EP1353301B1 - Lecteur de fil de sécurité magnétique - Google Patents

Lecteur de fil de sécurité magnétique Download PDF

Info

Publication number
EP1353301B1
EP1353301B1 EP03015595A EP03015595A EP1353301B1 EP 1353301 B1 EP1353301 B1 EP 1353301B1 EP 03015595 A EP03015595 A EP 03015595A EP 03015595 A EP03015595 A EP 03015595A EP 1353301 B1 EP1353301 B1 EP 1353301B1
Authority
EP
European Patent Office
Prior art keywords
digital representation
thread
magnetic
digital
peak
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.)
Expired - Lifetime
Application number
EP03015595A
Other languages
German (de)
English (en)
Other versions
EP1353301A3 (fr
EP1353301A2 (fr
Inventor
John Buttiant
Peter Dilwyn Evans
John Alan Skinner
Michael Potter
Peter Alan Pullan
Barry Clifford Scowen
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.)
Glory Global Solutions Holdings Ltd
Original Assignee
De la Rue International 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 De la Rue International Ltd filed Critical De la Rue International Ltd
Priority claimed from EP02250079A external-priority patent/EP1221679B1/fr
Publication of EP1353301A2 publication Critical patent/EP1353301A2/fr
Publication of EP1353301A3 publication Critical patent/EP1353301A3/fr
Application granted granted Critical
Publication of EP1353301B1 publication Critical patent/EP1353301B1/fr
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
    • G07D7/00—Testing 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
    • G07D7/004—Testing 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 digital security elements, e.g. information coded on a magnetic thread or strip
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
    • G07D7/00—Testing 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
    • G07D7/04—Testing magnetic properties of the materials thereof, e.g. by detection of magnetic imprint

Definitions

  • the present invention relates to a method and apparatus for detecting a security thread, for example for identifying a document from a security code carried by the thread.
  • a security thread This may be a simple metallic thread or comprise segments of magnetic material and segments of non-magnetic material.
  • the segments of magnetic and non-magnetic material such that they represent a code.
  • the segments are arranged into fixed length elements so that they represent a binary word and this word may be repeated several times along the thread.
  • the fixed length elements may have a length of 2mm, the presence of magnetic material indicating a binary 1 and the absence indicating a binary O.
  • This code can be read using a magnetic head or an array of heads. This would typically be done by temporarily magnetising the magnetic material constituting a thread and arranging for the documents to be carried by a transport mechanism so that they pass by the array of magnetic heads, the magnetic material of the thread being in close proximity to the heads.
  • GB 2098768B discloses a coded magnetic thread reader in which a linear array of magnetic heads scans a security thread embedded in a document and stores samples of the signals produced by the magnetic heads in storage devices which are sequentially scanned and compared against a fixed threshold in order to produce a binary bit stream corresponding to the variation of magnetisation along the thread.
  • US-A-5889271 describes a method of detecting a coded thread using multiple reading channels to detect each thread segment but is complex to implement.
  • DE19921653 discloses a banknote with a magnetic security thread. It also discloses a reading head in which a signal is produced by the passage of the thread. The code borne by the thread can be discerned from this signal.
  • a disadvantage of prior techniques is that in order to identify the document bearing the thread, it was necessary to use a technique that involves rotating the code read from the thread through each possible permutation and comparing each of these permutations against each entry in a database. Hence, a 16 bit code would have to be rotated and compared 16 times against each entry in a database.
  • a method of identifying a coded magnetic thread comprising generating a digital representation of the thread and comparing the digital representation with one or more known digital representations characterised in that the method further comprises storing the digital representation of the thread in a circular buffer scanning the digital representation to locate a predetermined code sequence; rotating the digital representation within the circular buffer to locate the predetermined code sequence in a predetermined position corresponding to the position of the predetermined code sequence in a stored version of the or each known digital representation prior to carrying out the comparison step.
  • This method overcomes the limitation of the above mentioned “sliding correlator” technique. By aligning the digital representation into the same format in which the known digital representations are stored, only one comparison is required for each known digital representation.
  • the digital representation is binary.
  • the code is asymmetric and in this case, the comparison is performed against reversed versions of the known digital representations, thereby determining the orientation of the coded magnetic thread.
  • this measurement may be used to determine which face of a sheet document containing the thread is uppermost.
  • the relative displacement of the thread from a known magnetic feature may be measured and used to determine which face of a sheet document containing the thread and magnetic feature is uppermost.
  • the digital representation is scanned for at least one feature which indicates the likelihood that the digital representation is valid.
  • an apparatus for identifying a coded magnetic thread comprising a magnetic field detector, a processing system for processing signals generated by the detector to generate a digital representation of the thread and for comparing the digital representation with one or more known digital representations characterised in that the processing system comprises a circular buffer for storing the digital representation of the thread and is further adapted to scan the digital representation to locate a predetermined code sequence; and to rotate a digital representation within the circular buffer to locate the predetermined code sequence in a predetermined position corresponding to the position of the predetermined code sequence in a stored version of the or each known digital representation prior to comparing the digital representation with the known digital representations.
  • the processing system is further adapted to compare the digital representation against reversed versions of the known digital representations, thereby determining the orientation of the coded magnetic thread in the case that the code is asymmetric.
  • the apparatus may further comprise a detector for measuring the lateral displacement of the thread for determining which face of a sheet document containing the thread is uppermost.
  • the processing system is further configured to scan the digital representation for features which indicate the likelihood of the digital representation being valid, prior to performing the comparison.
  • an apparatus is provided according to the second aspect of the present invention for carrying out a method according to the first aspect of the invention.
  • FIG. 1 A schematic representation of a coded magnetic thread reader suitable for reading the codes stored in a coded magnetic thread on a sheet document is shown in Figure 1.
  • the reader comprises a linear array 1 of twelve magnetic heads 2a to 21, each of which is connected to an individual signal processor 3a to 31.
  • the analogue signals generated by the magnetic heads are converted into a digital form by the signal processors 3a to 31, which interface with a microprocessor system 4.
  • Software executed by the microprocessor system 4 performs further processing on the digitised signal in order to align the code into a known format and compare it against a database of known codes.
  • the software also captures the peak positive and negative excursions of the analogue signal using the analogue to digital converters of the microprocessor system 4 and calculates appropriate thresholds from these. These thresholds are set on the signal processors 3a to 31 using digital to analogue converters of the microprocessor system 4.
  • FIG. 2 shows the linear array 1 of magnetic heads 2a-2l and two sheet documents 5,6 being conveyed by a document transport system (not shown) such that they will pass by the array 1 of magnetic heads 2a-2l.
  • Each sheet 5,6 has a magnetic coded thread 7a,7b.
  • a permanent magnet incorporated within the array 1, temporarily magnetises the magnetic material that constitutes the threads 7a,7b.
  • FIG. 8 The arrangement of one of the magnetic heads is shown in Figure 8. It comprises a core 60, which may be made from a ferrite, around the arms of which are wound two coils 61a,61b. Permanent magnet 62 provides a magnetic bias at the air gap of the core which causes the magnetic material to be temporarily magnetised.
  • Sheet 5 is being conveyed by the document transport system such that the thread 7a is presented perpendicularly to the linear array 1 of magnetic heads 2a-2l. It can be seen that the thread 7a will pass directly under magnetic head 2f and the code held by thread 7a can be reconstructed from the signal produced by magnetic head 2f alone.
  • FIG. 9 An alternative arrangement for the array 1 of magnetic heads is shown in Figure 9.
  • the array 1 comprises thirteen magnetic heads 100a to 100m.
  • these magnetic heads 100a to 100m are not configured in a linear fashion. Instead, they are configured on two parallel axes with magnetic heads 100a to 100g being on the first axis and magnetic heads 100h to 100m being on the second axis.
  • the signals developed by the magnetic heads 100a to 100m Before the signals developed by the magnetic heads 100a to 100m can be processed they must be time-shifted appropriately. Either the signals produced by the magnetic heads 100a to 100g lying on the first axis or the signals produced by the magnetic heads 100h to 100m lying on the second axis or both must be time-shifted so that they are in alignment. This can be done using analogue or digital processing techniques involving using the predetermined distance between the first and second axes and the velocity of a sheet document passing under the array 1 to determine the amount by which the signals produced by the magnetic heads lying on one axis must be time-shifted so that they are aligned with the signals produced by the magnetic heads lying on the other axis. The velocity of this sheet document can either be measured directly or the velocity of the document transport system can be determined.
  • the magnetic heads used in these examples are inductive heads but magnetoresistive heads could be used.
  • the signal processors 3a-31 and software incorporate features that enable detection of the thread irrespective of its lateral displacement and enable reconstruction of the code from the signal generated by several magnetic heads in the case of a skewed thread.
  • n refers to the relevant channel number and takes an integer value from 0 to 11.
  • the analogue signal HEAD [n] generated by the magnetic head forms the input to a paraphase amplifier 10a,10b.
  • the inverted and non-inverted outputs of the paraphase amplifier 10a, 10b are connected to the inputs of a pair of comparators 11a, 11b and to the inputs of a 2:1 multiplexer 17a, 17b.
  • Comparators 11a and 11b compare the output signals from the paraphase amplifier with separate variable thresholds. If the inverted output from the paraphase amplifier 10a, 10b exceeds the threshold input of comparator 11a then the output of comparator 11a is driven low which subsequently drives the output of AND gate 12 low and, since this output is connected to the clear input of D-type latch 14, the Q output of D-type latch 14 is also driven low. Similarly, if input CLR [n] from the microprocessor system 4 is driven low then the Q output of D-type latch 14 will be driven low in response.
  • the value of the thresholds of comparator 11a and 11b are determined by the microprocessor system 4 via output DAC [n]. This output is presented to a unity gain buffer 15 before being connected to the threshold input of comparator 11b.
  • the output of unity gain buffer 15 is also connected to potential divider 16 which reduces the threshold presented to comparator 11a by a factor of two.
  • Both comparators 11a and 11b incorporate a degree of hysteresis to improve noise immunity and to prevent false switching.
  • the 2:1 multiplexer 17a,17b connects either the inverted or non-inverted output of the paraphase amplifier 10a,10b to the integrating peak detector 19.
  • the logic state of signal MUX[n] determines which of these two outputs is connected to the integrating peak detector 19.
  • Inverter 18 inverts the logic state of signal MUX[n] so that either analogue switch 17a or analogue switch 17b is closed.
  • Integrating peak detector 19 detects and stores the positive peak of the signal applied to it. This is presented to the microprocessor system 4 as signal PEAK [n]. The integrating peak detector 19 can be reset by asserting signal RESET [n].
  • FIG. 4 A typical signal generated by one particular magnetic head similar to those described above, due to a 2mm magnetic element passing by, is shown in Figure 4. As the magnetic element approaches the head a negative peak 21 is generated. When the magnetic element is underneath the head, the direction of magnetic flux reverses and a positive peak 22 is generated. Finally, as the magnetic element moves away from the head, the flux reverses a second time and a second negative peak 23 is generated. This represents the signal HEAD[n] shown in Figure 4.
  • This signal is buffered by a unity gain inverter 15, the output of which determines a positive threshold.
  • This output is also potentially divided, for example by a factor of 2, using a potential divider 16 which sets a negative threshold.
  • the positive threshold may be set at four times the peak noise level, the negative threshold consequently being half this magnitude.
  • These thresholds may then be adapted and can be modified for each magnetic element scanned by the head. For example, the running average of the positive peak generated by a magnetic element could be calculated and used to determine a suitable positive threshold.
  • the value of the thresholds can be stored in a non-volatile memory so that they are not lost when the apparatus is switched off.
  • the signal MUX[n] is now driven high and the signal processors 3a to 31 await the arrival of a valid signal generated by a magnetic element.
  • a negative going excursion is induced in signal HEAD[n].
  • This negative going excursion is inverted at the inverted output of paraphase amplifier 10a,10b and integrating peak detector 19 stores the peak value of this excursion.
  • the direction of magnetic flux reverses and a positive going signal excursion is induced.
  • a clock pulse is supplied to the D-type latch 14 causing the Q output to be driven high. This generates an interrupt to the microprocessor system 4, the time of which is recorded.
  • the value of the negative peak is recorded, the integrating peak detector 19 is reset and the signal MUX[n] is driven low so that the positive peak can be detected by the integrating peak detector 19.
  • the direction of magnetic flux reverses a second time and a second negative going excursion is generated.
  • the thresholds are now adjusted so that the negative threshold has a value derived from the immediately preceding negative peak.
  • the signal MUX[n] is driven high so that the integrating peak detector 19 is monitoring for the presence of a negative going peak and the clear input of the D-type latch 14 is asserted so that the interrupt to the microprocessor system 4 is cleared.
  • the time of this event is recorded and hence the duration of the interrupt pulse can be determined.
  • the length of the magnetic element can be derived from this duration using a clock that is driven by the transport system.
  • the value of the positive peak is recorded, the integrating peak detector 19 is reset and the signal MUX[n] driven high so that the next negative peak can be acquired.
  • the value of the positive peak is used to determine when a skewed thread has passed from one magnetic head to an adjacent one as will be described subsequently.
  • the system can tolerate a wide variation in flux density of the magnetic material. Such variation may be caused by differing condition of the sheet documents bearing the threads, variance in the displacement between the magnetic heads and the threads or changing the speed of the document transport system.
  • FIG. 5 shows a typical signal produced when a 6mm length magnetic element passes by a magnetic head. The rate of change of the flux approaches zero when a long element such as this is directly beneath the head. Hence, the induced electromotive force also approaches zero. This can be seen as the dip 30 in Figure 5. It can be seen that since there are positive and negative thresholds, the interrupt commences at point 31 and ends at point 32 as required. However, if only a positive threshold were used, then two interrupts would be generated, the first commencing at point 31 and ending at point 33 and the second commencing at point 34 and ending at point 35.
  • the software in the microprocessor system 4 is responsible for providing the appropriate outputs to the signal processors 3a to 31 at the correct time and responding to their inputs such that the data read from the magnetic code can be reconstructed. To do this, the software is split into two major sections. These are six synchronously executed processes and three interrupt service routines.
  • FIG. 7 A software flow diagram is shown in Figure 7. The operation of the individual software processes and interrupt service routines will now be described with reference to Figure 7.
  • the microprocessor system 4 commences by executing process IDLE 50. This process is responsible for basic initialisation functions including checking whether any faults have been recorded by other software processes, reporting these faults if appropriate and checking whether any non-runtime communications have occurred. Execution now proceeds to process RUN-UP 51 on request, as long as there are no current faults.
  • Process RUN-UP 51 performs various other initialisation routines in order that background noise measurements can be performed to set the positive thresholds appropriately. Hence, all interrupts are disabled, the output MUX[n] is driven low so that the integrating peak detector 19 is recording positive peak values. Finally, the storage arrays for the head data are initialised by setting pointers to their beginning and if no faults have been logged, execution proceeds to process CALIBRATION 52.
  • This process is responsible for recording the peak background noise present on all heads. This is done by measuring the peak noise for 32 blocks of 1 millisecond each and averaging the peak detected for each of these blocks. The digital thresholds are now set with respect to the measured noise and if no faults have been logged, execution proceeds to process SET-CAPTURE 53.
  • Process SET-CAPTURE 53 commences by asserting signal MUX[n] so that the integrating peak detector 19 is recording negative peak values.
  • the code storage array is initialised and interrupts are enabled. If no faults have been logged then execution proceeds to process RUNNING 54.
  • the remaining two processes RUNNING 54 and CALCULATE-CODE 55 collate the data read from the signal processors 3a to 31, align it into a known format and compare it with a database of known codes. It is important to realise that the data from the signal processors 3a to 31 is actually captured by two interrupt service routines that are executed in response to interrupts generated by signal processors 3a to 31.
  • a particular advantage of using interrupts is that it is not necessary to scan the entire array 1 of magnetic heads 2a to 21 until a valid signal is detected. Hence, the microprocessor system 4 can perform other tasks when no valid signal is present.
  • the interrupt service routines will be described subsequently.
  • Process RUNNING 54 is responsible for constructing a bit sequence from the data supplied by the interrupt service routines and for reconstructing the code if the thread is skewed. This process also monitors the bit sequence for the presence of a marker section and when sufficient bits have been captured and no errors have been logged execution proceeds to process CALCULATE-CODE 55.
  • An example of a marker section of a possible code is shown in Figure 6, the marker section in this case being a reversal pattern 1010.
  • the software can reconstruct the code simply by logically ORing the data captured by all heads that were primary or secondary as the thread passed the array 1 of magnetic heads 2a to 21.
  • An advantage of this is that it is only necessary to store pertinent information; the signals generated by magnetic heads that were not primary or secondary can be ignored and discarded.
  • Process CALCULATE-CODE 55 begins by finding the start and end of the captured code. It works from the centre of the captured code outwards since this is less likely to be corrupted by other magnetic features that may be present and tears in the edge of the sheet.
  • the process searches for a repetition of this code which is used as a confidence check that the code is correct.
  • the code is then aligned to a known format and compared against a database of known codes to find the best match. If a match is found then a flag is set to indicate this fact to the relevant software process. Execution then returns to process SET-CAPTURE 53 so that the next code can be captured.
  • the code is aligned by storing it in a circular buffer and rotating it until the marker section is in a known position. This has the advantage that only one comparison is necessary against each database entry whereas a sliding correlation technique requires rotating a m-bit code through each of its m permutations and comparing each permutation against each database entry.
  • the aligned code is compared with the database entries by logically exclusively ORing it with each entry.
  • the number of bits set in the aligned code is divided by the result of this exclusive OR operation. The smallest value indicates the best match.
  • scoring techniques can be used to determine the likelihood of the captured code being in error. This can be done by searching for certain features of the code, for example:
  • interrupt service routines responsible for recording the code held by the magnetic thread.
  • INTERRUPT LEADING EDGE SERVICE ROUTINE 56 is executed. This routine records the value of the negative peak and this value is used to set the negative threshold for the subsequent negative peak. The event is also time stamped and converted into displacement of the transport system using a clock that is synchronous with the transport system drive mechanism. The peak detector 19 is then reset and signal MUX[n] negated so that the multiplexer 17a,17b presents positive signals to the integrating peak detector 19.
  • INTERRUPT TRAILING EDGE SERVICE ROUTINE 57 is executed. This routine records the positive peak value from the integrating peak detector 19. This value is used to track the thread if it is skewed and moves from one magnetic head to another. The event is time stamped in a similar manner to the leading edge so that the length of the magnetic element can be determined. The peak value stored in the integrating peak detector 19 is cleared and the multiplexer 17a,17b is set to look for negative peaks. The pointer to the storage array is advanced to the next bit.
  • a third interrupt service routine, ADC AUTOSCAN SERVICE ROUTINE 58 is responsible for performing regular conversions of the twelve signals PEAK[n] from the signal processors 3a-3l using the analogue to digital converters of the microprocessor system 4. These conversions are triggered automatically by a timer interrupt. This is done to reduce the processor overhead. The converted values are only permanently stored if required such as on detection of the leading or trailing edge of an interrupt.
  • the captured code may be successfully compared with a database entry, it may be possible to determine certain information about the sheet document. For example, if the sheet document is a bank note, it may be possible to determine its denomination. On the basis of this it would be possible to send the note to a desired destination for example to split a stack of notes into two denominations. Alternatively, it would be possible to stop the document transport if the thread's code is unreadable or say a rogue denomination note is discovered in a stack of notes of a single denomination.
  • the code is asymmetric, it is possible to detect the orientation of the sheet. If it is possible to detect the location of a feature of the sheet that is offset from its centre then it is possible to detect which face of the sheet is uppermost. For example, using an optical detector it is possible to determine the lateral position of the thread and this can be used to determine which face of the sheet is uppermost. Alternatively, the position of a known magnetic feature relative to the thread may be determined and this can be used to determine which face of the sheet is uppermost.
  • FIG. 10 illustrates a modified apparatus.
  • the heads 2a-2l are connected to an analogue to digital converter (ADC) 200 which is connected to a digital signal processor (DSP) 205.
  • ADC analogue to digital converter
  • DSP digital signal processor
  • the purpose of the DSP 205 is to process the digitised data and generate a series of digital waveforms representing the code stored within the magnetic feature. These waveforms are presented to the microprocessor 4 where pattern matching algorithms are applied to determine the authenticity and denomination of the note.
  • ADC analogue to digital converter
  • DSP digital signal processor
  • the ADC 200 samples the analogue signal every 0.25mm, generates a digital representation and transmits this to the DSP. Whilst the ADC 200 is busy converting the current sample, the DSP 205 is processing the previous sample obtained from an adjacent channel in a pipeline structure. This process repeats until all the note data has been acquired, thus processing is performed in real-time.
  • Sampling for a pair of channels is governed by a free running timer at a fixed period of 9.4 ⁇ s.
  • the system requires a measurement of the linear note speed. This is provided by a timing wheel consisting of a slotted-opto sensor (not shown) in a conventional manner. This provides a pulse corresponding to 4.42 mm linear travel. By measuring the number of timer pulses that have occurred within a timing wheel slot, the system can determine a sampling delay that is introduced to ensure the required sampling pitch.
  • Note sampling and processing is enabled under instruction of the microprocessor 4 and a track sensor (not shown).
  • the track sensor is a reflective optical sensor that provides an indication of the presence of a note under the detector.
  • the DSP 205 performs three main processing tasks
  • the algorithm used to generate a digital waveform for the microprocessor 4 comprises peak detection and a priori signal conditioning. Peak detection is used since the signals generated from the inductive magnetic heads are based on rate of change of magnetic material past the head. Therefore transitions occur at boundaries between magnetic and non-magnetic features. Example idealised waveforms for various sized magnetic features are shown in Figure 11.
  • peak detection could be used to determine the extent of the magnetic regions along a single plane.
  • the problem with using a peak detector is that like any rate of change detector, it is susceptible to signal noise. In practice, noise will be present on the input signals and therefore mechanisms are required to reduce the effect of these artifacts.
  • Two schemes may be used to provide a level of noise resistance; calibrated thresholds and larger peak detection window.
  • Calibration is required so that the system can generate a suitable threshold for each channel. These thresholds will be used to stop processing of low amplitude signals which although they may satisfy the peak detector, are due to system noise rather than valid magnetic material passing across the head.
  • the calibration scheme is as follows.
  • the microprocessor 4 will instruct the DSP 205 to go into calibration mode.
  • the DSP 205 takes 32 samples and generates an average absolute level. A threshold being a constant multiple of the average level is created and stored.
  • the DSP 205 examines the 32 samples to see if any exceed the calculated threshold. If so, calibration is reported as having failed otherwise calibration is a success and note processing can continue. The process is repeated for the remaining channels. If calibration has failed, the DSP 205 will report to the microprocessor 4 as being not ready and requires intervention.
  • the calibration process is performed on every bundle.
  • the second scheme to provide a level of noise resistance is in the choice of peak detection kernel applied to the data. Rather than a rate of change kernel that looks at the differences between direct neighbouring values (size 3), the approach taken in this design is to look at the next nearest neighbour (size 5).
  • a simple example that illustrates the benefit of a size 5 over a size 3 in terms of number of peaks detected is shown in Figure 13.
  • a signal whose amplitude varies in a similar manner to the example above e.g. noise will produce a large number of peaks with a size 3 kernel whereas a substantially reduced number will be produced by the size 5 kernel. Since the peak transitions due to magnetic/non-magnetic boundaries take place over more than three samples, the size 5 kernel is sufficiently small enough to track these transitions whilst providing a level of noise immunity.
  • the system applies the size 5 peak detector to data in real-time as it is acquired and adds valid peaks (i.e. a local minimum or maximum that is greater than the threshold band) to a list that contains information about peaks that have been found on a given channel.
  • the data that is stored is the position along the note parallel to the short edge where the peak was detected, the type of peak detected (i.e. a positive or negative peak) and the location in DSP 205 memory where the raw analogue data from the ADC 200 for that peak is stored.
  • the advantage of this is that the amount of data that has to subsequently be searched and processed is greatly reduced. This allows additional flexibility for more sophisticated algorithms since the volume of data has been reduced.
  • the DSP 205 has produced (for all 12 channels) a set of events that contain all peaks that satisfy the thresholding criterion.
  • the next process is to examine these peaks and determine which of those are valid and indicate true magnetic transition events and which are due to signal artifacts.
  • Each of these voltage peaks is individually checked against more stringent criteria. These criteria encapsulate the key characteristics of valid magnetic transitions, including checks on the absolute levels of the induced voltages and checks on the signature of the voltage peak. Any voltage peak that fails the criteria check is disregarded. Figure 14 illustrates this showing that each of the peaks that pass the initial assembly level routine check is classified as either valid or invalid.
  • This resultant subset of the initial voltage peaks is processed to further remove any erroneous signals. This is done in part by assessing the relative locations, magnitudes and shapes of each peak with those of any other peaks in close spatial proximity to it. This ensures that peaks that occur because of an increase in magnetic flux in the detector are matched with those peaks that correspond to a decrease in magnetic flux in the detector. Because of the complex note dynamics that occur as a note passes a detector, situations can arise in which there is ambiguity as to how the peaks should be joined together. For example, two voltage maxima may occur with no voltage minima between them.
  • One peak has been rejected because a falling edge peak should have been preceded by a corresponding rising edge peak within a given distance (where the distance corresponds to the length including a tolerance of the longest magnetic region expected).
  • the other peak has been rejected based on the peak properties because there are two rising edge peaks with only one falling edge peak.
  • This refined set of peaks is checked to make sure that a long magnetic region has not appeared to be made up of two shorter magnetic transitions. Again, this is done by evaluating the relative properties of a given group of peaks with those determined from empirical studies of the notes.
  • the data required by the microprocessor 4 is a digital bit stream stored in DSP memory for each channel. This stream is broken into chunks that can be stored in individual memory locations with 1 bit corresponding to a 0.25mm sample. Therefore, for 16 bit storage, each location would correspond to 4mm of note. As validated transition events are confirmed, the bit stream is constructed for each channel. Once an individual location has had all bits written to, the DSP moves on to the next location.
  • An example is shown in Figure 16.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Computer Security & Cryptography (AREA)
  • Investigating Or Analyzing Materials By The Use Of Magnetic Means (AREA)
  • Inspection Of Paper Currency And Valuable Securities (AREA)
  • Signal Processing For Digital Recording And Reproducing (AREA)

Claims (18)

  1. Procédé d'identification d'un fil de sécurité magnétique codé, comprenant la génération d'une représentation numérique du fil de sécurité et la comparaison de la représentation numérique avec une ou plusieurs représentations numériques connues, caractérisé en ce que le procédé comprend en outre le stockage de la représentation numérique du fil de sécurité dans un tampon circulaire ; le balayage de la représentation numérique afin de situer la séquence de code prédéterminée ; la rotation de la représentation numérique dans le tampon circulaire afin de situer la séquence de codage prédéterminée dans une position prédéterminée correspondant à la position de la séquence de code prédéterminé, dans une version stockée de la ou de chaque représentation numérique connue avant la réalisation de la phase de comparaison.
  2. Procédé selon la revendication 1, dans lequel la représentation numérique est binaire.
  3. Procédé selon la revendication 1 ou 2, dans lequel le code est asymétrique.
  4. Procédé selon la revendication 3, dans lequel la comparaison est réalisée par rapport aux versions inversées des représentations numériques connues, en déterminant ainsi l'orientation du fil de sécurité magnétique codé.
  5. Procédé selon la revendication 4, dans lequel le déplacement latéral du fil de sécurité est mesuré et utilisé pour déterminer quelle face d'un document en feuille contenant le fil de sécurité est située la plus au dessus.
  6. Procédé selon la revendication 4, dans lequel le déplacement relatif du fil de sécurité, à partir d'une caractéristique magnétique connue, est mesuré et utilisé pour déterminer quelle face d'un document en feuille contenant le fil de sécurité et la caractéristique magnétique est située la plus au dessus.
  7. Procédé selon l'une quelconque des revendications 2 à 6, dans lequel, avant la réalisation de la comparaison, la représentation numérique est scannée pour vérifier au moins une caractéristique qui indique la probabilité que la représentation numérique soit valable.
  8. Procédé selon la revendication 7, dans lequel la représentation numérique est scannée pour confirmer que les bits les moins significatifs et les plus significatifs sont établis.
  9. Procédé selon la revendication 7 ou la revendication 8, dans lequel la représentation numérique est scannée pour confirmer que le nombre de changements de bit est compris dans des limites prédéfinies.
  10. Procédé selon l'une quelconque des revendications 7 à 9, dans lequel la représentation numérique est scannée pour confirmer que le nombre de bits établis est dans des limites prédéfinies.
  11. Procédé selon l'une quelconque des revendications 7 à 10, dans lequel la représentation numérique est scannée pour confirmer que la séquence de code prédéterminée est présente et à l'emplacement correct.
  12. Procédé selon l'une quelconque des revendications 7 à 11, dans lequel la représentation numérique est scannée pour confirmer que le code est asymétrique.
  13. Procédé selon l'une quelconque des revendications 7 à 12, dans lequel des scores différents sont appliqués à différentes caractéristiques dépendant de leur importance relative.
  14. Appareil d'identification d'un fil de sécurité magnétique codé comprenant un détecteur de champ magnétique, un système de traitement pour traiter des signaux générés par le détecteur pour générer une représentation numérique du fil de sécurité et pour comparer la représentation numérique avec une ou plusieurs représentations numériques connues, caractérisé en ce que le système de traitement comprend un tampon circulaire pour stocker la représentation numérique du fil de sécurité et est en outre adapté pour scanner la représentation numérique afin de situer une séquence de codage prédéterminée ; et pour tourner la représentation numérique dans le tampon circulaire, afin de situer la séquence de codage prédéterminé dans une position prédéterminée, correspondant à la position de la séquence de codage prédéterminée dans une version stockée de la ou de chaque représentation numérique connue, avant la comparaison de la représentation numérique aux représentations numériques connues.
  15. Appareil selon la revendication 14, dans lequel le système de traitement est en outre adapté pour comparer la représentation numérique aux versions inversées des représentations numériques connues, en déterminant ainsi l'orientation du fil de sécurité magnétique codé.
  16. Appareil selon la revendication 14 ou la revendication 15, dans lequel l'appareil comprend en outre un détecteur pour mesurer le déplacement latéral du fil de sécurité, afin de déterminer quelle face d'un document en feuille contenant le fil de sécurité est située la plus au dessus.
  17. Appareil selon l'une quelconque des revendications 14 à 16, dans lequel le système de traitement est en outre configuré pour scanner la représentation numérique pour vérifier des caractéristiques qui indiquent la probabilité que la représentation numérique soit valable, avant d'effectuer la comparaison.
  18. Appareil selon l'une quelconque des revendications 14 à 17, pour réaliser un procédé selon l'une quelconque des revendications 1 à 13.
EP03015595A 2001-01-08 2002-01-07 Lecteur de fil de sécurité magnétique Expired - Lifetime EP1353301B1 (fr)

Applications Claiming Priority (5)

Application Number Priority Date Filing Date Title
GB0100451 2001-01-08
GBGB0100451.4A GB0100451D0 (en) 2001-01-08 2001-01-08 Method for reading coded magnetic threads
GB0114212 2001-06-11
GBGB0114212.4A GB0114212D0 (en) 2001-01-08 2001-06-11 Magnetic thread reader
EP02250079A EP1221679B1 (fr) 2001-01-08 2002-01-07 Lecteur de fil de sécurité magnétique

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
EP02250079A Division EP1221679B1 (fr) 2001-01-08 2002-01-07 Lecteur de fil de sécurité magnétique

Publications (3)

Publication Number Publication Date
EP1353301A2 EP1353301A2 (fr) 2003-10-15
EP1353301A3 EP1353301A3 (fr) 2003-10-22
EP1353301B1 true EP1353301B1 (fr) 2006-11-02

Family

ID=9906433

Family Applications (2)

Application Number Title Priority Date Filing Date
EP03015595A Expired - Lifetime EP1353301B1 (fr) 2001-01-08 2002-01-07 Lecteur de fil de sécurité magnétique
EP03015597A Expired - Lifetime EP1353302B1 (fr) 2001-01-08 2002-01-07 Lecteur de fil de sécurité magnétique

Family Applications After (1)

Application Number Title Priority Date Filing Date
EP03015597A Expired - Lifetime EP1353302B1 (fr) 2001-01-08 2002-01-07 Lecteur de fil de sécurité magnétique

Country Status (3)

Country Link
EP (2) EP1353301B1 (fr)
GB (2) GB0100451D0 (fr)
PT (1) PT1353301E (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10922913B2 (en) 2016-12-27 2021-02-16 Giesecke+Devrient Currency Technology Gmbh Method and apparatus for detecting a security thread in a value document

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
ITTO20040753A1 (it) * 2004-10-29 2005-01-29 Photovox S R L Testina magnetica modulare per la lettura di bande o fili magnetizzati
EP2046513B1 (fr) 2006-06-28 2012-10-31 Talaris Holdings Limited Appareil de manipulation de documents
DE102008048043A1 (de) * 2008-09-19 2010-03-25 Giesecke & Devrient Gmbh Kalibrieren eines Sensors zur Wertdokumentbearbeitung
JP6198932B2 (ja) * 2014-03-27 2017-09-20 三菱電機株式会社 情報読取装置及び情報読取方法
CN105160756A (zh) * 2015-08-18 2015-12-16 深圳怡化电脑股份有限公司 一种纸币面向识别方法与装置
GB2615816A (en) * 2022-02-21 2023-08-23 The Governor & Company Of The Bank Of England Document recognition device

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2098768B (en) * 1981-04-13 1984-07-04 Bank Of England The Governor & Card readers
GB8921435D0 (en) * 1989-09-22 1989-11-08 Bank Of England Sensor system for document sorting machines
US5378885A (en) * 1991-10-29 1995-01-03 Mars Incorporated Unshielded magnetoresistive head with multiple pairs of sensing elements
AU689968B2 (en) * 1994-11-18 1998-04-09 Governor And Company Of The Bank Of England, The Security thread detector assembly
DE19921653A1 (de) * 1999-03-24 2000-09-28 Kraus Sicherheitstechnik Geldscheinsicherheitscodefaden

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10922913B2 (en) 2016-12-27 2021-02-16 Giesecke+Devrient Currency Technology Gmbh Method and apparatus for detecting a security thread in a value document

Also Published As

Publication number Publication date
GB0100451D0 (en) 2001-02-21
PT1353301E (pt) 2007-01-31
EP1353301A3 (fr) 2003-10-22
EP1353302A3 (fr) 2003-10-22
EP1353302B1 (fr) 2005-07-20
EP1353301A2 (fr) 2003-10-15
GB0114212D0 (en) 2001-08-01
EP1353302A2 (fr) 2003-10-15

Similar Documents

Publication Publication Date Title
US7434732B2 (en) Magnetic thread reader
JP2739157B2 (ja) 調整可能なしきい値を有する磁気コード読取装置
EP1353301B1 (fr) Lecteur de fil de sécurité magnétique
GB2098768A (en) Card readers
US4232216A (en) Method and device for reading coded information in the form of intervals of predetermined sizes
AU689968B2 (en) Security thread detector assembly
US4845348A (en) Method and apparatus for reading bar code magnetic ink characters
US6928183B2 (en) System for and method of reading MICR using high and low gain signals
US3879707A (en) Character recognition system for bar coded characters
JPS5848956B2 (ja) 紙葉類鑑別装置
US5257319A (en) Character recognition apparatus
JPH08180239A (ja) 書類識別装置および方法
JP3599941B2 (ja) データ記録媒体及びその処理装置
JP3597530B2 (ja) 記録データの処理装置及び処理方法
JPH03214390A (ja) 磁気バーコード偽造検出システム

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

PUAL Search report despatched

Free format text: ORIGINAL CODE: 0009013

AC Divisional application: reference to earlier application

Ref document number: 1221679

Country of ref document: EP

Kind code of ref document: P

AK Designated contracting states

Kind code of ref document: A2

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LI LU MC NL PT SE TR

AX Request for extension of the european patent

Extension state: AL LT LV MK RO SI

AK Designated contracting states

Kind code of ref document: A3

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LI LU MC NL PT SE TR

AX Request for extension of the european patent

Extension state: AL LT LV MK RO SI

RIC1 Information provided on ipc code assigned before grant

Ipc: 7G 07D 7/00 B

Ipc: 7G 07D 7/04 A

17P Request for examination filed

Effective date: 20040422

AKX Designation fees paid

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LI LU MC NL PT SE TR

AXX Extension fees paid

Extension state: AL

Payment date: 20030731

Extension state: MK

Payment date: 20030731

Extension state: RO

Payment date: 20030731

Extension state: SI

Payment date: 20030731

Extension state: LV

Payment date: 20030731

Extension state: LT

Payment date: 20030731

17Q First examination report despatched

Effective date: 20041110

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

AC Divisional application: reference to earlier application

Ref document number: 1221679

Country of ref document: EP

Kind code of ref document: P

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LI LU MC NL PT SE TR

AX Request for extension of the european patent

Extension state: AL LT LV MK RO SI

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: IT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT;WARNING: LAPSES OF ITALIAN PATENTS WITH EFFECTIVE DATE BEFORE 2007 MAY HAVE OCCURRED AT ANY TIME BEFORE 2007. THE CORRECT EFFECTIVE DATE MAY BE DIFFERENT FROM THE ONE RECORDED.

Effective date: 20061102

REG Reference to a national code

Ref country code: GB

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: CH

Ref legal event code: EP

REF Corresponds to:

Ref document number: 60215768

Country of ref document: DE

Date of ref document: 20061214

Kind code of ref document: P

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MC

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20070131

REG Reference to a national code

Ref country code: PT

Ref legal event code: SC4A

Free format text: AVAILABILITY OF NATIONAL TRANSLATION

Effective date: 20061211

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20070202

Ref country code: SE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20070202

REG Reference to a national code

Ref country code: CH

Ref legal event code: NV

Representative=s name: BOVARD AG PATENTANWAELTE

Ref country code: GR

Ref legal event code: EP

Ref document number: 20070400192

Country of ref document: GR

LTIE Lt: invalidation of european patent or patent extension

Effective date: 20061102

ET Fr: translation filed
REG Reference to a national code

Ref country code: ES

Ref legal event code: FG2A

Ref document number: 2274147

Country of ref document: ES

Kind code of ref document: T3

PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

26N No opposition filed

Effective date: 20070803

REG Reference to a national code

Ref country code: GB

Ref legal event code: 732E

Free format text: REGISTERED BETWEEN 20090122 AND 20090128

REG Reference to a national code

Ref country code: PT

Ref legal event code: PC4A

Owner name: TALARIS HOLDINGS LIMITED, GB

Effective date: 20090211

NLS Nl: assignments of ep-patents

Owner name: TALARIS HOLDINGS LIMITED

Effective date: 20090205

REG Reference to a national code

Ref country code: CH

Ref legal event code: NV

Representative=s name: KELLER & PARTNER PATENTANWAELTE AG

Ref country code: CH

Ref legal event code: PUE

Owner name: TALARIS HOLDINGS LIMITED

Free format text: DE LA RUE INTERNATIONAL LIMITED#DE LA RUE HOUSE, JAYS CLOSE, VIABLES#BASINGSTOKE, HANTS RG22 4BS (GB) -TRANSFER TO- TALARIS HOLDINGS LIMITED#TALARIS HOUSE CROCKFORD LANE CHINEHAM BUSINESS PARK#BASINGSTOKE, HAMPSHIRE RG24 8QZ (GB)

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: CY

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20061102

Ref country code: LU

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20070107

REG Reference to a national code

Ref country code: FR

Ref legal event code: TP

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: TR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20061102

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: CH

Payment date: 20100113

Year of fee payment: 9

Ref country code: ES

Payment date: 20100128

Year of fee payment: 9

Ref country code: IE

Payment date: 20100118

Year of fee payment: 9

Ref country code: PT

Payment date: 20100106

Year of fee payment: 9

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: FI

Payment date: 20100114

Year of fee payment: 9

Ref country code: FR

Payment date: 20100208

Year of fee payment: 9

Ref country code: IT

Payment date: 20100116

Year of fee payment: 9

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: AT

Payment date: 20100113

Year of fee payment: 9

Ref country code: BE

Payment date: 20100125

Year of fee payment: 9

Ref country code: DE

Payment date: 20091231

Year of fee payment: 9

Ref country code: GB

Payment date: 20100106

Year of fee payment: 9

Ref country code: GR

Payment date: 20091215

Year of fee payment: 9

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: NL

Payment date: 20100101

Year of fee payment: 9

REG Reference to a national code

Ref country code: PT

Ref legal event code: MM4A

Free format text: LAPSE DUE TO NON-PAYMENT OF FEES

Effective date: 20110707

BERE Be: lapsed

Owner name: TALARIS HOLDING LTD

Effective date: 20110131

REG Reference to a national code

Ref country code: NL

Ref legal event code: V1

Effective date: 20110801

REG Reference to a national code

Ref country code: CH

Ref legal event code: PL

GBPC Gb: european patent ceased through non-payment of renewal fee

Effective date: 20110107

REG Reference to a national code

Ref country code: FR

Ref legal event code: ST

Effective date: 20110930

REG Reference to a national code

Ref country code: IE

Ref legal event code: MM4A

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: PT

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110707

Ref country code: LI

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110131

Ref country code: FR

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110131

Ref country code: CH

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110131

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: GR

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110802

Ref country code: FI

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110107

Ref country code: BE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110131

Ref country code: GB

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110107

Ref country code: AT

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110107

REG Reference to a national code

Ref country code: DE

Ref legal event code: R119

Ref document number: 60215768

Country of ref document: DE

Effective date: 20110802

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: NL

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110801

Ref country code: IT

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110107

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: IE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110107

REG Reference to a national code

Ref country code: ES

Ref legal event code: FD2A

Effective date: 20120220

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: ES

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110108

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20110802