[go: up one dir, main page]

ES2549461B1 - METHOD AND DEVICE FOR THE CHARACTERIZATION OF THE STATE OF USE OF BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION - Google Patents

METHOD AND DEVICE FOR THE CHARACTERIZATION OF THE STATE OF USE OF BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION Download PDF

Info

Publication number
ES2549461B1
ES2549461B1 ES201430239A ES201430239A ES2549461B1 ES 2549461 B1 ES2549461 B1 ES 2549461B1 ES 201430239 A ES201430239 A ES 201430239A ES 201430239 A ES201430239 A ES 201430239A ES 2549461 B1 ES2549461 B1 ES 2549461B1
Authority
ES
Spain
Prior art keywords
classification
circulation
image
interest
characterization
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 - Fee Related
Application number
ES201430239A
Other languages
Spanish (es)
Other versions
ES2549461R1 (en
ES2549461A2 (en
Inventor
Eduardo KROPNICK CARVAJAL
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.)
Banco De Espana
Original Assignee
BANCO DE ESPANA
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 BANCO DE ESPANA filed Critical BANCO DE ESPANA
Priority to ES201430239A priority Critical patent/ES2549461B1/en
Priority to EP15155865.7A priority patent/EP2911123B1/en
Priority to ES15155865.7T priority patent/ES2591228T3/en
Publication of ES2549461A2 publication Critical patent/ES2549461A2/en
Publication of ES2549461R1 publication Critical patent/ES2549461R1/en
Application granted granted Critical
Publication of ES2549461B1 publication Critical patent/ES2549461B1/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G07D7/181Testing mechanical properties or condition, e.g. wear or tear
    • G07D7/187Detecting defacement or contamination, e.g. dirt
    • GPHYSICS
    • G07CHECKING-DEVICES
    • G07DHANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
    • G07D7/00Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
    • G07D7/06Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency using wave or particle radiation
    • G07D7/12Visible light, infrared or ultraviolet radiation
    • GPHYSICS
    • G07CHECKING-DEVICES
    • G07DHANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
    • G07D7/00Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
    • G07D7/06Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency using wave or particle radiation
    • G07D7/12Visible light, infrared or ultraviolet radiation
    • G07D7/1205Testing spectral properties

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Toxicology (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Inspection Of Paper Currency And Valuable Securities (AREA)
  • Image Analysis (AREA)

Abstract

Método y dispositivo para la caracterización del estado de uso de los billetes de banco y su clasificación en aptos y no aptos para la circulación, que evita errores en dicha clasificación, impidiendo que billetes no aptos sean puestos de nuevo en circulación, y que billetes aptos sean erróneamente destruidos de forma prematura. El método comprende capturar (1) al menos una imagen de al menos una de las caras de un billete a analizar, y analizar (4) en el dominio espacial la imagen captada. Además se analiza (5) en el dominio de la frecuencia la imagen captada y se combinan (13) ambos análisis (4, 5), mediante un algoritmo matemático, para obtener un único valor denominado FV. Finalmente, FV se compara (15) con un valor FV umbral fijado (14) y en función de que dicho FV calculado sea menor o mayor que el FV umbral, el billete se clasifica como apto o no apto para la circulación.Method and device for the characterization of the state of use of banknotes and their classification in apt and unfit for circulation, which avoids errors in said classification, preventing unsuitable bills from being put back into circulation, and that apt bills are mistakenly destroyed prematurely. The method comprises capturing (1) at least one image of at least one of the faces of a bill to be analyzed, and analyzing (4) in the spatial domain the captured image. In addition, the captured image is analyzed (5) in the frequency domain and both analyzes (4) are combined (13), using a mathematical algorithm, to obtain a single value called PV. Finally, PV is compared (15) with a fixed threshold PV value (14) and depending on whether said calculated PV is less than or greater than the threshold PV, the bill is classified as suitable or not suitable for circulation.

Description

DESCRIPCIONDESCRIPTION

Metodo y dispositivo para la caracterizacion del estado de uso de los billetes de banco, y su clasificacion en aptos y no aptos para la circulationMethod and device for the characterization of the state of use of banknotes, and their classification in apt and unfit for circulation

55

OBJETO DE LA INVENCIONOBJECT OF THE INVENTION

La presente invention, tal y como se expresa en el enunciado de esta memoria descriptiva, consiste en un metodo y dispositivo para la caracterizacion del estado de uso 10 de los billetes de banco y su clasificacion en aptos y no aptos para la circulacion, que tiene por objeto evitar errores en dicha clasificacion, impidiendo que billetes no aptos sean erroneamente clasificados como aptos y puestos de nuevo en circulacion debiendo haber sido destruidos, y evitando que billetes aptos sean erroneamente clasificados como no aptos y en consecuencia sean destruidos de forma prematura.The present invention, as expressed in the statement of this specification, consists of a method and device for characterizing the state of use 10 of banknotes and their classification as suitable and not suitable for circulation, which has in order to avoid errors in said classification, preventing unfit bills from being misclassified as fit and put back into circulation must have been destroyed, and preventing apt bills from being misclassified as unfit and consequently destroyed prematurely.

15fifteen

ANTECEDENTES DE LA INVENCIONBACKGROUND OF THE INVENTION

Una de las labores que llevan a cabo los Bancos Centrales Nacionales en relation con el efectivo, es la clasificacion por estado de uso de los billetes que retornan de la 20 circulacion, separando los aptos para seguir en circulacion de aquellos que no lo son.One of the tasks carried out by the National Central Banks in relation to cash, is the classification by state of use of the notes that return from the circulation, separating the apt ones to continue in circulation from those that are not.

Normalmente, los billetes aptos se preparan para ser puestos de nuevo en circulacion y los billetes no aptos se destruyen.Normally, suitable bills are prepared to be put back into circulation and unfit bills are destroyed.

Esta clasificacion por estado de uso se lleva a cabo mediante la detection y evaluation de diversas caracterlsticas en los billetes, como por ejemplo el grado de ensuciamiento 25 de los billetes, la existencia de agujeros y determination de su tamano, la existencia de esquinas dobladas y/o que faltan y determinacion de su tamano, la existencia de rasgaduras y determinacion de su tamano, la existencia de cintas adhesivas y determinacion de su tamano, la existencia de grafitis y determinacion de su tamano, la existencia de manchas y determinacion de su tamano, etc.This classification by state of use is carried out by detecting and evaluating various characteristics on the bills, such as the degree of fouling of the bills, the existence of holes and determination of their size, the existence of bent corners and / or missing and determination of its size, the existence of tears and determination of its size, the existence of adhesive tapes and determination of its size, the existence of graffiti and determination of its size, the existence of spots and determination of its size , etc.

30 De todas ellas, la que presenta una mayor dificultad es la medicion del grado de ensuciamiento de los billetes, siendo ademas la que origina que un mayor numero de billetes sean clasificados como no aptos, y por tanto, destruidos.30 Of all of them, the one that presents the greatest difficulty is the measurement of the degree of fouling of the bills, being also the one that causes a greater number of bills to be classified as unfit, and therefore, destroyed.

Los sistemas y metodos existentes en la actualidad se basan exclusivamente en elThe systems and methods currently in existence are based exclusively on the

1010

15fifteen

20twenty

2525

analisis de las imagenes de los billetes en el dominio espacial, es decir en descomponer las imagenes en los tres canales RGB (Red (rojo), Green (verde), Blue (azul)) y en medir sobre ellos o sobre ciertas combinaciones de estos canales, la media y desviacion estandar de los pixeles encuadrados en determinadas regiones de interes. En este sentido puede citarse el documento de patente WO 2012165959 referente a "Method and device for classifying security documents such as banknotes”.analysis of the images of the bills in the spatial domain, that is to say in decomposing the images in the three RGB channels (Red (red), Green (green), Blue (blue)) and in measuring on them or on certain combinations of these channels, the mean and standard deviation of the pixels framed in certain regions of interest. In this regard, the patent document WO 2012165959 referring to "Method and device for classifying security documents such as banknotes" can be cited.

El anterior documento, al igual que el resto de los metodos y dispositivos existentes en la actualidad, presentan una serie de problemas para la caracterizacion de la suciedad de los billetes y su clasificacion en aptos y no aptos. Estos problemas basicamente son de tres tipos:The previous document, like the rest of the methods and devices currently in existence, present a series of problems for the characterization of the dirt of the bills and their classification in apt and unfit. These problems are basically of three types:

1. Baja repetividad y reproducibilidad: El mismo billete, procesado en el mismo sistema varias veces o en sistemas diferentes del mismo tipo, produce resultados diferentes.1. Low repeatability and reproducibility: The same ticket, processed in the same system several times or in different systems of the same type, produces different results.

2. Tasas de errores cruzados ("crossovers”): Existe la tasa "False Fit Rate FFR” (billetes sucios erroneamente clasificados como limpios) y la tasa "False Unfit Rate FUR” (billetes limpios erroneamente clasificados como sucios). Con los sistemas actuales, para conseguir FFR por debajo del 10% se obtienen FUR por encima del 15 y del 20%. Mientras que la FFR "contamina” los billetes en circulacion con billetes sucios, la FUR representa directamente una perdida economica para los Bancos Centrales, al tratarse de billetes aptos para seguir en circulacion que se estan destruyendo de una forma prematura.2. Cross-error rates ("crossovers"): There is the "False Fit Rate FFR" (dirty tickets wrongly classified as clean) and the "False Unfit Rate FUR" (clean tickets wrongly classified as dirty). to obtain FFR below 10%, FURs are obtained above 15 and 20%, while the FFR "pollutes" the banknotes in circulation with dirty bills, the FUR directly represents an economic loss for the Central Banks, at these are banknotes that are suitable for circulation that are being destroyed prematurely.

3. Baja correlation con la perception humana de la suciedad en los billetes: En muchas ocasiones, los billetes clasificados como sucios por los sistemas actuales no son percibidos como tales por los seres humanos y viceversa, billetes clasificados como limpios presentan para el observador humano un grado de suciedad inaceptable.3. Low correlation with the human perception of the dirt on the banknotes: In many cases, banknotes classified as dirty by current systems are not perceived as such by human beings and vice versa, banknotes classified as clean present for the human observer a degree of dirt unacceptable.

Si se tiene en cuenta que los Bancos Centrales Nacionales de todo el mundo clasifican por estado de uso miles de millones de billetes cada ano, se puede llegar a tener una idea de la magnitud del problema que representa la inconsistencia en la caracterizacion de la suciedad de los billetes por los metodos o sistemas actuales.If one takes into account that the National Central Banks around the world classify billions of bills each year by use, one can get an idea of the magnitude of the problem represented by the inconsistency in the characterization of the dirtiness of banknotes by current methods or systems.

DESCRIPCION DE LA INVENCIONDESCRIPTION OF THE INVENTION

Para conseguir los objetivos y resolver los inconvenientes anteriormente senalados, la invention proporciona un nuevo metodo y dispositivo para la caracterizacion del estado 5 de uso de los billetes de banco, y su clasificacion en aptos y no aptos para la circulation. La invencion tiene por objetivo principal la caracterizacion y detection de la suciedad de los billetes, aunque tambien permite detectar otras caracterlsticas como por ejemplo manchas y escrituras o “grafittis”.In order to achieve the objectives and solve the aforementioned problems, the invention provides a new method and device for the characterization of the state of use of banknotes, and their classification as suitable and not suitable for circulation. The invention has as its main objective the characterization and detection of the dirt of the bills, although it also allows the detection of other characteristics such as stains and writings or "graffiti".

Es conocido que para la caracterizacion de la suciedad de los billetes de banco se 10 empleen metodos que comprenden capturar, por reflexion, al menos una imagen de alIt is known that for the characterization of the dirt of banknotes 10 methods are used that comprise capturing, by reflection, at least one image of the

menos una de las caras de un billete, mediante un dispositivo digital de captura de imagenes por reflexion. Seguidamente se almacena la imagen capturada en un formato estandar, se corrige el color de la imagen captada, utilizando el perfil de color ICC (perfil de color estandarizado por el International Color Consortium) del dispositivo empleado 15 para la captura de la imagen. A continuation se analiza en el dominio espacial la imagenminus one of the faces of a bill, using a digital device for capturing images by reflection. Then, the captured image is stored in a standard format, the color of the captured image is corrected, using the ICC color profile (color profile standardized by the International Color Consortium) of the device used 15 for image capture. The image is then analyzed in the space domain

captada, para lo que su imagen se descompone en los canales de color RGB y se calcula en al menos uno de los tres canales RGB y sobre el valor de los pixeles comprendidos en dicha imagen captada uno o mas estadlsticos (variables) de los pixeles de la imagen de la imagen captada. Los estadlsticos que se calculan estan seleccionados entre la media 20 aritmetica, desviacion, tlpica, maximo, mlnimo, mediana, moda, coeficiente de Kurtosis, y coeficiente de simetrla.captured, for which its image is broken down into RGB color channels and is calculated on at least one of the three RGB channels and on the value of the pixels included in said image captured one or more statistics (variables) of the pixels of The image of the captured image. The statistics that are calculated are selected from the arithmetic mean, deviation, typical, maximum, minimum, median, mode, Kurtosis coefficient, and symmetry coefficient.

La principal novedad de la invencion radica en que tras el analisis en el dominio espacial, se realiza un analisis en el dominio de la frecuencia de la imagen captada para lo que se aplica la transformation matematica conocida como transformada rapida de Fourier (Fast 25 Fourier Transform, FFT) a dicha imagen captada en al menos uno de los canales R, G y B, de forma que se obtiene una nueva imagen denominada espectro de potencias (“power spectrum”). Ademas para realizar el analisis en el dominio de la frecuencia, se calcula, sobre los pixeles de la imagen captada del espectro de potencias, uno o mas de los estadlsticos comentados anteriormente.The main novelty of the invention is that after the analysis in the spatial domain, an analysis is made in the domain of the frequency of the captured image for which the mathematical transformation known as Fast Fourier Transform is applied (Fast 25 Fourier Transform , FFT) to said image captured in at least one of the R, G and B channels, so that a new image called "power spectrum" is obtained. In addition to performing the analysis in the frequency domain, one or more of the statistics mentioned above are calculated on the pixels of the image captured from the power spectrum.

30 La realization del analisis de las imagenes en el dominio de la frecuencia, proporciona una informacion complementaria a la obtenida mediante el analisis tradicional en el dominio espacial.30 The realization of the analysis of the images in the frequency domain provides complementary information to that obtained through the traditional analysis in the spatial domain.

El metodo de la invention tambien comprende calificar visual e individualmente, como aptos o no aptos para la circulation, un conjunto de billetes usados de la misma denomination que los billetes a analizar. Esta clasificacion se realiza por un grupo de expertos prestando especial interes a la region de la imagen captada del billete a 5 analizar.The method of the invention also includes qualifying visually and individually, as suitable or unfit for circulation, a set of used bills of the same denomination as the bills to be analyzed. This classification is carried out by a group of experts paying special interest to the region of the image captured from the ticket to be analyzed.

Se preve realizar la captura de las imagenes individuales del conjunto de los billetes usados que anteriormente fueron calificados visualmente, para seguidamente analizar estas imagenes en todos los canales RGB, tanto en el dominio espacial como en el de la frecuencia, y obtener todos los estadlsticos mencionados anteriormente. Las imagenes 10 captadas se han de corresponder con la imagen captada del billete a analizar.It is planned to capture the individual images of the set of used banknotes that were previously qualified visually, to then analyze these images in all RGB channels, both in the spatial domain and in the frequency domain, and obtain all the mentioned statistics previously. The images captured must correspond to the image captured from the bill to be analyzed.

A continuation se analiza la correlation existente entre la clasificacion visual realizada por el grupo de expertos y los datos obtenidos en el analisis en los dominios espacial y de la frecuencia de dichas imagenes individuales de los billetes calificados visualmente, para determinar los canales RGB y los estadlsticos que ofrecen una mayor correlacion y una 15 mejor caracterizacion del estado de uso del billete. Este analisis de la correlacion se realiza de forma convencional por lo que no se describe en mayor detalle.The correlation between the visual classification carried out by the group of experts and the data obtained in the analysis in the spatial domains and of the frequency of said individual images of the visually qualified banknotes is analyzed below, to determine the RGB and statistic channels that offer a greater correlation and a better characterization of the state of use of the ticket. This correlation analysis is carried out in a conventional manner, so it is not described in greater detail.

Seguidamente se combina al menos un estadlstico obtenido mediante el analisis en el dominio espacial con al menos un estadlstico obtenido mediante el analisis en el dominio de la frecuencia, del billete en analisis, para calcular un unico valor numerico, 20 denominado FV (Fitness Value). Para realizar este calculo se emplean los estadlsticos del analisis descrito en el parrafo anterior que ofrecen una mayor correlacion en la clasificacion realizada visualmente por el grupo de expertos.Next, at least one statistic obtained by means of the analysis in the spatial domain is combined with at least one statistic obtained by means of the analysis in the frequency domain, of the banknote in analysis, to calculate a unique numerical value, 20 called FV (Fitness Value) . To perform this calculation, the statistics of the analysis described in the previous paragraph are used, which offer a greater correlation in the classification carried out visually by the group of experts.

Posteriormente se obtienen unos valores de referencia de los estadlsticos con mayor correlacion del punto anterior, para un billete en perfecto estado del mismo tipo que el 25 billete a analizar. Para ello se captan imagenes individuales de un conjunto de billetes nuevos de la misma denominacion que el billete a analizar, correspondientes a la imagen captada del billete a analizar, se calculan los estadlsticos seleccionados que ofrecen una mayor correlacion en la clasificacion realizada visualmente, y se toman los valores mlnimos como valores de referencia.Subsequently, reference values are obtained for statisticians with greater correlation of the previous point, for a ticket in perfect condition of the same type as the 25 ticket to be analyzed. For this, individual images of a set of new banknotes of the same denomination as the banknote to be analyzed are captured, corresponding to the image captured of the banknote to be analyzed, the selected statistics are calculated that offer a greater correlation in the classification carried out visually, and they take the minimum values as reference values.

30 Ademas la invencion preve fijar un valor FV umbral arbitrario y finalmente se compara el FV calculado, con el FV umbral, y se clasifica el billete analizado como apto o no apto para la circulacion, en funcion de que dicho FV calculado sea menor o mayor que el FV umbral.In addition, the invention provides for setting an arbitrary threshold PV value and finally the calculated PV is compared with the threshold PV, and the analyzed ticket is classified as suitable or not suitable for circulation, depending on whether said calculated PV is lower or higher. than the threshold PV.

Con el procedimiento descrito se evitan los problemas comentados en el estado de la tecnica referentes a baja repetividad y reproducibilidad, tasas de errores cruzados (“crossovers”) y baja correlation con la perception humana de la suciedad en los billetes; ya que la invention, en comparacion con los sistemas existentes en la actualidad, 5 presenta alta repetitividad y reproducibilidad, mejores valores de FFR y FUR y alta correlacion con la percepcion humana de la suciedad en los billetes.With the described procedure, the problems mentioned in the state of the art regarding low repeatability and reproducibility, cross error rates ("crossovers") and low correlation with the human perception of the dirt on the bills are avoided; since the invention, in comparison with the existing systems at present, 5 presents high repeatability and reproducibility, better values of FFR and FUR and high correlation with the human perception of the dirt in the bills.

En la realization preferente de la invencion el FV se calcula como una distancia, desde el billete objeto de analisis hasta un billete perfectamente limpio tomado como referencia, comentado anteriormente, mediante la ecuacion: FV = (£™ 1 ct * (Xt - XRt)n)1/n; donde Xi 10 son los estadlsticos seleccionados, XRi son los valores de referencia para esos estadlsticos, Ci son los factores de correction de escala, n un exponente que consiste en un numero positivo y m es el numero de estadlsticos seleccionados en el calculo de FV. Por lo tanto para la obtencion del FV calculado, se emplea un algoritmo que se calcula como la ralz n de la suma de las diferencias de los valores elevados an de los 15 estadlsticos seleccionados con respecto a los valores de referencia correspondientes a un billete en perfecto estado con nivel de suciedad nulo. Los factores de correccion de escala Ci se pueden obtener convencionalmente.In the preferred embodiment of the invention, the FV is calculated as a distance, from the ticket under analysis to a perfectly clean ticket taken as a reference, discussed above, by the equation: FV = (£ ™ 1 ct * (Xt - XRt) n) 1 / n; where Xi 10 are the selected statistics, XRi are the reference values for those statistics, Ci are the scale correction factors, an exponent consisting of a positive number and m is the number of statistics selected in the calculation of VF. Therefore, to obtain the calculated PV, an algorithm is used that is calculated as the root n of the sum of the differences of the high values of the 15 selected statistics with respect to the reference values corresponding to a perfect ticket state with zero dirt level. The scale correction factors Ci can be obtained conventionally.

En una realizacion de la invencion se preve que entre la fase de corregir el color de la imagen captada y la fase de analizar en el dominio espacial, comprende realizar una 20 selection de al menos una region de interes de la imagen captada para la caracterizacionIn an embodiment of the invention it is anticipated that between the phase of correcting the color of the captured image and the phase of analyzing in the spatial domain, it comprises making a selection of at least one region of interest of the captured image for characterization.

del estado de uso de los billetes. En este caso tanto el analisis espacial como el analisis en frecuencia, unicamente se realiza sobre dicha region de interes; e igualmente la calificacion visual por un grupo de expertos, se realiza prestando especial interes en la region de interes seleccionada. Tambien, en este caso, los valores de referencia se 25 obtienen seleccionando dicha region de interes en las imagenes individuales del conjunto de billetes nuevos de la misma denomination que el billete a analizar.of the state of use of the bills. In this case, both spatial analysis and frequency analysis are only carried out on said region of interest; and also the visual qualification by a group of experts, is carried out paying special interest in the region of interest selected. Also, in this case, the reference values are obtained by selecting said region of interest in the individual images of the set of new bills of the same denomination as the ticket to be analyzed.

Cuando se selecciona al menos una region de interes de la imagen captada, la invencion preve la posibilidad de que dentro de dicha region de interes se seleccione al menos una zona de la region de interes del espectro de potencias. En este caso el analisis espacial 30 y en frecuencia se realiza sobre dicha zona de la region de interes. De la misma forma la calificacion visual por un grupo de expertos, se realiza prestando especial interes en la zona de la region de interes seleccionada, y los valores de referencia se obtienen seleccionando dicha zona de dicha region de interes en las imagenes individuales delWhen at least one region of interest is selected from the captured image, the invention provides for the possibility that within said region of interest at least one area of the region of interest of the power spectrum is selected. In this case, the spatial analysis 30 and in frequency is performed on said area of the region of interest. In the same way the visual qualification by a group of experts, is carried out paying special interest in the area of the region of interest selected, and the reference values are obtained by selecting said area of said region of interest in the individual images of the

conjunto de billetes nuevos de la misma denomination que el billete a analizar.set of new bills of the same denomination as the ticket to be analyzed.

En la invention se establece que cuando el valor FV calculado es menor que el valor FV umbral, el billete analizado es clasificado como apto y cuando es mayor es clasificado como no apto.In the invention it is established that when the calculated PV value is less than the threshold PV value, the analyzed bill is classified as apt and when it is higher it is classified as unfit.

5 Ademas se preve que las imagenes capturadas han de tener al menos una resolution de 60 puntos por cm, con una profundidad de color de al menos 24 bits por pixel y en un formato sin compresion con perdidas.5 It is also expected that the captured images must have at least a resolution of 60 dots per cm, with a color depth of at least 24 bits per pixel and in a format without compression with losses.

En la fase de selection de una region de interes que se efectua para caracterizar la suciedad de los billetes, se selecciona una region que carezca de impresion y queIn the selection phase of a region of interest that is carried out to characterize the dirtiness of the bills, a region that lacks impression and that is selected is selected.

10 represente un porcentaje igual o mayor a un porcentaje previamente establecido de la10 represents a percentage equal to or greater than a previously established percentage of the

superficie total del billete. En la realization preferente este porcentaje es igual o menor que el 10% de la superficie total de billete.total area of the ticket. In the preferred embodiment this percentage is equal to or less than 10% of the total ticket area.

Ademas la invention se refiere a un dispositivo que funciona de acuerdo con el metodo descrito que se caracteriza por que comprende un modulo de captura digital de imagenes 15 por reflexion, mediante el que se capta al menos una imagen digital de al menos una de las caras de un billete. Ademas incluye un procesador de la imagen digital captada, que esta configurado para procesar las imagenes captadas de acuerdo con el procedimiento anteriormente descrito.In addition, the invention relates to a device that operates in accordance with the described method characterized in that it comprises a digital image capture module 15 by reflection, whereby at least one digital image of at least one of the faces is captured of a ticket. It also includes a processor of the captured digital image, which is configured to process the captured images according to the procedure described above.

Respecto al modulo de captura digital de imagenes por reflexion se ha previsto que 20 pueda ser un escaner, una camara digital, o un array CCD (Dispositivo de CargaRegarding the module of digital capture of images by reflection, it has been foreseen that 20 can be a scanner, a digital camera, or a CCD array (Load Device

Acoplada, Charge Coupled Device).Coupled, Charge Coupled Device).

A continuation, para facilitar una mejor comprension de esta memoria descriptiva y formando parte integrante de la misma, se acompana una unica figura en la que con 25 caracter ilustrativo y no limitativo se ha representado el objeto de la invention.Next, in order to facilitate a better understanding of this descriptive report and forming an integral part thereof, a single figure is attached in which the object of the invention has been represented with an illustrative and non-limiting character.

BREVE DESCRIPCION DE LA FIGURABRIEF DESCRIPTION OF THE FIGURE

Figura 1.- Muestra una representation esquematica de las fases que constituyen el metodo de la invencion.Figure 1.- Shows a schematic representation of the phases that constitute the method of the invention.

3030

DESCRIPCION DE UN EJEMPLO DE REALIZACION DE LA INVENCIONDESCRIPTION OF AN EXAMPLE OF EMBODIMENT OF THE INVENTION

A continuation para facilitar una mejor comprension de la invention se describe unA continuation to facilitate a better understanding of the invention describes a

77

ejemplo de realization de la invention.example of realization of the invention.

Como ya ha sido indicado, la invention se refiere a un metodo y a un dispositivo que permiten caracterizar la suciedad en los billetes de banco y su clasificacion en aptos y no aptos para la circulation.As already indicated, the invention relates to a method and a device that allow the characterization of dirt on banknotes and their classification as suitable and not suitable for circulation.

5 Referente al metodo, cabe senalar que en el estado de la tecnica es conocido que para determinar el estado de suciedad de los billetes de banco y clasificarlos como aptos o no aptos para la circulation, se realiza la captura (1) de imagenes por reflexion de los billetes a analizar. Esta captura puede ser de solo el anverso, solo del reverso o del anverso y reverso, mediante el empleo de un escaner, una camara, un array CCD (Dispositivo de 10 Acoplo de Carga) o cualquier otro tipo de modulo de captura digital de imagenes que forma parte del dispositivo de la invencion. En el ejemplo de realization se capta la imagen del anverso del billete a analizar y la imagen captada de cada billete ha de tener una resolution suficiente, considerandose el mlnimo recomendable 60 puntos por cm, una profundidad de color minima de 24 bits por pixel y ha de estar en un formato sin 15 compresion con perdidas.5 Regarding the method, it should be noted that in the state of the art it is known that in order to determine the state of dirtiness of banknotes and classify them as suitable or unfit for circulation, the capture (1) of images by reflection is performed of the bills to analyze. This capture can be of only the obverse, only of the reverse or of the obverse and reverse, by means of the use of a scanner, a camera, a CCD array (Device of 10 Load Attachment) or any other type of digital image capture module which is part of the device of the invention. In the realization example, the image on the front of the banknote to be analyzed is captured and the image captured from each banknote must have a sufficient resolution, considering the minimum recommended 60 points per cm, a minimum color depth of 24 bits per pixel and has of being in a format without 15 compression with losses.

La imagen capturada se almacena (2) en un formato estandar sin compresion con perdidas y se realiza una correction (3) del color de la imagen captada utilizando el perfil de color ICC del modulo empleado para la captura de la imagen.The captured image is stored (2) in a standard format without compression with losses and a correction (3) of the color of the captured image is made using the ICC color profile of the module used for image capture.

Es conocido que el perfil ICC del modulo de captura se construye mediante un proceso 20 de calibration, capturando una tarjeta de calibration y comparando el resultado con losIt is known that the ICC profile of the capture module is constructed by a calibration process 20, capturing a calibration card and comparing the result with the

valores de referencia proporcionados por el suministrador de la tarjeta de calibration, por lo que no se describe en mayor detalle.reference values provided by the supplier of the calibration card, so it is not described in greater detail.

El objeto de la fase de correction de color es garantizar la repetitividad y reproducibilidad del metodo y del dispositivo, ya que gracias a esta correction se minimizan las 25 variaciones debidas al modulo de captura. Mediante esta correction se logra que imagenes del mismo billete procedentes de distintos modulos de captura, o del mismo modulo pero capturadas en diferentes momentos, sean semejantes.The purpose of the color correction phase is to guarantee the repeatability and reproducibility of the method and the device, since thanks to this correction the 25 variations due to the capture module are minimized. Through this correction it is possible that images of the same ticket from different capture modules, or of the same module but captured at different times, are similar.

Sobre la imagen corregida en color se selecciona (16) una o varias regiones de interes. En el ejemplo de realization de la invencion se selecciona una unica region de interes de 30 forma rectangular que comprende la zona de la marca de agua de los billetes, que es la zona en la que el papel no tiene impresion. En cualquier caso, puede utilizarse mas deOne or more regions of interest are selected on the color corrected image (16). In the example of realization of the invention, a unique region of interest of rectangular shape is selected, which comprises the area of the watermark of the bills, which is the area in which the paper has no impression. In any case, more than

una region de interes y de formas diferentes a la rectangular.a region of interest and in ways other than rectangular.

Seguidamente se realiza el analisis (4) en el dominio espacial de la region de interes seleccionada. En el caso en que se seleccionaran mas regiones de interes, dicho analisis en el dominio espacial se realiza sobre dichas regiones de interes seleccionadas.The analysis (4) is then carried out in the spatial domain of the region of interest selected. In the case where more regions of interest were selected, said analysis in the spatial domain is performed on said selected regions of interest.

5 En caso de no definirse ninguna una region de interes, el analisis se llevara a cabo sobre toda la superficie del billete.5 If no region of interest is defined, the analysis will be carried out over the entire surface of the bill.

Para realizar el analisis (4) en el dominio espacial, la imagen corregida en color se descompone (4a) en los tres canales Rojo (R), Verde (G) y Azul (B). En cada uno de estos canales se calculan (4b) unos estadlsticos seleccionados entre media aritmetica, 10 desviacion tlpica, mediana, moda, maximo, mlnimo, coeficiente de Kurtosis y coeficiente de simetrla. En el ejemplo de realizacion de la invention, el analisis se realiza para un billete de 50 € y se ha determinado experimentalmente, tal y como sera explicado con posterioridad, que para el analisis en el dominio espacial se obtiene una buena caracterizacion del estado de uso, calculando la media en el canal verde y la desviacion 15 tlpica en el canal azul de los pixeles encuadrados en la region de interes. En caso de haberse definido mas de una region de interes, estos calculos se realizan sobre los pixeles encuadrados en estas regiones de interes y si no se ha definido ninguna region de interes, estos calculos se haran sobre todos los pixeles de la imagen.To perform the analysis (4) in the spatial domain, the color corrected image is decomposed (4a) into the three channels Red (R), Green (G) and Blue (B). In each of these channels, statistics selected from arithmetic mean, typical deviation, median, mode, maximum, minimum, Kurtosis coefficient and symmetry coefficient are calculated (4b). In the example of carrying out the invention, the analysis is performed for a € 50 bill and it has been determined experimentally, as will be explained later, that for the analysis in the spatial domain a good characterization of the state of use is obtained , calculating the mean in the green channel and the standard deviation 15 in the blue channel of the pixels framed in the region of interest. If more than one region of interest has been defined, these calculations are made on the pixels framed in these regions of interest and if no region of interest has been defined, these calculations will be made on all the pixels of the image.

En el ejemplo de realizacion la imagen corregida en color es de 24 bits por pixel y seIn the embodiment, the color corrected image is 24 bits per pixel and

20 descomponen en los tres canales RGB, cada uno de los cuales tiene 8 bits por pixel.20 decompose into the three RGB channels, each of which has 8 bits per pixel.

Ademas los calculos realizados para el ejemplo de realizacion se obtuvieron como valor de la media en el canal verde = 45,724 y como valor de la desviacion tlpica en el canal azul = 8,142.In addition, the calculations made for the example of realization were obtained as a value of the mean in the green channel = 45,724 and as a value of the typical deviation in the blue channel = 8,142.

Seguidamente sobre el canal verde, se realiza el analisis (5) en el dominio de la 25 frecuencia, para lo que se aplica (5a) la transformada rapida de Fourier o FFT (FastThen on the green channel, the analysis (5) is performed in the 25 frequency domain, for which the fast Fourier transform or FFT (Fast) is applied (5a)

Fourier Transform), obteniendo en una nueva imagen que recibe el nombre de PowerFourier Transform), obtaining in a new image that is called Power

Spectrum. A su vez sobre el Power Spectrum se puede definir una o varias zonas deSpectrum In turn on the Power Spectrum you can define one or several zones of

interes de cualquier forma, calculandose (5b) al menos un estadlstico de entre todos los comentados con anterioridad en la zona o zonas de interes del Power Spectrum. En el 30 ejemplo de realizacion no se ha definido ninguna zona de interes de la region de interes, por lo que este calculo se hace sobre todos los pixeles del Power Spectrum y se ha determinado experimentalmente, tal y como sera explicado con posterioridad, que seinterest in any way, calculating (5b) at least one statistic among all those previously mentioned in the area or areas of interest of the Power Spectrum. In the example of embodiment, no area of interest of the region of interest has been defined, so this calculation is made on all Power Spectrum pixels and has been determined experimentally, as will be explained later, which

obtiene una buena caracterizacion del estado de uso mediante el calculo de la desviacion tlpica del valor de los pixeles en la imagen que se obtiene al aplicar la transformada rapida de Fourier en la imagen de la region de interes.it obtains a good characterization of the state of use by calculating the typical deviation of the value of the pixels in the image that is obtained when applying the fast Fourier transform in the image of the region of interest.

Concretamente en el ejemplo de realization el valor de la desviacion tlpica en el espectro 5 de potencias = 19,436Specifically in the realization example, the value of the standard deviation in the power spectrum 5 = 19,436

En caso de no haberse definido ninguna region de interes en las imagenes corregidas en color, la FFT se aplicara a la totalidad de la imagen del billete.If no region of interest has been defined in the color corrected images, the FFT will be applied to the entire bill image.

Para establecer experimentalmente los estadlsticos que se han de calcular, previamente se realiza una clasificacion (6) visual e individual de un conjunto de billetes de la misma 10 denomination que los billetes a analizar, como aptos o no aptos para la circulation. Esta clasificacion (6) se realiza por un grupo de expertos prestando especial interes en la region de interes.To experimentally establish the statistics that have to be calculated, a visual and individual classification (6) of a set of bills of the same denomination as the bills to be analyzed is previously carried out, as suitable or not suitable for circulation. This classification (6) is carried out by a group of experts paying special interest in the region of interest.

Ademas se capturan (7) las imagenes individuales del conjunto de billetes anteriormente calificados visualmente y se analizan (8), en todos los canales RGB, en el dominio 15 espacial y de la frecuencia, las imagenes individuales de los billetes calificados visualmente, y se obtienen todos los estadlsticos mencionados anteriormente (media aritmetica, desviacion tlpica, mediana, moda, maximo, mlnimo, coeficiente de Kurtosis y coeficiente de simetrla).In addition, the individual images of the set of previously qualified banknotes are captured (7) and analyzed (8), on all RGB channels, in the spatial and frequency domain, the individual images of the banknotes visually qualified, and they obtain all the statistics mentioned above (arithmetic mean, typical deviation, median, mode, maximum, minimum, Kurtosis coefficient and symmetry coefficient).

Seguidamente se analiza (9) la correlation entre la clasificacion visual realizada por el 20 grupo de expertos y los datos obtenidos en el analisis en los dominios espacial y de la frecuencia de las imagenes individuales de dichos billetes calificados (6) visualmente, para establecer los canales RGB y los estadlsticos que ofrecen una mayor correlacion y una mejor caracterizacion del estado de uso del billete y se toman los valores mlnimos como valores de referencia. En el ejemplo de realizacion los estadlsticos calculados que 25 ofrecen una mejor correlacion se corresponden con los especificados anteriormente; esto es desviacion media en el canal verde, desviacion tlpica en el canal azul, y desviacion tlpica en el espectro de potencias del canal verde.Next, the correlation between the visual classification carried out by the group of experts and the data obtained in the analysis in the spatial domains and the frequency of the individual images of said banknotes (6) visually is analyzed (9), to establish the RGB channels and statistics that offer a higher correlation and a better characterization of the state of use of the bill and the minimum values are taken as reference values. In the exemplary embodiment, the calculated statistics that offer a better correlation correspond to those specified above; this is mean deviation in the green channel, typical deviation in the blue channel, and typical deviation in the power spectrum of the green channel.

En el ejemplo de realizacion los estadlsticos calculados de los billetes de referencia son: media en el canal verde = 39,440, desviacion tlpica en el canal azul = 6,940, y desviacion 30 tlpica en el espectro de potencias = 10,183.In the exemplary embodiment, the statistics calculated from the reference banknotes are: mean in the green channel = 39,440, typical deviation in the blue channel = 6,940, and typical deviation in the power spectrum = 10,183.

Se obtienen unos valores de referencia (12) de los estadlsticos con mayor correlacion delReference values (12) of the statisticians with greater correlation of the

punto anterior, correspondientes a un billete en perfecto estado del mismo tipo que el billete a analizar, para lo que se captan (10) imagenes individuales de un conjunto de billetes nuevos de la misma denomination que el billete a analizar, se selecciona la misma region de interes que en los billetes a analizar, se calculan (11) los estadlsticos 5 seleccionados que ofrecen una mayor correlation en la clasificacion realizada visualmente, y se toman los valores mlnimos como valores de referencia (12).previous point, corresponding to a ticket in perfect condition of the same type as the ticket to be analyzed, for which (10) individual images of a set of new bills of the same denomination as the ticket to be analyzed are captured, the same region is selected of interest that in the banknotes to be analyzed, the selected statistics 5 are calculated (11) that offer a greater correlation in the classification carried out visually, and the minimum values are taken as reference values (12).

Seguidamente se combina (13) al menos un estadlstico obtenido mediante el analisis en el dominio espacial con al menos un estadlstico obtenido mediante el analisis en el dominio de la frecuencia, del billete en analisis, para calcular un valor unico denominado 10 FV, en cuyo calculo se emplean los estadlsticos del analisis (9) descrito en el punto anterior que ofrecen una mayor correlacion en la clasificacion realizada visualmente por el grupo de expertos.Next, at least one statistic obtained by analyzing in the spatial domain is combined with at least one statistic obtained by analyzing in the frequency domain, of the bill under analysis, to calculate a unique value called 10 FV, in which I calculate the statistics used in the analysis (9) described in the previous point that offer a greater correlation in the classification carried out visually by the group of experts.

El valor FV se calcula como una distancia, desde el billete objeto de analisis hasta un billete perfectamente limpio tomado como referencia, comentado anteriormente, mediante 15 la ecuacion: FV = (££!! 1Q * (X* - XRn)1/n ;The PV value is calculated as a distance, from the ticket under analysis to a perfectly clean ticket taken as a reference, discussed above, using the equation: FV = (££ !! 1Q * (X * - XRn) 1 / n ;

donde Xi son los estadlsticos seleccionados, XRi son los valores de referencia para esos estadlsticos (p.e. los valores mlnimos), Ci son los factores de correction de escala, el exponente n es un numero positivo y m es el numero de estadlsticos seleccionados en el calculo de FV.where Xi are the selected statistics, XRi are the reference values for those statistics (eg the minimum values), Ci are the scale correction factors, the exponent n is a positive number and m is the number of statistics selected in the calculation of PV.

20 El valor del exponente n se obtiene experimentalmente de forma que se maximice el porcentaje de coincidencia entre la clasificacion visual llevada a cabo y el analisis realizado de dichos billetes clasificados visualmente, tal y como ya fue descrito con anterioridad.20 The value of the exponent n is obtained experimentally so as to maximize the percentage of coincidence between the visual classification carried out and the analysis performed of said visually classified notes, as already described previously.

En el ejemplo de realization se han seleccionado los tres estadlsticos ya comentados y 25 por tanto m = 3 que fue indicado, se obtiene experimentalmente (media canal verde, desviacion estandar canal azul y desviacion estandar de la FFT del canal verde). Normalmente m esta comprendido entre 2 y 4, ya que con mayor numero de estadlsticos el calculo de FV aumenta mucho en complejidad y los resultados en la clasificacion de los billetes como aptos o no aptos no mejora casi nada.In the example of realization, the three statistics already mentioned have been selected and 25 therefore m = 3 which was indicated, is obtained experimentally (half green channel, standard deviation blue channel and standard deviation of the FFT of the green channel). Normally m is between 2 and 4, since with a greater number of statisticians the calculation of VF increases greatly in complexity and the results in the classification of bills as suitable or unfit do not improve almost anything.

30 En este ejemplo los factores de correccion de escala Ci son 3,37, 107,88 y 26.96, que se obtienen convencionalmente, y el exponente n = 2,4.30 In this example the correction factors of scale Ci are 3.37, 107.88 and 26.96, which are obtained conventionally, and the exponent n = 2.4.

Por lo tanto, de acuerdo con el ejemplo de realizacion el FV = (3,37*(45,724 - 39,440)2,4 +Therefore, according to the embodiment example the FV = (3.37 * (45,724 - 39,440) 2,4 +

11eleven

107,88* (8,142 - 6,940)24 + 26.96* (19,436 - 10,183)2'4)1/2'4 = 62,66.107.88 * (8,142 - 6,940) 24 + 26.96 * (19,436 - 10,183) 2'4) 1 / 2'4 = 62.66.

Ademas se fija (14) un valor FV umbral que en el caso de los billetes de 50 € del ejemplo de realization, es de 50.In addition, a threshold PV value is set (14), which in the case of the € 50 banknotes of the realization example is 50.

Finalmente se compara (15) el FV calculado, con el FV umbral, para clasificar el billete 5 analizado como apto o no apto para la circulation, en funcion de que dicho FV calculado sea menor o mayor que el FV umbral.Finally, the calculated PV is compared (15) with the threshold PV, to classify the analyzed banknote 5 as suitable or not suitable for circulation, depending on whether said calculated PV is less than or greater than the threshold PV.

De acuerdo con el ejemplo de realizacion si el valor obtenido para el billete objeto del estudio es superior al valor umbral, el billete no es apto para seguir en circulacion, y en caso contrario si que es apto para seguir en circulacion. En el caso que nos ocupa como 10 el valor FV obtenido es 62,66 que es mayor que el valor FV umbral de 50; entonces el billete analizado no es apto para seguir en circulacion.According to the example of realization if the value obtained for the ticket object of the study is higher than the threshold value, the ticket is not suitable to continue in circulation, and otherwise it is apt to continue in circulation. In the case that concerns us as 10, the PV value obtained is 62.66 which is greater than the threshold PV value of 50; Then the analyzed ticket is not suitable to continue in circulation.

De acuerdo con la descripcion realizada para el calculo del FV se toman todas o parte de las mediciones obtenidas mediante los analisis en el dominio espacial y en el dominio de la frecuencia, y cuales se utilizaran y cuales se descartaran, depende de las 15 caracterlsticas de los billetes analizados y especialmente del tinte del papel. En cualquier caso, el mlnimo a utilizar en el calculo del FV es una media o desviacion procedente del analisis en dominio espacial y una media o desviacion tlpica procedente del analisis en el dominio de la frecuencia.According to the description made for the calculation of the VF, all or part of the measurements obtained through the analyzes in the spatial domain and in the frequency domain are taken, and which ones will be used and which will be discarded, depends on the 15 characteristics of the banknotes analyzed and especially the paper dye. In any case, the minimum to be used in the calculation of the VF is a mean or deviation from the analysis in the spatial domain and a mean or standard deviation from the analysis in the frequency domain.

Ademas la invention se refiere a un dispositivo que, tal y como fue descrito comprende 20 un modulo de captation de imagenes por reflexion (escaner, etc.) que se complementa con un procesador de la imagen digital captada que esta configurado para procesar la imagen digital captada de acuerdo con las anteriores fases descritas del metodo.In addition, the invention relates to a device that, as described, comprises a module for capturing images by reflection (scanner, etc.) that is complemented by a digital image processor that is configured to process the digital image. captured according to the previous described phases of the method.

Claims (10)

1010 15fifteen 20twenty 2525 REIVINDICACIONES 1.- METODO PARA LA CARACTERIZACION DEL ESTADO DE USO DE LOS BILLETES DE BANCO, Y SU CLASIFICACION EN APTOS Y NO APTOS PARA LA CIRCULACION, que comprende:1.- METHOD FOR THE CHARACTERIZATION OF THE STATE OF USE OF BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION, which includes: - capturar (1) por reflexion al menos una imagen de al menos una de las caras de un billete a analizar, mediante un modulo digital de captura de imagenes por reflexion,- capture (1) by reflection at least one image of at least one of the faces of a bill to be analyzed, by means of a digital module for capturing images by reflection, - almacenar (2) la imagen captada en un formato estandar;- store (2) the image captured in a standard format; - corregir (3) el color de la imagen captada, utilizando el perfil de color ICC del modulo empleado para la captura de la imagen,- correct (3) the color of the captured image, using the ICC color profile of the module used to capture the image, - analizar (4) en el dominio espacial la imagen captada de acuerdo con las siguientes fases:- analyze (4) in the spatial domain the image captured according to the following phases: o Descomponer (4a) dicha imagen captada en los tres canales RGB: rojo (Red), verde (Green) y azul (Blue),o Decompose (4a) said image captured in the three RGB channels: red (Red), green (Green) and blue (Blue), o calcular (4b) en al menos uno de los tres canales R, G y B, y sobre el valor de los pixeles comprendidos en dicha imagen captada, al menos un estadlstico seleccionado entre : media aritmetica, desviacion tlpica, mediana, moda, maximo, mlnimo, coeficiente de Kurtosis, y coeficiente de simetrla;or calculate (4b) on at least one of the three channels R, G and B, and on the value of the pixels included in said captured image, at least one statistic selected from: arithmetic mean, typical deviation, median, mode, maximum , minimum, Kurtosis coefficient, and symmetry coefficient; caracterizado por que ademas comprende:characterized in that it also includes: - Analizar (5) en el dominio de la frecuencia la imagen captada de acuerdo con las siguientes fases:- Analyze (5) in the frequency domain the image captured according to the following phases: o Aplicar (5a) la transformation matematica conocida como transformada rapida de Fourier (Fast Fourier Transform, FFT) a la imagen captada en al menos uno de los canales R, G y B, para obtener una nueva imagen denominada espectro de potencias ("power spectrum”),o Apply (5a) the mathematical transformation known as Fast Fourier Transform (FFT) to the image captured in at least one of the R, G and B channels, to obtain a new image called a power spectrum ("power spectrum ”), o Calcular (5b) sobre los pixeles de la imagen captada, del espectro de potencias al menos un estadlstico seleccionado entre: media aritmetica, desviacion tlpica, mediana, moda, maximo, mlnimo, coeficiente de Kurtosis y coeficiente de simetrla,o Calculate (5b) on the pixels of the captured image, of the power spectrum of at least one statistic selected from: arithmetic mean, typical deviation, median, mode, maximum, minimum, Kurtosis coefficient and symmetry coefficient, - calificar (6) visual e individualmente, como aptos o no aptos para la circulation, un conjunto de billetes de la misma denomination que los billetes a analizar, por un grupo de expertos, prestando especial interes en la region de la imagen captada del billete a analizar,- Qualify (6) visually and individually, as suitable or not suitable for circulation, a set of bills of the same denomination as the bills to be analyzed, by a group of experts, paying special interest in the region of the image captured from the ticket to analyze, 5 - capturar (7) las imagenes individuales del conjunto de billetes anteriormente calificados5 - capture (7) the individual images of the set of previously qualified banknotes visualmente, donde dichas imagenes se corresponden a la imagen captada del billete a analizar,visually, where these images correspond to the image captured from the bill to be analyzed, - analizar (8) en todos los canales RGB, en los dominios espacial y de la frecuencia, las imagenes individuales de los billetes calificados visualmente, y obtener todos los- analyze (8) in all RGB channels, in the spatial and frequency domains, the individual images of the visually qualified banknotes, and obtain all 10 estadlsticos mencionados anteriormente,10 statisticians mentioned above, - analizar (9) la correlation entre la clasificacion visual realizada por el grupo de expertos y los datos obtenidos en el analisis en los dominios espacial y de la frecuencia de las imagenes individuales de dichos billetes calificados visualmente, para establecer los canales RGB y los estadlsticos que ofrecen una mayor correlacion y una mejor- analyze (9) the correlation between the visual classification performed by the group of experts and the data obtained in the analysis in the spatial domains and of the frequency of the individual images of said visually qualified banknotes, to establish the RGB and statistic channels that offer a higher correlation and a better 15 caracterizacion del estado de uso del billete,15 characterization of the state of use of the ticket, - combinar (13) al menos un estadlstico obtenido mediante el analisis en el dominio espacial con al menos un estadlstico obtenido mediante el analisis en el dominio de la frecuencia, del billete en analisis, para calcular un unico valor numerico, denominado FV (Fitness Value), en cuyo calculo se emplean los estadlsticos del analisis (9) descrito en- combine (13) at least one statistic obtained by analysis in the spatial domain with at least one statistic obtained by analysis in the frequency domain, of the ticket under analysis, to calculate a unique numerical value, called FV (Fitness Value ), in whose calculation the statistics of the analysis (9) described in 20 el punto anterior que ofrecen una mayor correlacion en la clasificacion (6) visual realizada por el grupo de expertos,20 the previous point that offer a greater correlation in the visual classification (6) made by the group of experts, - obtener unos valores de referencia (12) de los estadlsticos con mayor correlacion del punto anterior, correspondientes a un billete en perfecto estado del mismo tipo que el billete a analizar, para lo que se captan (10) imagenes individuales de un conjunto de- obtain reference values (12) of the statisticians with greater correlation of the previous point, corresponding to a bill in perfect condition of the same type as the ticket to be analyzed, for which (10) individual images of a set of 25 billetes nuevos de la misma denominacion que el billete a analizar, correspondiente a la imagen captada del billete a analizar, se calculan (11) los estadlsticos seleccionados que ofrecen una mayor correlacion en la clasificacion realizada visualmente, y se toman los valores mlnimos como valores de referencia (12),25 new banknotes of the same denomination as the banknote to be analyzed, corresponding to the image captured of the banknote to be analyzed, are selected (11) the selected statistics that offer a greater correlation in the classification carried out visually, and the minimum values are taken as values reference (12), - fijar (14) un FV umbral y comparar (15) el FV calculado, con el FV umbral, para 30 clasificar el billete analizado como apto o no apto para la circulacion, en funcion de que- set (14) a threshold PV and compare (15) the calculated PV, with the threshold PV, to classify the banknote analyzed as suitable or not suitable for circulation, depending on whether dicho FV calculado sea menor o mayor que el FV umbral.said calculated PV is less than or greater than the threshold PV. 2.- METODO PARA LA CARACTERIZACION DEL ESTADO DE USO DE LOS2.- METHOD FOR THE CHARACTERIZATION OF THE STATE OF USE OF THE BILLETES DE BANCO, Y SU CLASIFICACION EN APTOS Y NO APTOS PARA LA CIRCULACION, segun reivindicacion 1, caracterizado por que el FV se calcula como una distancia, desde el billete objeto de analisis hasta el billete perfectamente limpio tomado como referencia, mediante la ecuacion: FV = (Z™ 1Ci * (Xi - XRj) n)1/n ; donde Xi son 5 los estadlsticos seleccionados, XRi son los valores de referencia para esos estadlsticos, Ci son factores de correction de escala, m es el numero de estadlsticos seleccionados en el calculo de FV que se obtiene experimentalmente; y el exponente n es un numero positivo que se obtiene experimentalmente de forma que se maximice el porcentaje de coincidencia entre la clasificacion visual (6) llevada a cabo y el analisis (9) realizado de 10 dichos billetes clasificados visualmente.BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION, according to claim 1, characterized in that the FV is calculated as a distance, from the ticket under analysis to the perfectly clean ticket taken as a reference, by the equation: FV = (Z ™ 1Ci * (Xi - XRj) n) 1 / n; where Xi are 5 the selected statistics, XRi are the reference values for those statistics, Ci are scale correction factors, m is the number of statistics selected in the calculation of VF that is obtained experimentally; and the exponent n is a positive number that is obtained experimentally so as to maximize the percentage of coincidence between the visual classification (6) carried out and the analysis (9) made of 10 said visually classified notes. 3. - METODO PARA LA CARACTERIZACION DEL ESTADO DE USO DE LOS BILLETES DE BANCO, Y SU CLASIFICACION EN APTOS Y NO APTOS PARA LA CIRCULACION, segun reivindicacion 1, caracterizado por que entre la fase de corregir (3) el color de la imagen captada y la fase de analizar (4) en el dominio espacial,3. - METHOD FOR THE CHARACTERIZATION OF THE STATE OF USE OF BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION, according to claim 1, characterized in that between the phase of correcting (3) the color of the captured image and the phase of analyzing (4) in the spatial domain, 15 comprende seleccionar (16) al menos una region de interes de la imagen captada para la caracterizacion del estado de uso de los billetes, realizandose el analisis espacial (4) y en frecuencia (5) unicamente sobre dicha region de interes; donde la calificacion (6) visual por un grupo de expertos, se realiza prestando especial interes en la region de interes seleccionada, y donde los valores de referencia (12) se obtienen seleccionando 20 dicha region de interes en las imagenes individuales del conjunto de billetes nuevos de la misma denominacion que el billete a analizar.15 comprises selecting (16) at least one region of interest from the image captured for the characterization of the state of use of the bills, the spatial analysis (4) and frequency (5) being performed only on said region of interest; where the visual qualification (6) by a group of experts is carried out paying special interest in the region of interest selected, and where the reference values (12) are obtained by selecting said region of interest in the individual images of the set of bills new of the same denomination as the ticket to analyze. 4. - METODO PARA LA CARACTERIZACION DEL ESTADO DE USO DE LOS BILLETES DE BANCO, Y SU CLASIFICACION EN APTOS Y NO APTOS PARA LA CIRCULACION, segun reivindicacion 3, caracterizado por que comprende seleccionar al4. - METHOD FOR CHARACTERIZING THE STATE OF USE OF BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION, according to claim 3, characterized in that it comprises selecting the 25 menos una zona de la region de interes, del espectro de potencias realizandose el analisis espacial (4) y en frecuencia (5) sobre dicha zona de la region de interes; donde la calificacion (6) visual por un grupo de expertos, se realiza prestando especial interes en la zona de la region de interes seleccionada, y donde los valores de referencia (12) se obtienen seleccionando dicha zona de dicha region de interes en las imagenes 30 individuales del conjunto de billetes nuevos de la misma denomination que el billete a analizar.25 minus one area of the region of interest, of the spectrum of powers, performing the spatial analysis (4) and frequency (5) on said area of the region of interest; where the visual qualification (6) by a group of experts is carried out paying special interest in the area of the region of interest selected, and where the reference values (12) are obtained by selecting said area from said region of interest in the images 30 individuals from the set of new bills of the same denomination as the ticket to be analyzed. 5. - METODO PARA LA CARACTERIZACION DEL ESTADO DE USO DE LOS BILLETES DE BANCO, Y SU CLASIFICACION EN APTOS Y NO APTOS PARA LA CIRCULACION, segun reivindicacion 1, caracterizado por que cuando el FV calculado es5. - METHOD FOR THE CHARACTERIZATION OF THE STATE OF USE OF BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION, according to claim 1, characterized in that when the calculated PV is 5 menor que el valor umbral, el billete analizado es clasificado como apto y cuando es mayor es clasificado como no apto.5 less than the threshold value, the analyzed ticket is classified as eligible and when it is higher it is classified as unfit. 6. - METODO PARA LA CARACTERIZACION DEL ESTADO DE USO DELOS6. - METHOD FOR CHARACTERIZING THE STATE OF USE OF DELOS BILLETES DE BANCO, Y SU CLASIFICACION EN APTOS Y NO APTOS PARA LA CIRCULACION, segun reivindicacion 1, caracterizado por que la al menos imagenBANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION, according to claim 1, characterized in that the at least image 10 capturada ha de tener al menos una resolution de 60 puntos por cm, con una10 captured must have at least a resolution of 60 dots per cm, with a profundidad de color de al menos 24 bits por pixel y en un formato sin compresion con perdidas.color depth of at least 24 bits per pixel and in an uncompressed format with losses. 7. - METODO PARA LA CARACTERIZACION DEL ESTADO DE USO DE LOS BILLETES DE BANCO, Y SU CLASIFICACION EN APTOS Y NO APTOS PARA LA7. - METHOD FOR THE CHARACTERIZATION OF THE STATE OF USE OF BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR THE 15 CIRCULACION, segun reivindicacion 3, caracterizado por que la selection de una region15 CIRCULATION, according to claim 3, characterized in that the selection of a region de interes para la caracterizacion de la suciedad de los billetes comprende seleccionarof interest for the characterization of the dirt of the bills includes selecting una region que carezca de impresion y que represente un porcentaje igual o mayor a un porcentaje previamente establecido de la superficie total del billete.a region that lacks impression and that represents a percentage equal to or greater than a previously established percentage of the total bill area. 8. - METODO PARA LA CARACTERIZACION DEL ESTADO DE USO DE LOS 20 BILLETES DE BANCO, Y SU CLASIFICACION EN APTOS Y NO APTOS PARA LA8. - METHOD FOR THE CHARACTERIZATION OF THE STATE OF USE OF THE 20 BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR THE CIRCULACION, segun reivindicacion 7, caracterizado por que el porcentaje previamente establecido que carezca de impresion ha de ser igual o menor que el 10% de la superficie total de billete.CIRCULATION, according to claim 7, characterized in that the previously established percentage that lacks printing must be equal to or less than 10% of the total bill area. 9. - DISPOSITIVO PARA LA CARACTERIZACION DEL ESTADO DE USO DELOS 25 BILLETES DE BANCO, Y SU CLASIFICACION EN APTOS Y NO APTOS PARA LA9. - DEVICE FOR THE CHARACTERIZATION OF THE STATE OF USE OF THE 25 BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR THE CIRCULACION, caracterizado por que comprende:CIRCULATION, characterized in that it comprises: - un modulo de captura digital de imagenes por reflexion, para captar al menos una imagen digital de al menos una de las caras de un billete,- a module for digital capture of images by reflection, to capture at least one digital image of at least one of the faces of a bill, - un procesador de la imagen digital captada, configurado para funcionar de acuerdo con 30 el metodo de las reivindicaciones 1 a 8.- a digital image processor captured, configured to operate according to the method of claims 1 to 8. 10.- DISPOSITIVO PARA LA CARACTERIZACION DEL ESTADO DE USO DELOS BILLETES DE BANCO, Y SU CLASIFICACION EN APTOS Y NO APTOS PARA LA CIRCULACION, segun reivindicacion 9, caracterizado por que el modulo de captura digital de imagenes por reflexion esta seleccionado entre un escaner, una camara digital, 5 y un array CCD (Dispositivo de Carga Acoplada, Charge Coupled Device).10.- DEVICE FOR THE CHARACTERIZATION OF THE STATE OF USE OF THE BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION, according to claim 9, characterized in that the module of digital capture of images by reflection is selected from a scanner, a digital camera, 5 and a CCD array (Charged Couple Device, Charge Coupled Device).
ES201430239A 2014-02-21 2014-02-21 METHOD AND DEVICE FOR THE CHARACTERIZATION OF THE STATE OF USE OF BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION Expired - Fee Related ES2549461B1 (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
ES201430239A ES2549461B1 (en) 2014-02-21 2014-02-21 METHOD AND DEVICE FOR THE CHARACTERIZATION OF THE STATE OF USE OF BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION
EP15155865.7A EP2911123B1 (en) 2014-02-21 2015-02-20 A method and device for characterising the state of use of banknotes, and their classification as fit and unfit for circulation
ES15155865.7T ES2591228T3 (en) 2014-02-21 2015-02-20 Method and device for characterizing the state of use of banknotes and their classification as suitable and not suitable for circulation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
ES201430239A ES2549461B1 (en) 2014-02-21 2014-02-21 METHOD AND DEVICE FOR THE CHARACTERIZATION OF THE STATE OF USE OF BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION

Publications (3)

Publication Number Publication Date
ES2549461A2 ES2549461A2 (en) 2015-10-28
ES2549461R1 ES2549461R1 (en) 2015-12-01
ES2549461B1 true ES2549461B1 (en) 2016-10-07

Family

ID=52544344

Family Applications (2)

Application Number Title Priority Date Filing Date
ES201430239A Expired - Fee Related ES2549461B1 (en) 2014-02-21 2014-02-21 METHOD AND DEVICE FOR THE CHARACTERIZATION OF THE STATE OF USE OF BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION
ES15155865.7T Active ES2591228T3 (en) 2014-02-21 2015-02-20 Method and device for characterizing the state of use of banknotes and their classification as suitable and not suitable for circulation

Family Applications After (1)

Application Number Title Priority Date Filing Date
ES15155865.7T Active ES2591228T3 (en) 2014-02-21 2015-02-20 Method and device for characterizing the state of use of banknotes and their classification as suitable and not suitable for circulation

Country Status (2)

Country Link
EP (1) EP2911123B1 (en)
ES (2) ES2549461B1 (en)

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3023694B2 (en) * 1990-07-27 2000-03-21 セイコーインスツルメンツ株式会社 Light pattern recognition method for multi-reference images
GB9413413D0 (en) * 1994-07-04 1994-08-24 At & T Global Inf Solution Apparatus and method for testing bank-notes
US6978036B2 (en) * 1998-07-31 2005-12-20 Digimarc Corporation Tamper-resistant authentication techniques for identification documents
JP4180715B2 (en) * 1998-12-14 2008-11-12 株式会社東芝 Device for determining the degree of contamination of printed matter
NL1011766C2 (en) * 1999-04-09 2000-10-10 Nl Bank Nv Machine for sorting securities.
EP1864825B1 (en) * 2005-03-30 2014-05-28 National Printing Bureau, Incorporated Administrative Agency Printed product, method and device for detecting such printed product, and authentication method and device
US20070140551A1 (en) * 2005-12-16 2007-06-21 Chao He Banknote validation
US8096584B2 (en) * 2006-07-24 2012-01-17 3M Innovative Properties Company Document authentication using template matching with fast masked normalized cross-correlation
NL2006990C2 (en) * 2011-06-01 2012-12-04 Nl Bank Nv Method and device for classifying security documents such as banknotes.

Also Published As

Publication number Publication date
EP2911123A1 (en) 2015-08-26
EP2911123B1 (en) 2016-07-06
ES2549461R1 (en) 2015-12-01
ES2549461A2 (en) 2015-10-28
ES2591228T3 (en) 2016-11-25

Similar Documents

Publication Publication Date Title
US10204424B2 (en) Color identifying system, color identifying method and display device
AU2016278954B2 (en) Methods and a computing device for determining whether a mark is genuine
TWI696450B (en) Information processing device, information processing system, and information processing program
CN106163384B (en) Information processing device, information processing method, and program
EP2859833A1 (en) Image processing device, image processing method, and image processing program
US10388099B2 (en) Paper currency fold recognition apparatus and method
CN103093274B (en) Method based on the people counting of video
US10607361B2 (en) Method for observing wildlife population size through remote sensing imagery
US10181198B2 (en) Data processing apparatus, color identification method, non-transitory computer readable medium, and color chart
CN105706147A (en) Hand-held device and method for authenticating a marking
CN104572971A (en) Image retrieval method and device
CN110490839A (en) The method, apparatus and computer equipment of failure area in a kind of detection highway
CN106295491A (en) Track line detection method and device
CN104573647A (en) Recognition method and device for illegal identity document
WO2020186043A1 (en) Systems and methods for detection of counterfeit documents
CN110059607B (en) Living body multiplex detection method, living body multiplex detection device, computer equipment and storage medium
CN108021913A (en) Certificate photograph information identifying method and device
ES2549461B1 (en) METHOD AND DEVICE FOR THE CHARACTERIZATION OF THE STATE OF USE OF BANK TICKETS, AND ITS CLASSIFICATION IN APTOS AND NOT SUITABLE FOR CIRCULATION
CN103200349B (en) Scanned image color cast automatic detection method
JP2019128188A (en) Mass spectrum processing apparatus and method
EP2960874A1 (en) Identity document digital image correction method, system and computer program
CN108537945B (en) Bill watermark detection method and system and self-service equipment
Hu et al. Jpeg ringing artifact visibility evaluation
CN202533362U (en) Printed matter authenticity identification device based on multispectral imaging technology
JP2015149008A (en) authentication system and authentication method

Legal Events

Date Code Title Description
FG2A Definitive protection

Ref document number: 2549461

Country of ref document: ES

Kind code of ref document: B1

Effective date: 20161007

FD2A Announcement of lapse in spain

Effective date: 20210929