 15中所示;
本发明中的处理器 17进一步包括如下模块(属于处理器下的软模块), 如图 18所示:
图像切割部 31 : 将对于识别有用的前景区域从背景区域中切割出来; 图像纠偏部 32: 判断前景区域是否存在倾斜和偏移, 采用现有纠偏技 术实施图像纠偏;
图像识别部 33: 根据当前有价文件的图像数据与存储部存储的标准模 板实施匹配鉴别;
图像传感器与磁传感器空间位置关系修定部 34: 利用参考单元在图像 传感器采集的到整个图像中的分布特点 和 参考单元的磁信号在磁传感器 采集到的整个磁信号中的分布特点, 修正图像传感器与磁传感器的空间位 置关系, 得到两种传感器当前位置下的真实位置约束;
图像信息与磁信息融合部 35: 图像信息和磁信息是由不同类型的数据 采集单元采集的对于同一待鉴别介质的不同类型信息描述。 虽然是两种不 同类型的信息, 但二者却存在某种关联, ,因为有价文件的印刷是十分精密 的, 即图像信息 (特别是图像特征信息)与磁信号 (特别是磁特征信号) 在同种有价文件上的物理位置分布是十分稳定可靠的, 同时当前有价文件 的传输速度以及图像传感器、 磁传感器的分布位置和分辨率是固定的。 利 用这种固定关系, 可根据图像信息, 结合图像传感器和磁传感器在物理位 置上的对应关系, 计算当前有价文件的有效磁信号分布。
磁信号识别部 36: 基于上述多源信息融合的基础上, 鉴别当前有价文 件的磁信号, 得到对当前有价文件的磁信号识别结果;
图像传感器与厚度传感器空间位置关系修定部 37: 利用识别系统参考 单元在图像传感器采集到的整个图像中的分布特点和参考单元的厚度信号 在厚度传感器采集到的整个厚度信号中的分布特点修正图像传感器与厚度 传感器的空间位置关系, 得到两种传感器当前状态下的真实位置约束; 图像信息与厚度信息融合部 38: 图像信息和厚度信息同样是由不同类 型的数据采集单元采集的对于同一待鉴别介质的不同类型信息描述。 二者 同样存在着类似于图像信号与磁信号般的相互关系。 利用这种关系, 可根 据图像信息, 结合图像传感器和厚度传感器在物理位置上的对应关系, 计 算当前有价文件的有效厚度信号分布。
厚度信号识别部 39: 基于上述多源信息融合的基础上, 鉴别当前有价 文件的厚度信号, 得到对当前有价文件的厚度信号识别结果;
结果输出部 40: 输出当前有价文件最终的识别结果。
本发明的实施流程如图 18所示:
在本发明所提出的有价文件鉴别装置中, 图像传感器、 磁传感器和厚 度传感器分别按时间序列获得了参考单元和当前有价文件的观测数据。 因 为在本发明提出的有价文件鉴别装置中, 当前有价文件的传输速度一定, 图像传感器、 磁传感和厚度传感器的空间分布位置是不变的, 故本发明首 先利用参考单元的图像信号、 磁信号和厚度信号确定图像传感器、 磁传感 器和厚度传感器的真实的空间位置关系, 再利用各个传感器位置之间的这 种约束关系, 将图像传感器、 磁传感器和厚度传感器获取的信号进行融合 和识别, 以获得对当前有价文件一致性的描述和识别。
结合上述装置中的部件, 下面筒要介绍这些部件之间的工作流程, 步骤 S21、 投入待鉴别的有价文件(如纸币), 流程开始;
步骤 S22、 图像传感器处理, 扫描参考单元和当前有价文件, 获取参 考单元和当前有价文件的图像信息;
步骤 S23、 磁传感器处理, 获取参照结构单元和当前有价文件的磁信 号数据;
步骤 S24、 厚度传感器处理, 获取参照结构单元和当前有价文件的厚 度信号数据;
步骤 S25、 图像切割, 首先求出前景区域在整个背景区域中的位置关 系, 将识别有用的图像前景区域从背景区域中切割出来;
步骤 S26、 图像纠偏, 对存在倾斜和偏移的图像进行纠偏, 方便后续 的图像匹配识别;
步骤 S27、 图像识别, 将图像切割所得到的有价文件的前景区域置于 坐标系中, 选择特定位置的图像数据与存储部所存储的标准模板数据进行 匹配识别, 得到图像识别结果。 如果图像识别失败, 则退出当前有价文件; 步骤 S28、 修正图像传感器和磁传感器物理位置关系, 利用参考单元 的图像信号和磁信号, 得到图像传感器左顶点和磁传感器左顶点在水平方 向和垂直方向的位移; 由此获得图像传感器和磁传感器之间真实的约束关 系;
步骤 S29、 图像信息与磁信息融合, 根据当前有价文件传输速度不变、 图像传感器和磁传感器物理位置约束关系, 并根据图像识别的结果, 利用 当前有价文件有效区域在背景图像中的位置(分布),计算出当前有价文件 有效的磁信号在磁传感器采集的整个磁信号中的位置(分布)。
步骤 S30、 磁信号识别, 根据图像信息与磁信息融合的情况, 判定磁 传感器采集到的磁信号数据特定位置磁信号规律是否满足当前有价文件图 像识别结果所要求的磁信号分布规律。 若磁信号识别失败, 则退出当前有 价文件;
步骤 S31、 修正图像传感器和厚度传感器物理位置关系, 利用参考单 元的图像信号和厚度信号, 确定图像传感器左顶点和厚度传感器左顶点在 水平方向和垂直方向的位移; 由此获得图像传感器和厚度传感器之间真实 的约束关系;
步骤 S32、 图像信息与厚度信息融合, 根据当前有价文件传输速度不 变、 图像传感器和厚度传感器物理位置不变的规则, 并根据图像识别的结 果, 利用当前有价文件有效区域在背景图像中的位置(分布), 计算出当前 有价文件有效的厚度信号在厚度传感器采集的整个厚度信号中的位置(分 布)。
步骤 S33、 厚度信号识别, 根据图像信息与厚度信息融合的情况, 判 定厚度传感器采集到的厚度信号数据特定位置厚度信号规律是否满足当前 有价文件图像识别结果所要求的厚度信号分布规律。若厚度信号识别失败, 则退出当前有价文件;
步骤 S34、 若厚度识别成功, 则完成了对当前有价文件的识别, 输出 识别结果(如币种、 面额、 版本等信息)。
需要说明的是,图 1至图 18所示的实施例只是本发明所介绍的优选实 施例, 本领域技术人员在此基础上, 完全可以设计出更多的实施例, 因此 不在此处赘述。
对这些实施例的多种修改对本领域的专业技术人员来说将是显而易 见的, 本文中所定义的一般原理可以在不脱离本发明的精神或范围的情况 下, 在其它实施例中实现。 因此, 本发明将不会被限制于本文所示的这些 实施例,而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。

Claims

1、 一种有价文件鉴别方法, 其特征在于, 包括:
通过图像传感器和第一传感器采集检测对象的图像信号和第一信号, 所述检测对象包括所述有价文件和参考单元;
在所述检测对象的图像信号中确定所述有价文件的图像信号; 当在所述有价文件的预设区域的图像信号与在模板信息的所述预设区 域的图像信号相匹配时, 则根权据所述图像传感器与所述参考单元的位置关 系和所述第一传感器与所述参考单利 _元的位置关系来确定所述图像传感器和 所述第一传感器的位置关系, 所述模 3板 1 信息包括图像信号和第二信号; 要
确定所述模板信息在预设位置上的第二信号;
确定所述检测对象的预设位置与所述图求像传感器的位置关系; 根据所述图像传感器与所述第一传感器的位置关系确定所述检测对象 的预设位置与所述第一传感器的位置关系;
判断所述第一传感器采集所述检测对象在预设位置上的第一信号与所 述模板信息在预设位置上的第二信号是否相同, 若是, 则显示所述有价文 件是真币; 否则, 显示所述有价文件为假币。
2、根据权利要求 1所述的有价文件鉴别方法, 其特征在于, 所述第一 传感器具体为磁传感器, 所述第一信号和第二信号均为磁信号。
3、根据权利要求 1所述的有价文件鉴别方法, 其特征在于, 所述第一 传感器具体为厚度传感器, 所述第一信号和第二信号均为厚度信号。
4、根据权利要求 1所述的有价文件鉴别方法, 其特征在于, 所述第一 传感器包括磁传感器和厚度传感器, 所述第一信号和第二信号均包括磁信 号和厚度信号。
5、 一种有价文件鉴别装置, 其特征在于, 包括:
图像传感器, 用于采集检测对象的图像信号, 所述检测对象包括所述 有价文件和参考单元;
第一传感器, 用于采集检测对象的第一信号;
第一确定模块, 用于在所述检测对象的图像信号中确定所述有价文件 的图像信号; 第二确定模块, 用于当在所述有价文件的预设区域的图像信号与在模 板信息的所述预设区域的图像信号相匹配时, 则根据所述图像传感器与所 述参考单元的位置关系和所述第一传感器与所述参考单元的位置关系来确 定所述图像传感器和所述第一传感器的位置关系, 所述模板信息包括图像 信号和第二信号;
第三确定模块, 用于确定所述模板信息在预设位置上的第二信号; 第四确定模块, 用于确定所述检测对象的预设位置与所述图像传感器 的位置关系;
第五确定模块, 根据所述图像传感器与所述第一传感器的位置关系确 定所述检测对象的预设位置与所述第一传感器的位置关系;
判断模块, 用于判断所述第一传感器采集所述检测对象在预设位置上 的第一信号与所述模板信息在预设位置上的第二信号是否相同;
显示模块, 用于显示所述有价文件是真币或假币。
6、根据权利要求 5所述的有价文件鉴别装置, 其特征在于, 所述第一 传感器具体为磁传感器, 所述第一信号和第二信号均为磁信号。
7、根据权利要求 5所述的有价文件鉴别装置, 其特征在于, 所述第一 传感器具体为厚度传感器, 所述第一信号和第二信号均为厚度信号。
8、根据权利要求 5所述的有价文件鉴别装置, 其特征在于, 所述第一 传感器包括磁传感器和厚度传感器, 所述第一信号和第二信号均包括磁信 号和厚度信号。
PCT/CN2013/073647 2012-10-08 2013-04-02 一种有价文件鉴别方法及装置 WO2014056311A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
AU2013330103A AU2013330103B2 (en) 2012-10-08 2013-04-02 Valuable document authentication method and device
EP13844647.1A EP2905754B1 (en) 2012-10-08 2013-04-02 Valuable document authentication method and device
US14/426,969 US9406183B2 (en) 2012-10-08 2013-04-02 Valuable document authentication method and device
ZA2015/02610A ZA201502610B (en) 2012-10-08 2015-04-17 Valuable document authentication method and device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2012103781916A CN102890841B (zh) 2012-10-08 2012-10-08 一种有价文件鉴别方法及装置
CN201210378191.6 2012-10-08

Publications (1)

Publication Number Publication Date
WO2014056311A1 true WO2014056311A1 (zh) 2014-04-17

Family

ID=47534332

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/073647 WO2014056311A1 (zh) 2012-10-08 2013-04-02 一种有价文件鉴别方法及装置

Country Status (7)

Country Link
US (1) US9406183B2 (zh)
EP (1) EP2905754B1 (zh)
CN (1) CN102890841B (zh)
AU (1) AU2013330103B2 (zh)
CL (1) CL2015000824A1 (zh)
WO (1) WO2014056311A1 (zh)
ZA (1) ZA201502610B (zh)

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102722708B (zh) * 2012-05-16 2015-04-15 广州广电运通金融电子股份有限公司 一种薄片类介质分类方法及其装置
CN102890841B (zh) 2012-10-08 2013-11-27 广州广电运通金融电子股份有限公司 一种有价文件鉴别方法及装置
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
JP2015095154A (ja) * 2013-11-13 2015-05-18 株式会社東芝 紙葉類処理装置
CN103809137B (zh) * 2014-02-21 2016-08-31 中国人民银行印制科学技术研究所 纸页检测装置和纸页检测方法
CN105469082B (zh) * 2014-09-12 2019-02-26 威海新北洋荣鑫科技股份有限公司 金融票据处理方法和装置
CN105528825B (zh) 2015-12-02 2018-08-31 广州广电运通金融电子股份有限公司 有价文件自适应识别方法和装置
CN106296970B (zh) * 2016-08-11 2022-05-20 威海华菱光电股份有限公司 薄膜的检测系统与检测方法
DE102017001947A1 (de) * 2017-02-28 2018-08-30 Giesecke+Devrient Currency Technology Gmbh Magnetische Prüfung von Wertdokumenten
CN107134046B (zh) * 2017-05-02 2019-08-23 深圳怡化电脑股份有限公司 一种纸币厚度异常检测方法及装置
CN111602178B (zh) * 2018-01-11 2022-03-11 三菱电机株式会社 图像读取装置
CN111476937B (zh) * 2020-04-18 2021-11-16 武汉卓目科技有限公司 金融机具二维信号辅助一维信号鉴伪的方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030021459A1 (en) * 2000-05-24 2003-01-30 Armando Neri Controlling banknotes
CN101520912A (zh) * 2008-02-29 2009-09-02 富士通先端科技株式会社 纸张种类鉴别装置及纸张种类鉴别方法
CN101925931A (zh) * 2008-01-25 2010-12-22 光荣株式会社 纸张类识别装置
CN102890841A (zh) * 2012-10-08 2013-01-23 广州广电运通金融电子股份有限公司 一种有价文件鉴别方法及装置

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5790693A (en) * 1990-02-05 1998-08-04 Cummins-Allison Corp. Currency discriminator and authenticator
JP4450966B2 (ja) * 2000-10-03 2010-04-14 富士通株式会社 紙葉類鑑別装置
WO2011086665A1 (ja) * 2010-01-12 2011-07-21 グローリー株式会社 紙葉類識別装置および紙葉類識別方法
CN101916470B (zh) * 2010-08-03 2013-04-24 广州广电运通金融电子股份有限公司 纸质有价文件识别方法及其识别系统
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
CN102289857B (zh) * 2011-05-19 2013-09-25 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系统
CN102592352B (zh) * 2012-02-28 2014-02-12 广州广电运通金融电子股份有限公司 纸页类介质识别装置及识别方法
CN103035061B (zh) * 2012-09-29 2014-12-31 广州广电运通金融电子股份有限公司 有价文件防伪特征生成方法及其鉴别方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030021459A1 (en) * 2000-05-24 2003-01-30 Armando Neri Controlling banknotes
CN101925931A (zh) * 2008-01-25 2010-12-22 光荣株式会社 纸张类识别装置
CN101520912A (zh) * 2008-02-29 2009-09-02 富士通先端科技株式会社 纸张种类鉴别装置及纸张种类鉴别方法
CN102890841A (zh) * 2012-10-08 2013-01-23 广州广电运通金融电子股份有限公司 一种有价文件鉴别方法及装置

Also Published As

Publication number Publication date
US20150220936A1 (en) 2015-08-06
CN102890841B (zh) 2013-11-27
AU2013330103B2 (en) 2016-05-19
ZA201502610B (en) 2016-01-27
AU2013330103A1 (en) 2015-03-26
US9406183B2 (en) 2016-08-02
CL2015000824A1 (es) 2015-06-12
EP2905754A1 (en) 2015-08-12
EP2905754A4 (en) 2015-10-07
CN102890841A (zh) 2013-01-23
EP2905754B1 (en) 2020-08-05

Similar Documents

Publication Publication Date Title
WO2014056311A1 (zh) 一种有价文件鉴别方法及装置
RU2527203C2 (ru) Устройство обработки бумажных листов и способ обработки бумажных листов
US9307206B2 (en) Automatic image capture
JP5616958B2 (ja) 紙幣検出器デバイスのための方法、および紙幣検出器デバイス
US20050169511A1 (en) Document processing system using primary and secondary pictorial image comparison
US20120150745A1 (en) Document imaging and processing system
CN102324134A (zh) 有价文件鉴别方法及其鉴别装置
US20100246928A1 (en) Banknote recognition apparatus and banknote recognition method
JP2008524683A (ja) シート物受入装置
WO2004081887A1 (ja) 紙葉類識別方法及び紙葉類識別装置
KR20030025187A (ko) 지엽류 계수장치 및 거래 처리장치
KR101112445B1 (ko) 수표 인식 장치 및 방법
JP4450966B2 (ja) 紙葉類鑑別装置
KR20120117049A (ko) 휴대용 단말기에서 위조 지폐 감별 방법 및 장치
CN105096446A (zh) 一种折角钞光变油墨鉴伪识别的方法及系统
JP6582548B2 (ja) 記番号認識装置、媒体鑑別装置、自動取引装置、記番号管理装置、記番号管理システム、及び記番号管理プログラム
JP5953553B2 (ja) ディスク判別方法、ディスク判別装置およびディスク選別装置
JP5953614B2 (ja) ディスク判別方法、ディスク判別装置およびディスク選別装置
JP3064739B2 (ja) 画像処理装置
KR20210081488A (ko) 수표인식장치
JPH07311867A (ja) 紙葉類認識装置
JP5216301B2 (ja) 媒体識別装置
KR20090073681A (ko) 자기잉크문자인식장치 및 그 방법
JP2005316727A (ja) 紙葉類識別装置および方法
TWM474980U (zh) 整鈔機系統

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: 13844647

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 14426969

Country of ref document: US

ENP Entry into the national phase

Ref document number: 2013330103

Country of ref document: AU

Date of ref document: 20130402

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 2013844647

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2015000824

Country of ref document: CL

NENP Non-entry into the national phase

Ref country code: DE