RU2158443C1 - Method for detection of authenticity and value of bank notes and bank note sorting mechanism - Google Patents

Abstract

FIELD: bank equipment. SUBSTANCE: method involves scanning bank note by means of special-purpose scanning devices, production of its image in three spectra: in visible light, infrared and ultraviolet light, transmitting image to computer and processing it using image recognition software. Authenticity of bank note rests on comparison of infrared and ultraviolet images against reference one. Device provides analysis of complete image of bank notes. EFFECT: increased functional capabilities. 2 cl, 6 dwg

Description

 The invention relates to methods and devices for counting, sorting and authenticating banknotes, and is intended for use in banks.

 A known machine for counting and sorting banknotes De La Rue 3700 company De La Rue [1], in which to determine the authenticity and denomination of banknotes using a method based on checking the presence of marks in certain places of the banknote using infrared and ultraviolet sensors.

 In order to protect against fakes, an image is applied to the banknote with the so-called metamer paints, this image is invisible to the ordinary eye and appears only in the infrared spectrum. Knowing a specific infrared image of a banknote, a sensor can be developed that checks for the presence or absence of metamer paint at several specific points.

 This sensor is a pair of infrared emitter and photodetector, working on reflection from paper.

 In the absence of metamer paint, the photodetector is illuminated by reflected light; in the presence of paint, the light flux on the photodetector decreases.

 When the banknote is illuminated with UV radiation, induced visible radiation occurs. Knowing the specific induced image of the banknote, a sensor can be developed that checks for the presence or absence of the induced image at several specific points.

 This sensor is a pair of UV emitter and photodetector working on reflection from paper.

 In the presence of an induced image, the photodetector is illuminated by reflected light; in the absence of an induced image, the light flux on the photodetector decreases.

 The disadvantage of this method is not checking the entire image, but only checking the presence or absence of the image at several points.

 The dilapidation of a banknote is checked by a microphone sensor according to the level of rustling of the paper, while significant signs of decay, such as dirty spots and inscriptions on the banknote, may not be noticed.

 Advantage is checked by a special sensor, which is a light source and many optical fibers that collect reflected illumination over several banknote strips, these optical fibers are collected at the output of the beam and transmit the integrated illumination to the photodetector, when the banknotes pass under the sensor on the photodetector, they receive various integral illumination curves depending on denominations of banknotes. This method does not guarantee absolute recognition of dignity with all other sensors disabled.

 The disadvantage of this method is the need to replace and rebuild the sensors in the transition to working with banknotes of a different kind, which takes time and effort to redesign the sensor system, as well as cash costs.

 The known machine has a feeder for pulling out a bundle of banknotes, a transport system for feeding banknotes to the sensors, diverters for sending banknotes to one of the receiving bins, depending on the decision made by the programmable controller based on data from the set of sensors, three receiving bins for folding the counted and sorted banknotes: one for suitable, the second for decrepit, the third for banknotes of dubious authenticity and value. A disadvantage of the known machine is the need to rebuild and replace a set of sensors when switching to another type of banknote, for example, banknotes of another country. Redesigning the sensor system to a new kind of banknote takes time, a certain amount of effort, as well as money.

 The present invention is to develop a method for determining the authenticity, dignity and dilapidation of banknotes, based on scanning banknotes using scanning devices to obtain complete images of the banknote in three spectra to achieve greater reliability, and creating a compact machine for counting and sorting banknotes that does not require changes in the hardware parts when changing the type of banknotes.

 This task is achieved by abandoning the system of sensors that read individual labels in different spectra and replacing the sensors with scanning devices that allow obtaining complete banknote images in three spectra, and then comparing the received banknote images with reference ones using a computer pattern recognition program.

 The essence of the method lies in the fact that the banknote is scanned by scanning devices to obtain images in three spectra, the received images are transmitted from the scanning devices to a computer, in which the received banknote images are compared with the reference ones and a decision is made on the authenticity, dignity and decay of the banknote in accordance with the computer pattern recognition program. A computer pattern recognition program is a software tool that allows you to train a computer to recognize the images presented to it and give an answer to the question to what extent the entered image matches the reference image. The program can determine the complete correspondence of images, partial correspondence of images, complete incompatibility of images, correspondence and inconsistency of individual fragments of the image. The standard tool for solving this kind of problem is the STARC program [2].

 To determine the value and decay of a banknote, a computer program compares the image of a banknote in the visible spectrum of a banknote with a reference. With the complete discrepancy of the images, the program makes a decision that the banknote is of another denomination, with a partial discrepancy (spots, inscriptions, tears on the banknote, etc.), the program makes a decision about the decay of the banknote. To determine the authenticity of the banknote, a computer program compares the infrared and ultraviolet images of the banknote with the reference. In case of complete discrepancy between the images, a decision is made on doubts about the authenticity of the banknote. In case of partial mismatch of images, the presence or absence of a pattern in special protection zones on a banknote is checked. If the protection zones do not comply, a decision is made on the doubt of the authenticity of the banknote.

 The proposed method is implemented in a BARS banknote recounting and sorting machine, including a feeder for pulling out a bundle of banknotes, a conveyor system for feeding banknotes, divertors for sending banknotes to one of the receiving bins, depending on the decision on authenticity, dignity and decay, three receiving bins for folding sorted banknotes: one for suitable, the second for decrepit, the third for banknotes of doubtful authenticity and dignity, in which scanning devices are installed to receive images of banknotes in three spectra, a computer with a computer program for pattern recognition to make a decision on the authenticity, dignity and decay of banknotes and transfer the decision to divertors for sending banknotes to one of the receiving bins.

 Figure 1 presents the structural diagram of technical means for implementing the method. 1 - scanning device for inputting banknote images in the visible and infrared spectra, 2 - scanning device for inputting banknote images in the ultraviolet spectrum 3 - computer for image processing.

 Figure 2 presents a block diagram of an algorithm for determining the value and decay of banknotes.

 Figure 3 shows a block diagram of an algorithm for determining the authenticity of banknotes.

 In FIG. 4 shows a structural diagram of a BARS machine, where. 4 - feeder, 5 - transport system, 1, 2 - scanning devices, 7 - divertors for highlighting the ultraviolet spectrum, 8 - receiving bunkers, 3 - computer.

 In FIG. 5 shows a scanning device for acquiring images in the visible and infrared spectra, where 9 is a scanned banknote, 10 is a visible spectrum illuminator, 11 is an infrared spectrum illuminator, 12 is a lens, 13 is a translucent mirror, 14 is a light filter to highlight the visible spectrum, 15 is a light filter for highlighting the infrared spectrum, 16 - CCD line, 17 - an electronic device for capturing images from CCD lines. Thin lines show the direction of light fluxes.

 In FIG. 6 shows a scanning device for obtaining banknote images in the infrared and ultraviolet spectra, where 9 is a scannable banknote, 18 is an ultraviolet illuminator, 19 is an infrared illuminator, 20 is a lens, 21 is a translucent mirror, 22 is an optical filter for highlighting the infrared spectrum, 23 - a light filter for emitting visible light, 24 - CCD - line, 25 - an electronic device for capturing images from CCD arrays. Thin lines show the direction of light fluxes.

 The machine operates as follows. The bundle of banknotes is placed in the feeder 4, from which the banknotes arrive one at a time in the transport system 5, the transport system feeds the banknote to the scanning devices 1, 2, the scanning devices transfer images of the banknote in the visible, infrared and ultraviolet spectra to the computer 3, the computer using the pattern recognition program compares them with the reference, makes a decision on the authenticity, dignity and dilapidation of banknotes and transmits a signal to divertors 7 to send the banknote to one of the bins 8 depending on the received about the decision.

 Scanning devices (scanners), commercially available and used to obtain images of paper documents and enter these images into a computer, do not allow you to get an image of a paper document simultaneously in two spectra (for example, visible and infrared, infrared and ultraviolet), which makes them unsuitable to determine the authenticity of banknotes [3,4,5,6]. They contain a illuminator, a lens, a CCD line, an electronic device for capturing images from a CCD line and transferring it to a computer. The paper document is illuminated and the image of this document with the lens is focused on the CCD line, where it is converted into electronic form using an electronic device for capturing images from CCD arrays and in this electronic form is transmitted to the computer.

 The proposed machine uses a device for acquiring banknote images in the visible and infrared spectra shown in FIG. 5.

 The specified technical result is achieved by installing after the lens 12 a translucent mirror 13, with which the banknote image is transmitted to two CCD arrays 16, while a filter 14 is installed in front of one CCD array to highlight the visible spectrum, and in front of the other, a filter 15 for infrared spectrum. In addition to a conventional illuminator, an infrared illuminator 11 is mounted on the scanning device.

 It is known that on genuine banknotes in special areas of the picture, the infrared image differs significantly from the visible. By viewing these areas on a computer screen, the operator is able to determine the authenticity of the banknote. When you install a pattern recognition program on your computer, the authenticity of a banknote can be determined automatically.

 A scanning device for obtaining banknote images in the visible and infrared spectra works as follows.

 The luminous flux is focused by the lens 12, then divided by a translucent mirror 13 into two streams and transmitted to the CCD arrays 16. The filters 14 and 15 standing in front of each ruler form a visible CCD array and an infrared image on the other CCD array. An electronic device 17 captures images from CCD arrays and transfers them to a computer.

 The proposed machine also uses a device for obtaining images of banknotes in the infrared and ultraviolet spectra. This device is presented in Fig.6.

 The specified technical result is achieved by installing after the lens a translucent mirror 21, with which the image is transmitted to two CCD arrays 24, with a light filter in front of one ruler to highlight the ultraviolet spectrum 22, and in front of another ruler a filter to highlight the infrared spectrum 23. Instead of a conventional illuminator infrared 19 and ultraviolet 18 illuminators are installed on the scanning device.

 When scanning a banknote, two images are taken - infrared and ultraviolet. It is known that on genuine banknotes in special areas of the picture, infrared and ultraviolet images differ significantly from the visible. By viewing these areas on a computer monitor, the operator can determine the authenticity of the banknote. If the computer has a special pattern recognition program, the authenticity of the banknote can be determined automatically.

 Thus, the proposed technical solution allows to increase the reliability of determining the authenticity, since it is not individual points that are analyzed, but the whole infrared and ultraviolet image, to guarantee the reliability of the denomination, since the image is analyzed as a whole, to avoid the need to redo the sensor system when transferring the machine to a new type of banknote, which requires time and money, since with the claimed method, to switch to a new currency, it is enough to train the software both sintering work with new images.

Literature
1. De La Rue Systems. Flyer.

 2. Certificate of official registration of the computer program "Universal system for pattern recognition STARC" (N 960128 from 04/18/96), copyright holder 000 "Firma Data Center".

 3. ProfScan 5 000. Brochure.

 4. PARAGON 800 SP. Flyer.

 5. ScanJet 5p. Flyer.

 6. FILEMASTER. Flyer.

Claims (2)

 1. The method of determining the authenticity and denomination of banknotes, which consists in the fact that the banknote is illuminated with visible, infrared or ultraviolet light, and a decision is made on the authenticity and denomination of the banknote, characterized in that the banknote is scanned with the help of scanning devices; spectra - visible, infrared and ultraviolet, compare them with the reference ones and at the same time the received images are transferred to a computer on which the pattern recognition program is installed in advance, after by means of which, to determine the authenticity of the banknote, the indicated full images of the banknote in the infrared and ultraviolet spectra are compared with the reference ones, if the images do not match, the banknote is questioned if the banknote is partially mismatched, the presence or absence of the pattern in the protection zones on the banknote is checked, if the protection zones do not match decide on the doubt of the authenticity of the banknote, to determine the value and decay, the image of the banknote obtained in the visible spectrum is compared According to the standard, in case of complete mismatch of images, a decision is made that a banknote of a different denomination, with a partial mismatch, a decision is made about the dilapidation of a banknote.  2. Banknote sorting machine, including a feeder for pulling out a bundle of banknotes and a transport system for feeding banknotes, diverters for sending banknotes to one of the receiving bins, depending on the decision on authenticity, dignity, three receiving bins for sorted banknotes - respectively suitable, shabby and for banknotes on which a decision has been made to doubt the authenticity, characterized in that the said transport system is designed to supply banknotes to scanning devices, one of which is intended about to obtain images of banknotes in the visible and infrared spectra and contains illuminators of the visible and infrared spectra, a translucent mirror, filters for separating the visible and infrared spectra and a device for capturing images and transferring them to a computer, another scanning device is designed to receive images of the banknote in infrared and ultraviolet spectra and contains illuminators of the ultraviolet and infrared spectra, a translucent mirror, filters for the separation of ultraviolet on and infrared spectra and apparatus for removing the image and transfer it to the computer for recognition with a view to deciding on the authenticity, dignity and dilapidation and transmission solutions for banknote diverters direction in one of the receiving bins.

Images