JPH10111965A - Method and device for deciding disfeatured state of body - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately detect the disfeatured state of a body by finding the density distribution of image data from total image data on the body and deciding the disfeatured state of the body. SOLUTION: An image input part 21 obtains reflected light on the entire surface of a coin 15 as image data by a coin detection sensor 17 and an A/D conversion part 22 converts the image data from analog to digital (multi-valued). A density histogram generation part 23 calculates pixel quantity distribution by density (gradation value) from the image data and a binarization part 24 converts the A/D-converted image data from the multi-valued data to binary data to decide the kind of the money. A discrimination part 25 decides the kind of the money by comparing the binarized data with reference density distributions corresponding to the heads and tails by the respective money kinds. A disfeaturing level decision part 26 decides the disfeatured state of the coin by comparing the pixel quantity distributions by density calculated by the density histogram generation part 234 with previously set disfeaturing decision levels by the money kinds.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for discriminating a stained state of an object, such as a method incorporated in various types of automatic teller machines for handling objects such as coins, and more particularly to an object for accurately discriminating the stained state of an object. The present invention relates to a method and apparatus for determining the state of fouling.

[0002]

2. Description of the Related Art In the following, a description will be given of an example of an apparatus for determining the dirty state of a coin. For example, as shown in Japanese Patent Application Laid-Open No. 61-18844, several points of a coin are detected at a detection time interval determined by using an optical fiber. The amount of reflected light is acquired, and the acquired reflected light amount is compared with a preset contamination determination level to locally detect the dirty state of the coin.

[0003]

However, in this case, since the dirty state of the coin is locally detected, the dirty state of the entire coin cannot be accurately detected, and uneven detection is likely to occur, resulting in inaccurate determination. Therefore, there has been a problem that the reliability of detection accuracy is poor, such as inducing erroneous determination.

Accordingly, the present invention provides a method for discriminating a stained state of an object, which detects the stained state of an object by detecting the whole state of the object uniformly, not locally, so that the stained state of the object can be accurately detected. And the provision of such a device.

[0005]

According to the first aspect of the present invention,
The entire image data of the object is obtained by the image data obtaining means, the density distribution of the image data is obtained from the obtained image data, and the contamination state of the object is determined from the obtained density distribution. There is a feature.

According to a second aspect of the present invention, the entire image data of the object is obtained by the image data obtaining means, the density distribution is calculated from the obtained image data, and the calculated density distribution and the type of the object are calculated. And a method of discriminating the contamination state of the object by comparing the reference density distribution stored according to the type of the identified object with the reference density distribution.

According to a third aspect of the present invention, there is provided an image data obtaining means for obtaining overall image data of an object, a density distribution calculating means for calculating a density distribution from the image data obtained by the image data obtaining means, Reference value setting means for setting a reference value to determine contamination, and determination means for determining the contamination state of the object from the density distribution calculated by the density distribution calculation means and the reference value set by the reference value setting means. Characterized in that it is an object contamination state determination device provided with

According to a fourth aspect of the present invention, there is provided an image data obtaining means for obtaining overall image data of an object, a density distribution calculating means for calculating a density distribution from the image data obtained by the image data obtaining means, Identification means for identifying the type of the object, storage means for pre-storing the reference density distribution for each of the plurality of types of objects, and the density distribution calculated by the density distribution calculation means and the type of the object identified by the identification means An object contamination state discriminating apparatus comprising: a discrimination means for discriminating a contamination state of an object by comparing the reference density distribution of the object stored in the storage means.

[0009]

According to the present invention, when determining the contamination state of an object, first, the entire image data of the object is obtained by the image data obtaining means, and the entire image data of the object is obtained from the obtained image data. A density distribution is obtained, and the contamination state of the object is determined from the obtained density distribution.

[0010] Further, when determining the contamination state of the object, the entire image data of the object is obtained by the image data obtaining means, and the density distribution is calculated from the obtained image data. The type of the object is identified, and the reference density distribution stored in accordance with the type of the identified object is compared with the reference density distribution to determine the contamination state of the object.

[0011]

As a result, it is possible to determine the overall contamination state of the object from the density distribution of the image data that captures the entire object.
Uniform and accurate discrimination can be made. In particular, since the detection is made by detecting the whole rather than the local detection, it is possible to perform the contamination determination close to human sensitivity. Therefore, erroneous determination can be naturally eliminated, and highly reliable stain determination can be performed.

[0012] Further, even in the case of discriminating the stained state of an object such as a coin which is frequently handled, if the discrimination is made by comparing it with a reference value set for a denomination, the coin is not stained and becomes dirty or damaged. It can be accurately distinguished from a damaged coin. In addition, when this highly reliable dirty state determination device is incorporated in an automatic transaction machine, a dirty coin and a normal coin are accurately selected, and the circulation of the dirty coin is regulated, so that only the normal coin can be reliably recycled.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below in detail with reference to the drawings. [First Embodiment] FIG. 1 shows a coin detecting structure of a dirty state discriminating device 11 for a coin.
A coin transport path 16 for pinching and transporting coins 15 in the horizontal direction is provided by combining transport members such as 2, 13 and feed rollers 14. The coin 15 guided here is discriminated by the denomination and the stained state.

In this case, the coin detecting sensor 17 irradiates the entire upper surface of the coin 15 conveyed in a horizontal state with the irradiation light 17b of the LED 17a arranged in a ring shape, and reflects the reflected light 17c of the upper surface uneven pattern of the coin on the CCD camera. At step 17d, image data representing the surface pattern of the coin and its dirty state is acquired. This coin detection sensor 17 instantaneously reads the coin being conveyed, which has been guided here, and the read coin is conveyed to the subsequent stage as it is.

FIG. 2 is a circuit diagram of the apparatus 11 for judging the dirty state of a coin, which includes an image input unit 21, an A / D conversion unit 22, a density histogram creation unit 23, and a binarization unit 24.
, An identification unit 25, and a contamination level determination unit 26.

The image input unit 21 acquires the reflected light of the entire surface of the coin 15 as image data by the CCD camera 17d of the coin detection sensor 17, and converts the acquired image data to A / A
The D converter 22 converts the signal from analog to digital (multi-value).

The density distribution and the denomination are obtained from the image data obtained through the A / D converter 22.
One density histogram creation unit 23 calculates a pixel number distribution for each density (gradation value) from the A / D-converted image data, and the other binarization unit 24 converts the A / D-converted image data. The multi-value is converted into a binary value for denomination determination.

The discriminator 25 stores a reference density distribution corresponding to the front and back of each denomination, and determines the denomination by comparing the reference density distribution with the binarized data.

The contamination level determination unit 26 stores a contamination determination level for each denomination set in advance in order to determine the contamination. The contamination determination level and the density for each density calculated by the density histogram creating unit 23 are stored. The coins are compared with the pixel number distribution to determine the dirty state of the coin.

FIG. 3 shows image data 31 of a beautiful normal coin suitable for circulation without contamination, and is represented by image data of, for example, 256 × 256 pixels (256 gradations / one pixel). The uneven pattern of the coin is obtained relatively clearly.

FIG. 4 shows an example of a density histogram 41 obtained by calculating a pixel number distribution for each density (tone value) from image data of a normal coin. The density histogram 41 obtained from this normal coin has a high tone value. The density position (gradation value = 105) at which the area ratio becomes a certain ratio value (about 15%) of the entire area is calculated from the density value (darker one), and this is set to a preset contamination determination level (reference gradation value = 64). If the gradation value calculated from the reference gradation value is larger than the reference gradation value, it is determined that there is no stain,
If it is smaller, it is determined that there is contamination. In the case of this normal coin, it is determined that there is no contamination because it is sufficiently large (64 <105).

FIG. 5 shows image data 51 of a dirty coin that is unsuitable for distribution. In the case of this dirty coin, the uneven pattern of the coin is obtained with low reflection and unclear, and the regular uneven pattern cannot be sufficiently obtained.

FIG. 6 shows an example of a density histogram 61 obtained by calculating the number-of-pixels distribution for each density (gradation value) from the image data of the dirty coin. The density position (gradation value = 1) where the area ratio is 15% of the whole is calculated from the higher value,
This and a preset contamination determination level (reference gradation value = 6)
4) is compared to determine the presence or absence of contamination. In the case of this dirty coin, it is determined that there is contamination since it is considerably small (64> 1).

The overall processing operation of the coin discrimination state determining device 11 thus constructed will be described with reference to the flowchart of FIG. Now, when the coin 15 is conveyed to the coin conveyance path 16 and guided to the position of the coin detection sensor 17, the reflected light irradiated on the coin 15 is received by the CCD camera, and the image data of the entire upper surface of the coin is imaged. It is obtained by the input unit 21 (step n1).

The obtained image data is converted from analog to digital (multi-valued) by the A / D converter 22, and then output to both the density histogram generator 23 and the binarizer 24 (step n2). ), The density histogram creation unit 23 creates a pixel number distribution for each density (tone value) from the A / D converted image data (step n3), while the binarization unit 24 performs A / D conversion. The converted image data is converted from multi-valued data to binary data for the denomination determination, and the authenticity and denomination of coins are identified by the identification unit 25 based on the converted binary data (steps n4 to n5).

If the coin is discriminated as a true coin based on the discrimination result, the stain level judging section 26 judges whether the coin is dirty or not, and judges whether the coin is valid or inappropriate as a circulation coin. On the other hand, if it is a fake coin, fake coin processing such as return and collection is performed (steps n6 to n7).

Next, a description will be given, with reference to the flowchart of FIG. When the stain level determination unit 26 determines the dirty state of the coin 15, the stain level determination unit 26 first obtains denomination data of the same coin recognized as a genuine coin from the identification unit 25 (step n 11). .

Subsequently, a tone value having an area ratio of 15% is calculated from the density histogram creating section 23 (step n12).
The calculated tone value is discriminated by comparing it with a contamination determination level set in advance for each denomination and for each side (step n13).

At this time, if it is determined that the coin has no stain,
The coin is determined as a coin suitable for distribution use (step n
14 to n15).

On the other hand, if it is determined that the coin is a dirty coin, the coin is dirty or damaged and is unsuitable for recycling.

[Second Embodiment] FIG. 9 shows a coin detecting structure of a coin soiling state discriminating device 91. This coin detecting structure is horizontal by combining conveying members such as conveying belts 92, 93, feed rollers 94, etc. A coin transport path 96 for pinching and transporting the coin 95 in the direction is provided, and a coin detection sensor 97 and a magnetic detection sensor 98 are arranged in this order along the coin transport path 96,
The denomination state of the coin 95 guided here and the denomination are determined.

In this case, the reflected light emitted by the annularly arranged LEDs 97a of the coin detection sensor 97 is reflected by the CCD camera 9
7b receives the image and obtains image data representing the surface pattern of the coin 95 and the stained state thereof, and acquires magnetic data of the coin for obtaining the outer diameter and material of the coin by the magnetic detection sensor 98 at the subsequent stage.

FIG. 10 shows a circuit diagram of a coin contamination state discriminating apparatus 91, which comprises an image input unit 101 and an image A /
A D conversion unit 102, a density histogram creation unit 103,
It comprises a magnetic input unit 104, a magnetic A / D converter 105, an identification unit 106, and a contamination level determination unit 107.

The image input unit 101 includes a coin detection sensor 97
The reflected light of the entire surface of the coin 95 is acquired as image data by the CCD camera 97b, and the acquired image data is converted from analog to digital (multi-valued) by the image A / D converter 102.

The density histogram creation unit 103 performs A / D
A pixel number distribution for each density (gradation value) is calculated from the converted image data.

On the other hand, the magnetic input unit 104 obtains magnetic data of the outer diameter and the material of the coin 95 by the magnetic detection sensor 98 and converts the obtained magnetic data into a magnetic A / D converter 105.
To convert from analog to digital (multi-valued) for denomination determination.

The identification section 106 stores reference magnetic data corresponding to the outer diameter and material of each denomination, and compares the reference magnetic data with the magnetic data detected by the magnetic detection sensor 98 to compare the denomination. judge.

The contamination level determination unit 107 stores a contamination determination level for each denomination set in advance to determine the contamination, and determines the contamination determination level and the pixel for each density calculated by the density histogram creating unit 103. Compare with the number distribution,
The dirty state of the coin 95 is determined. When the magnetic detection sensor 98 for determining the denomination of a coin is provided as described above, the denomination determination can be further narrowed based on the detection information of this sensor, and the identification accuracy of the denomination determination can be increased.

The processing operation of the coin discrimination state determining device 91 thus configured will be described with reference to the flowchart of FIG. Now, when the coin 95 is conveyed to the coin conveyance path 96 and guided to the position of the coin detection sensor 97, the reflected light illuminated on the coin 95 is received by a CCD camera, and image data of the entire upper surface of the coin is imaged. It is acquired by the input unit 101 (step n21).

The obtained image data is converted from analog to digital (multi-valued) by the image A / D converter 102, and then output to the density histogram generator 103 (step n22). A / D
A pixel number distribution for each density (gradation value) is created from the converted image data (step n23), and the created pixel number distribution is used as a stain state determination element as a stain level determination unit 107.
(Step n24).

On the other hand, the coin 9 via the magnetic detection sensor 98
The magnetic input unit 104 having acquired the magnetic data of No. 5 converts the magnetic data from analog to digital (multi-valued) by the magnetic A / D converter 105 (steps n25 to n26).
From the magnetic data, the discrimination unit 106 discriminates the authenticity and denomination of the coin (step n27).

If the coin is discriminated as a true coin based on the discrimination result, the stain level judging section 107 judges whether the coin is dirty or not, and judges whether the coin is suitable or not as a circulation coin. On the other hand, if it is a fake coin, fake coin processing such as return and collection is performed (step n28).

As described above, since the contamination state of the entire coin can be determined from the density distribution of the image data that captures the entire coin, detection unevenness is eliminated when the coin is detected, and uniform and accurate determination of the contamination can be performed. In particular, since the detection is performed not by the local detection but by the entire detection, the contamination determination close to the human sensitivity can be performed. Therefore, erroneous determination can be reliably eliminated,
In addition, highly reliable contamination determination can be performed. In addition, when determining the dirty state of coins with many types of handling, by comparing with a reference gradation value set for each denomination, it is possible to accurately distinguish normal coins without stains from dirty or damaged dirty coins. be able to. In addition, when this highly-reliable defacement state determination device is incorporated in an automatic transaction machine, it is necessary to accurately sort out defaced coins and normal coins, regulate the distribution of defaced coins, and properly recycle only normal coins. Can be.

In correspondence between the present invention and the configuration of the above-described embodiment, the apparatus for determining the soiled state of an object according to the present invention corresponds to the apparatus 11 or 91 for determining the dirty state of coins of the embodiment. Corresponds to the coins 15 and 95, the image data acquisition means corresponds to the coin detection sensors 17 and 97 and the image input sections 21 and 101, the density distribution calculation means corresponds to the density histogram creation sections 23 and 103, The reference value corresponds to a reference gradation value, and the reference value setting means and the identification means
Although the identification means and the storage means correspond to the identification units 25 and 106 and the stain level determination units 26 and 107, the present invention is not limited to the configuration of the above-described embodiment.

[Brief description of the drawings]

FIG. 1 is a side view showing a coin detecting structure of a coin soiling state discriminating apparatus according to a first embodiment of the present invention.

FIG. 2 is a circuit configuration diagram of the coin dirty state determining apparatus according to the first embodiment of the present invention.

FIG. 3 is an image data diagram of a normal coin having no stain according to the first embodiment of the present invention.

FIG. 4 is a table showing an example of a density histogram of normal coins according to the first embodiment of the present invention.

FIG. 5 is an image data diagram of a dirty coin according to the first embodiment of the present invention.

FIG. 6 is a chart showing an example of a density histogram of a dirty coin according to the first embodiment of the present invention.

FIG. 7 is a flowchart showing the overall processing operation of the coin defacement state determination device according to the first embodiment of the present invention.

FIG. 8 is a flowchart showing a stain discrimination processing operation of the coin stain condition discriminating apparatus according to the first embodiment of the present invention.

FIG. 9 is a side view showing a coin detecting structure of the coin soiling state discriminating apparatus according to the second embodiment of the present invention.

FIG. 10 is a circuit configuration diagram of a coin dirty state determination device according to a second embodiment of the present invention.

FIG. 11 is a flowchart showing a processing operation of the coin contamination state determination device according to the second embodiment of the present invention.

[Explanation of symbols]

 11, 91: coin stain state determination device 15, 95: coin 17, 97 ... coin detection sensor 21, 101 ... image input unit 23, 103 ... density histogram creation unit 25, 106 ... identification unit 26, 107 ... stain level Judgment unit 31 ... Normal coin image data 41 ... Normal coin density histogram 51 ... Soiled coin image data 61 ... Soiled coin density histogram 98 ... Magnetic detection sensor 104 ... Magnetic input unit

Claims (4)

[Claims] An image data acquiring unit acquires overall image data of an object, obtains a density distribution of the image data from the obtained image data, and determines a contamination state of the object from the obtained density distribution. A method for determining the state of contamination of a featured object. 2. An image data acquiring means for acquiring overall image data of an object, calculating a density distribution from the acquired image data, identifying the calculated density distribution and a type of the object, and identifying the object. And a method of determining a soiled state of an object by comparing the stored reference density distribution according to the type of the object. 3. An image data obtaining means for obtaining overall image data of an object, a density distribution calculating means for calculating a density distribution from the image data obtained by the image data obtaining means, and a method for determining contamination. A reference value setting unit that sets a reference value; and a determination unit configured to determine a contamination state of the object from the density distribution calculated by the density distribution calculation unit and the reference value set by the reference value setting unit. State determination device. 4. An image data obtaining means for obtaining overall image data of an object, a density distribution calculating means for calculating a density distribution from the image data obtained by the image data obtaining means, and identifying a type of the object. Identification means, storage means for previously storing a reference density distribution for each of a plurality of types of objects, and storage means for storing the density distribution calculated by the density distribution calculation means and the type of the object identified by the identification means An object contamination state discriminating device comprising: a discrimination means for comparing a reference density distribution of the object with the object to determine the contamination state of the object.