KR101706368B1 - Method and Apparatus for Generating Efficiency Map for Identifying Money - Google Patents

Abstract

The present invention relates to a control apparatus for performing a logical operation, comprising: a measurement unit for outputting a measured physical value obtained by measuring a specific physical value; And a controller for performing a logical operation using the used physical value as a parameter, wherein the controller uses the measured physical value as the used physical value when it is in a normal state, The measured physical value immediately before the transition from the steady state to the ignore state is used as the used physical value. If the measured physical value falls within the preset normal range even in the neglected state, Wherein the correction value obtained by correcting the measured physical value is used as the used physical value and the fault counter value increases with time when the measured physical value deviates from the normal range and decreases with time when entering the normal range , And the ignore state is a state in which the fault counter value is larger than the first counter value and smaller than the counter threshold value It relates to the device.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a method and apparatus for generating an effective map,

Field of the Invention [0002] The present invention relates to a method and apparatus for generating an effective map for currency identification, and more particularly, to a method and apparatus for generating a valid map of a valid region on a currency to be identified when money is inserted in the money identification apparatus.

In the case where a plurality of kinds of currencies are mixed and supplied in a batch, an apparatus or an apparatus for discriminating specific currencies from each other and discriminating each inserted currencies has been conventionally used. Such a money discriminating apparatus is used as a device for money discrimination itself, and is also widely used for a counter for counting money such as banknotes and an automatic money depositing machine for money.

However, conventionally, complicated image analysis has to be performed for accurate and quick money discrimination, so that there has been a problem in that expensive equipment or high-grade equipment must be used. In particular, various currencies exist in each country, and there are various types of currencies (bills) in each country, and there are four types of monetary images, such as front / back and two directions, Therefore, in order to accurately identify such various kinds of money by the money discriminating apparatus and to secure the identification accuracy thereof, there is a problem in that it requires more analysis time and efforts and increases the time required for identification.

As a result, conventionally, in order to accurately and quickly identify such a variety of image patterns, expensive and high-performance image analysis equipments have to be completed, and analysis time is also required.

The background art of the present invention is disclosed in Korean Patent Laid-Open Publication No. 2006-0101937 (published on Mar. 27, 2006).

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and it is an object of the present invention to provide a currency discrimination apparatus and a currency discrimination apparatus which can reduce a processing load on a currency discrimination apparatus, And to provide an effective map generation method and apparatus for monetary identification.

According to an aspect of the present invention, there is provided a method of processing an image, the method comprising: inputting a plurality of training images for a specific face of a specific currency; A reference image generating unit generates a reference image by averaging the input training images; A correlation map generating section generates an intra-class correlation map showing a correlation between the corresponding first monetary images and the reference image of the same denomination as the reference image, and a correlation image in a class different from the first monetary images, Generating an inter-class correlation map indicating a correlation between the reference image and the reference image; And a validity map generating unit for generating a validity map concerning a valid area on the currency to be identified at the time of denomination, based on the intra-class correlation map and the inter-class correlation map. And provides an effective map generation method for money identification.

In the present invention, the reference image is generated by averaging pixel values of corresponding pixels of the plurality of training images.

In the present invention, the correlation map generator may calculate a correlation value between each of the first monetary images and the reference image for each pixel, average the correlation values for the calculated pixel on a pixel by pixel basis, And generates a map.

In the present invention, the correlation map generating unit may calculate a correlation map based on a pixel value of each pixel, an average value of pixel values of all pixels included in the first monetary image, a pixel value of each pixel of the reference image And calculating an average value of pixel values of all pixels included in the reference image to calculate a correlation value for each pixel included in the intra-class correlation map.

In the present invention, the correlation map generator may calculate a correlation value between each of the second monetary images and the reference image on a pixel-by-pixel basis, average the correlated values for the calculated pixel on a pixel by pixel basis, And generates a map.

In the present invention, the correlation map generating unit may calculate a correlation map based on a pixel value of each pixel, an average value of pixel values of all pixels included in each second monetary image, a pixel value of each pixel of the reference image And calculating a correlation value for each pixel included in the inter-class correlation map through an operation of calculating an average value of pixel values of all pixels included in the reference image.

In the present invention, each of the second monetary images may be a currency image different from the reference image or a monetary image of another denomination, a monetary image of another facet of the same denomination as the reference image, Directional currency image.

In the present invention, the step of generating the effective map relating to the valid region may include: generating the valid map based on the pixel value of each pixel included in the intra-class correlation map and the pixel value of each pixel included in the inter- Calculating a similarity value between images based on a ratio between pixel values for each pixel to generate the similarity map; And extracting a valid code effective for denomination reflecting the frequency of pixel values of each pixel included in the similarity map.

In the present invention, in extracting the valid code, the valid map generation unit extracts the valid code such that the validity code generator starts from a pixel having a high pixel value and the cumulative value of the frequency number corresponding to each pixel falls within a threshold range .

According to another aspect of the present invention, there is provided an image processing apparatus including an input unit for inputting a plurality of training images for a specific face of a specific currency; A reference image generating unit for generating a reference image by averaging the input training images; An intra-class correlation map showing a correlation between the corresponding first monetary images and the reference image of the same denomination as the reference image, and a correlation between the second monetary images of the kind different from the first monetary images and the reference image A correlation map generating unit for generating a correlation map between classes indicating a relation; And a validity map generator for generating a validity map relating to a valid region on the currency to be identified at the time of denomination, based on the intra-class correlation map and the inter-class correlation map. A valid map generation device is provided.

The method and apparatus for generating effective maps for monetary identification according to an aspect of the present invention not only can alleviate the processing burden on the monetary identification of the monetary identification device in identifying the monies input to the monetary identification device, And an effective map for money identification, which can improve the accuracy of money discrimination. This also improves the accuracy of currency identification, the speed of currency identification and the system burden of identification.

1 is a block diagram illustrating a configuration of an effective map generating apparatus for money identification according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating an effective map generation method for money identification according to an embodiment of the present invention. Referring to FIG.
FIG. 3 is a reference diagram illustrating a process of generating a reference image from a training image in this embodiment.
4 is a reference diagram showing a process of generating an intra-class correlation map from a reference image in the present embodiment.
FIG. 5 is a reference diagram illustrating a process of generating an inter-class correlation map from a reference image in the present embodiment.
FIG. 6 is a flowchart illustrating a step of extracting a binary valid code shown in FIG. 2 to generate a valid map for denomination identification.
FIG. 7 is a reference view showing a process of extracting a binary valid code from a similarity map to generate an effective map for money identification.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and like parts are denoted by similar reference numerals throughout the specification.

Throughout the specification, when an element is referred to as "comprising ", it means that it can include other elements as well, without excluding other elements unless specifically stated otherwise.

FIG. 1 is a block diagram illustrating a configuration of an effective map generating apparatus for money identification according to an embodiment of the present invention. FIG. 2 is a flowchart illustrating an effective map generating method for money identification according to an embodiment of the present invention. FIG. 3 is a reference view showing a process of generating a reference image from a training image in this embodiment. FIG. 4 is a flowchart illustrating a process of generating a correlation map in a class from a reference image in this embodiment. FIG. 5 is a reference view showing a process of generating an interclass correlation map from a reference image in the present embodiment, FIG. 6 is a flowchart illustrating a process of generating a validity map for identifying a currency by extracting the binary- FIG. 7 is a flow chart for explaining a procedure for extracting a binary valid code from a similarity map and generating an effective map for money identification A. An embodiment according to the present invention will be described with reference to FIGS. 1 to 7. FIG.

1, the valid map generating apparatus for money identification according to the present embodiment includes an input unit 110, a reference image generating unit 120, a correlation map generating unit 130, and an effective map generating unit 140 ). ≪ / RTI >

The input unit 110 receives a plurality of training images for a specific face of a specific currency. Here, the plurality of training images may be an image of the same size as the original image of money, or may be an abbreviated image.

The reference image generator 120 generates a reference image by averaging the input training images. Here, the reference image may be generated by averaging pixel values of corresponding pixels of the plurality of training images.

The correlation map generating unit 130 generates an in-class correlation map indicating a correlation between the reference images and corresponding first monetary images of the same denomination as the reference image, Class correlation map indicating a correlation between the second monetary images and the reference image different from those of the images.

The correlation map generating unit 130 calculates a correlation value between each of the first monetary images and the reference image on a pixel by pixel basis and averages the correlation values for the calculated pixel on a pixel by pixel basis, . At this time, the correlation value for each pixel included in the intra-class correlation map is calculated by calculating a pixel value of each pixel with respect to each first monetary image, an average value of pixel values of all pixels included in each first monetary image , A pixel value of each pixel of the reference image, and an average value of pixel values of all pixels included in the reference image, which will be described in more detail below.

On the other hand, the correlation map generator 130 calculates the correlation value between each of the second monetary images and the reference image on a pixel-by-pixel basis, averages the correlation values for the calculated pixel on a pixel-by-pixel basis, Create a dmap. At this time, the correlation value for each pixel included in the inter-class correlation map is calculated by calculating a pixel value of each pixel with respect to each second monetary image, an average value of pixel values of all pixels included in each second monetary image , The pixel value of each pixel of the reference image, and the average value of the pixel values of all the pixels included in the reference image, which will be described in more detail below.

When generating a validity map for the valid region, the validity map generation unit 140 generates a validity map for the validity region by comparing the pixel value of each pixel included in the intra-class correlation map with the pixel value of each pixel included in the inter- The similarity map is generated by calculating the similarity value between images on the basis of each pixel and the validity code valid for the money identification is extracted by reflecting the frequency of pixel values of each pixel included in the similarity map.

The valid map generating unit 140 extracts the valid codes such that the cumulative value of the number of frequencies corresponding to each pixel falls within a critical range, starting from a pixel having a high pixel value.

The operation and operation of the present embodiment thus configured will be described in detail with reference to Figs. 1 to 7. Fig.

As shown in FIG. 2, first, the input unit 110 receives a plurality of training images for a specific face of a specific currency (S210). Here, the image of a particular face of a specific currency refers to a monetary image in a specific direction of one of the denominations of the denominations of various denominations of various countries. That is, the input unit 110 receives a plurality of training images for a specific direction of a specific face of the specific currency. Further, the plurality of training images may be an image of the same size as the original image of money, or may be an abbreviated image.

Subsequently, the reference image generator 120 generates a reference image by averaging the input training images (S220). 3, the reference image generator 120 averages pixel values of corresponding pixels of the plurality of training images to generate a reference image. 3, the average value of the pixel values (for example, the luminance values of 0 to 255) of the pixels at the (0, 0) position of each training image is obtained, and the pixel value of (0.0) I decide. Then, this operation is performed for all the pixels of each training image, so that a reference image is generated.

Next, the correlation map generating unit 130 generates an in-class correlation map indicating a correlation between the reference images and corresponding first monetary images of the same denomination as the reference image. And a second inter-class correlation map indicating a correlation between the second monetary images and the reference images different from the first monetary images (S230).

First, an in-class correlation map is described first. The first monetary image means an image in the same plane and direction of the same denomination as the reference image, that is, an image of the same class (in-class), and a plurality of first monetary images are processed in step S210. And may be a plurality of money images input separately.

As shown in FIG. 4, the correlation map generator 130 calculates a correlation value (CC) between each of the first monetary images and the reference image for each pixel. Here, the correlation value CC is a numerical value indicating the degree of relevance (for example, similarity of pixel values) between pixels of each first monetary image and the reference image, 1. ≪ / RTI > The correlation value calculated by Equation (1) becomes smaller as the degree of correlation between the reference image and each corresponding pixel of the first monetary image (e.g., similarity degree) is higher. Therefore, a pixel having a relatively small correlation value means that the reference image and the first monetary image have a high correlation between the reference image and the corresponding pixel of the first monetary image. Since the reference image and the first monetary image are images of the same class, .

Of course, unlike Equation (1), there may be a correlation value calculation expression of various types and types that can indicate the degree of correlation between pixels. Also, unlike in Equation 1, if the correlation value is relatively large, another correlation value calculation expression indicating that the correlation between the pixels is high can be used.

(Where CC (i, j) is the correlation value at the (i, j) pixel position,

I ₁ (i, j) is the pixel value at the (i, j) pixel position of each first monetary image,

₁는 각 제 1화폐 이미지의 전체 픽셀의 픽셀값의 평균,

₁ denotes an average of pixel values of all pixels of each first monetary image,

R (i, j) is the pixel value at the (i, j) pixel location of the reference image,

은 기준이미지의 전체 픽셀의 픽셀값의 평균)

Is the average of the pixel values of all pixels of the reference image)

The correlation map generating unit 130 generates the correlation map in the class by averaging the calculated correlation values for the corresponding pixel by pixel. For example, when there are 100 first monetary images, the correlation map generator 130 firstly calculates the correlation value between each of the 100 first monetary images and the reference image for each pixel, Correlation values are obtained. Then, a correlation map is generated as shown in FIG. 4 by averaging 100 correlation values per pixel for each pixel and implementing it as a map. That is, in brief, the correlation map generating unit 130 obtains a correlation value between the reference image and the first monetary images of the same class (in-class) and implements the correlation value in the form of a map, . Here, each pixel value of the intra-class correlation map may be the correlation value itself or another pixel value set to correspond to the correlation value.

Next, referring to the inter-class correlation map, the second monetary image means a monetary image different from the reference image (or the first monetary image). Wherein each of the second monetary images is a currency image different from the reference image or a monetary image of another denomination, a monetary image of another facet of the same denomination as the reference image, . ≪ / RTI >

As shown in FIG. 5, the correlation map generator 130 calculates a correlation value (CC) between each of the second monetary images and the reference image for each pixel. Here, the correlation value CC may be expressed as a numerical value of the degree of relevance (for example, similarity of pixel values) for each pixel between each second monetary image and the reference image, And can be calculated by the following equation (2). The correlation value calculated by Equation (2) becomes smaller as the degree of correlation (e.g., similarity) between the corresponding pixels of the reference image and the second monetary image is higher. Therefore, a pixel having a relatively small correlation value means that the relation between the reference image and the corresponding pixel of the second monetary image is high by that much. Since the reference image and the second monetary image are images of different classes, A large number of pixels having a relatively large correlation value are displayed rather than the first monetary image.

Of course, unlike Equation (2), there may be various types of correlation value calculating equations that can express the degrees of correlation between pixels. Further, unlike in Equation (2), if the correlation value is relatively large, another correlation value calculation expression indicating that the correlation between the pixels is high can be used.

(Where CC (i, j) is the correlation value at the (i, j) pixel position,

I ₂ (i, j) is the pixel value at the (i, j) pixel position of each second monetary image,

₂는 각 제 2화폐 이미지의 전체 픽셀의 픽셀값의 평균,

₂ is an average of pixel values of all pixels of each second monetary image,

R (i, j) is the pixel value at the (i, j) pixel location of the reference image,

은 기준이미지의 전체 픽셀의 픽셀값의 평균)

Is the average of the pixel values of all pixels of the reference image)

The correlation map generating unit 130 averages the calculated correlation values for the corresponding pixels on a pixel-by-pixel basis to generate the inter-class correlation map. For example, when there are 100 second monetary images, the correlation map generator 130 firstly calculates the correlation value between each of the 100 second monetary images and the reference image for each pixel, Correlation values are obtained. Then, 100 correlation values for each pixel are averaged for each pixel and are implemented as a map to generate a class-to-class correlation map as shown in FIG. That is, in brief, the correlation map generator 130 obtains a correlation value between the reference image and the second monetary images of different classes, and implements it in the form of a map, . Here, each pixel value of the inter-class correlation map may be the correlation value itself or another pixel value set to correspond to the correlation value.

Next, the validity map generation unit 140 generates a validity map for the effective region on the currency to be identified at the time of denomination, based on the intra-class correlation map and the inter-class correlation map (S240). To be more specific, the validity map generation unit 140 generates a validity map based on the ratio between the pixel value of each pixel included in the intra-class correlation map and the pixel value of each pixel included in the inter-class correlation map The similarity value between images is calculated for each pixel to generate a similarity map (S241). Here, the similarity value of each pixel constituting the similarity map can be obtained by the following equation (3).

As described above, if the degree of correlation (similarity) per pixel of an image to be compared is high, the correlation value is small, and if it is low, it is calculated large. As will be described below, the validity map to be used for identification of a currency is a map for defining an area of a monetary image to be used in actual currency discrimination. Therefore, the similarity map on which this effective map is generated has a small correlation value (i.e., high pixel-to-pixel correlation) between images of the same class (in-class) as the reference image, It is important to effectively define and describe pixels that exhibit large correlation values (i.e., low correlation between pixels).

Therefore, if the similarity value obtained by Equation (3) is relatively large, the pixel can be said to be an effective pixel which can be usefully used for money identification (especially denomination identification), and vice versa, Which is not very useful for identification of denomination. On the other hand, each pixel value of the similarity map may be the similarity value itself or another pixel value set to correspond to the similarity value. Through this process, a similarity map as shown in FIG. 7 is generated.

Then, the validity map generator 140 reflects the frequency of the pixel values of each pixel included in the similarity map, and extracts a valid code effective for denomination (S242).

6 and 7, first, the validity map generator 140 performs a histogram operation (histogram generation) showing the relationship between the pixel value of each pixel included in the similarity map and the frequency of the pixel value (S61). The horizontal axis of the histogram of FIG. 7 is the pixel value, and the vertical axis is the frequency at which the pixel value appears.

Then, the validity map generator 140 calculates a threshold pixel value as shown in FIG. 7 (S62). This threshold pixel value means a pixel value starting from a pixel having a high pixel value in the similarity map and causing the cumulative value of the frequency corresponding to each pixel to belong to a certain threshold range, It is the standard for judgment.

Then, the validity map generator 140 performs binarization on the basis of the threshold pixel value by setting a pixel having a larger pixel value (i.e., similarity) to 1 and a pixel having a lower pixel value to 0 (S63). As shown in FIG. 7, finally, the validity map generator 140 generates an effective code from the information obtained through the binarization of the step S63 (S64). The generated effective map may be a valid map corresponding to the reduced image or a valid map corresponding to the same size as the actual money in order to reduce the system burden according to the identification capacity. The code of the portion corresponding to the valid pixel in the valid map may be set to 0, and the code of the portion corresponding to the non-valid pixel may be set to 1.

The money identification device performs the money identification operation using the generated effective map. At this time, the money discriminating apparatus can perform currency discrimination using subspace analysis such as LDA (Linear Discriminant Analysis), PCA (Principal Component Ananlysis) and LFA (Local Feature Analysis), or various other image analysis tools You may.

As described above, the effective map generating method and apparatus for money identification according to the present embodiment not only can reduce the processing burden on the money identification of the money identification apparatus in identifying the money to be input to the money identification apparatus, It is possible to generate a reference image for monetary identification, which makes it possible to improve the monetary identification processing speed and the currency discrimination accuracy.

While the invention has been shown and described in detail in the foregoing description, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art, Of the right.

110: input unit
120: Reference image generation unit
130: correlation map generating unit
140: valid map generating unit

Claims (20)

The input unit receiving a plurality of training images for a specific face of a specific currency;
A reference image generating unit generates a reference image by averaging the input training images;
A correlation map generating section generates an intra-class correlation map showing a correlation between the corresponding first monetary images and the reference image of the same denomination as the reference image, and a correlation image in a class different from the first monetary images, Generating an inter-class correlation map indicating a correlation between the reference image and the reference image; And
Wherein the effective map generating unit includes a step of generating a valid map relating to a valid region on the currency to be identified at the time of denomination, based on the correlation map in the class and the correlation map between the classes, A valid map generation method for identification.
The method according to claim 1,
Wherein the reference image is generated by averaging pixel values of corresponding pixels of the plurality of training images.
The method according to claim 1,
The correlation map generator may calculate a correlation value between each of the first monetary images and the reference image on a pixel by pixel basis and averages the calculated correlation values for the corresponding pixel on a pixel by pixel basis to generate a correlation map in the class Wherein the validity map generation means generates the validity map for currency identification.
The method of claim 3,
Wherein the correlation map generating unit generates the correlation map based on the pixel value of each pixel of the first monetary image, the average value of the pixel values of all the pixels included in the first monetary image, the pixel value of each pixel of the reference image, Calculating a correlation value for each pixel included in the correlation map in the class through an operation between an average value of pixel values of all the pixels included in the image.
The method according to claim 1,
The correlation map generating unit may calculate a correlation value between each of the second monetary images and the reference image for each pixel and average the correlated values for the calculated pixel for each pixel to generate the inter- Wherein the validity map generation means generates the validity map for currency identification.
6. The method of claim 5,
Wherein the correlation map generating unit generates the correlation map based on the pixel value of each pixel, the average value of the pixel values of all the pixels included in the second monetary image, the pixel value of each pixel of the reference image, Calculating a correlation value for each pixel included in the inter-class correlation map through an operation between an average value of pixel values of all pixels included in the image.
The method according to claim 1,
Wherein each of the second monetary images is a currency image different from the reference image or a monetary image of another denomination, a monetary image of another facet of the same denomination as the reference image, Wherein the validity map generation means generates the validity map for currency identification.
The method according to claim 1,
The correlation map generator may calculate a correlation value between each of the first monetary images and the reference image on a pixel by pixel basis and averages the correlation values for the calculated pixel on a pixel by pixel basis to generate a correlation map in the class ,
The correlation map generating unit may calculate a correlation value between each of the second monetary images and the reference image for each pixel and average the correlated values for the calculated pixel for each pixel to generate the inter- Wherein the validity map generation means generates the validity map for currency identification.
9. The method of claim 8,
Wherein the generating of the validity map for the valid region comprises:
The effective map generating unit calculates the similarity value between images based on the ratio between the pixel value of each pixel included in the intra-class correlation map and the pixel value of each pixel included in the inter-class correlation map Generating a similarity map; And
And extracting an effective code effective for denomination reflecting the frequency of pixel values of each pixel included in the similarity map.
10. The method of claim 9,
In extracting the valid code,
Wherein the valid map generating unit extracts the valid code starting from a pixel having a high pixel value and storing the valid code so that the cumulative value of the frequency corresponding to each pixel falls within a threshold range.
An input unit for receiving a plurality of training images for a specific face of a specific currency;
A reference image generating unit for generating a reference image by averaging the input training images;
An intra-class correlation map showing a correlation between the corresponding first monetary images and the reference image of the same denomination as the reference image, and a correlation between the second monetary images of the kind different from the first monetary images and the reference image A correlation map generating unit for generating a correlation map between classes indicating a relation; And
And a validity map generation unit for generating a validity map relating to a valid region on the currency to be identified at the time of denomination, based on the intra-class correlation map and the inter-class correlation map. Effective map generator.
12. The method of claim 11,
Wherein the reference image is generated by averaging pixel values of corresponding pixels of the plurality of training images.
12. The method of claim 11,
The correlation map generator may calculate a correlation value between each of the first monetary images and the reference image on a pixel by pixel basis and averages the calculated correlation values for the corresponding pixel on a pixel by pixel basis to generate a correlation map in the class Wherein the validity map generating unit is configured to generate a validity map for money identification.
14. The method of claim 13,
Wherein the correlation map generating unit generates the correlation map based on the pixel value of each pixel of the first monetary image, the average value of the pixel values of all the pixels included in the first monetary image, the pixel value of each pixel of the reference image, And calculating a correlation value for each pixel included in the correlation map in the class through an operation between an average value of pixel values of all pixels included in the image.
12. The method of claim 11,
The correlation map generating unit may calculate a correlation value between each of the second monetary images and the reference image for each pixel and average the correlated values for the calculated pixel for each pixel to generate the inter- Wherein the validity map generating unit is configured to generate a validity map for money identification.
16. The method of claim 15,
Wherein the correlation map generating unit generates the correlation map based on the pixel value of each pixel, the average value of the pixel values of all the pixels included in the second monetary image, the pixel value of each pixel of the reference image, And calculating a correlation value for each pixel included in the inter-class correlation map through an operation between an average value of pixel values of all pixels included in the image.
12. The method of claim 11,
Wherein each of the second monetary images is a currency image different from the reference image or a monetary image of another denomination, a monetary image of another facet of the same denomination as the reference image, Wherein the validity map generating unit is configured to generate the validity map for money identification.
12. The method of claim 11,
The correlation map generator may calculate a correlation value between each of the first monetary images and the reference image on a pixel by pixel basis and averages the correlation values for the calculated pixel on a pixel by pixel basis to generate a correlation map in the class ,
The correlation map generating unit may calculate a correlation value between each of the second monetary images and the reference image for each pixel and average the correlated values for the calculated pixel for each pixel to generate the inter- Wherein the validity map generating unit is configured to generate a validity map for money identification.
19. The method of claim 18,
Upon creation of the validity map for the validity zone,
The validity map generator calculates the similarity value between images based on the ratio between the pixel value of each pixel included in the intra-class correlation map and the pixel value of each pixel included in the inter-class correlation map A similarity map is generated,
Wherein validity code valid for money identification is extracted by reflecting the frequency of pixel values of each pixel included in the similarity map.
20. The method of claim 19,
Upon extracting the valid code,
Wherein the effective map generating unit extracts the valid code starting from a pixel having a high pixel value and accumulating the cumulative value of the frequency corresponding to each pixel in a threshold range.

Images