JP2019185385A - Recognition processing apparatus, recognition processing method, and recognition processing program - Google Patents

Abstract

To provide a technique for improving the recognition accuracy of a subject.SOLUTION: The recognition processing apparatus includes an acquisition unit, a detection unit, and a determination unit. The acquisition unit acquires a captured image of a region including an object to be placed in a placement position. When the object is placed in the placement position, a size of the object is a first size, the detection unit extracts a feature of a background, and if the size of the object is a second size, which is larger than the first size, the detection unit detects a first feature of a first area included in the captured image in which the feature of the object is extracted. When the object is placed at the placement position, if the first feature indicates the feature of the background, the determination unit determines that the size of the object is the first size, and if the first characteristic indicates the feature of the object, the determination unit determines that the size of the object is the second size.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technique for recognizing an object.

When using a beverage vending machine using a cup, the user purchases a cup corresponding to a desired beverage, and places the purchased cup along a guide of the vending machine, thereby placing the cup at the placement position. Put. And a user pushes down the selection button of the drink corresponding to the purchased cup from the buttons for drink selection which a vending machine has. Thereby, a user receives supply of a desired drink to the purchased cup. In the following description, a beverage vending machine using a cup is also referred to as a beverage supply device. The beverage selection button is also simply referred to as a selection button.
However, when selecting a beverage selection button corresponding to a purchased cup from among selection buttons of the beverage supply device, the user may erroneously select a beverage selection button that does not correspond to the purchased cup. In order to cope with this, a technology in which the vending machine recognizes the type of cup placed at the placement position and automatically supplies beverages corresponding to the purchased cup according to the recognized cup type. It has been known.

  As a related technique, the image registration unit registers the image of the beverage container photographed by the photographing unit in the storage unit in advance. Thereafter, when the purchaser purchases a beverage container and photographs it with the photographing unit, the container recognition unit compares the image information registered in advance in the storage unit with the photographed image information to determine the beverage container. A technique is known in which the display unit displays the beverage menu and the beverage size corresponding to the beverage container determined by the container recognition unit, and the beverage discharge unit discharges the beverage to the beverage container based on the purchaser's operation (for example, Patent Document 1).

  Another related technology is a method for managing orders and supplying goods. Including requesting the user to select a product type at the order entry station, printing a display indicating at least the selected product type on the container, reading the display at the product supply station, and supplying the selected product type to the container. A technique is known (for example, Patent Document 2).

As another related technology, the cup sensor is a box-type beverage vending machine with a front opening that is shaped like a box, a front door (not shown) is provided, and there is a My Cup on the cup stage. Is detected. A cup upper end sensor detects the upper end part of my cup. The cup diameter sensor detects the width of the my cup. The stage position sensor detects the stop position of the cup stage. The stage lifting device moves the cup stage 11 up and down. And the technique which raises / lowers a cup stage according to the height dimension of the my cup mounted on the cup stage by the user, and stops the upper end of my cup in a predetermined position is known (for example, patent documents). 3).

JP 2017-159941 A JP2013-540294A JP 2007-128379 A

In the recognition technology described above, when recognizing an object, it recognizes an object such as a plastic cup with ice that looks different in terms of how ice enters, in order to compare the pre-registered image with the captured image of the object. Accuracy may be reduced.
One aspect of the present invention provides a technique for improving the recognition accuracy of an object.

  One of the recognition processing devices disclosed in this specification is a recognition processing device including an acquisition unit, a detection unit, and a determination unit. An acquisition part acquires the picked-up image of the area | region containing the object mounted in a mounting position. When the target is placed at the placement position, the detection unit extracts a background feature if the target size is the first size, and if the target size is the second size larger than the first size, the target The first characteristic of the first region included in the photographed image from which the feature is extracted is detected. The determination unit determines that the size of the target is the first size when the first characteristic indicates the background feature when the target is placed at the mounting position, and the target when the first characteristic indicates the feature of the target Is determined as the second size.

  According to one embodiment, the recognition accuracy of an object can be improved.

It is FIG. (1) which shows one Example of a drink supply apparatus. It is FIG. (2) which shows one Example of a drink supply apparatus. It is FIG. (3) which shows one Example of a drink supply apparatus. It is FIG. (4) which shows one Example of a drink supply apparatus. It is FIG. (5) which shows one Example of a drink supply apparatus. It is a functional block diagram which shows one Example of a recognition processing apparatus. It is a figure (the 1) which shows the frequency distribution of the brightness | luminance of a detection area. FIG. 6 is a diagram (part 1) illustrating a hue frequency distribution in a detection region. It is a figure which shows an example of type information. It is a figure which shows an example of size information. It is a flowchart (the 1) which shows an example of recognition processing. It is a flowchart (the 2) which shows an example of a recognition process. It is a block diagram which shows one Example of a computer apparatus. It is FIG. (6) which shows one Example of a drink supply apparatus. It is a figure explaining the recognition process of the cup in which the character was described. It is FIG. (7) which shows one Example of a drink supply apparatus. FIG. 10 is a second diagram illustrating a frequency distribution of luminance in a detection region; FIG. 6 is a diagram (part 2) illustrating a hue frequency distribution in a detection region. It is FIG. (8) which shows one Example of a drink supply apparatus.

[Embodiment 1]
The recognition processing apparatus of Embodiment 1 is demonstrated.
The recognition processing apparatus of Embodiment 1 recognizes the kind and size of the beverage container placed on the placement table of the beverage supply device. And a recognition processing apparatus performs the process of supplying the drink of a kind and quantity according to the kind and size of the recognized drink container from a drink supply apparatus to a drink container.
In the following description, as an example, a process for recognizing a cup-shaped beverage container into which coffee is poured will be described. However, the recognition processing device of the present invention is not limited to a beverage container, and at least one of a type and a size is different. Can be used for the recognition process.

  1-5 is a figure which shows one Example of a drink supply apparatus. With reference to FIGS. 1-5, the outline | summary of the process in which the drink supply apparatus 100 provided with the recognition processing apparatus 1 of this invention recognizes at least one of the kind and size of a drink container is demonstrated.

  In the following description, it is assumed that there are two types of beverage containers: a transparent plastic cup with ice for iced coffee and a white paper cup for hot coffee. The two types of cups have two sizes, a regular size (first size) and a large size (second size) larger than the regular size. In the following description, the transparent plastic cup with ice for iced coffee is also simply referred to as a plastic cup. A white paper cup for hot coffee is also simply called a paper cup.

  The beverage container 3a shown in FIG. 2 is a regular size plastic cup. A beverage container 3b shown in FIG. 3 is a large plastic cup. The beverage container 3c shown in FIG. 4 is a regular size paper cup. The beverage container 3d shown in FIG. 5 is a large-size paper cup. In the following description, the beverage containers 3a to 3d are also referred to as the beverage containers 3 unless particularly distinguished.

A description will be given with reference to FIG.
The beverage supply device 100 includes a display device 110, selection buttons 121 to 124, an opening 130, a mounting table 141, a guide 142, a supply port 150, an auxiliary member 161, a photographing hole 170, and a recognition processing device. 1 and the imaging device 2. The selection buttons 121 to 124 are also referred to as selection buttons 120 unless particularly distinguished.

  The display device 110 displays various types of information such as advertisements, advertisements, and operation screens using images and moving images. The display device 110 may include a touch panel, and may accept adjustment of the amount of sugar and milk added to the coffee, selection of coffee strength, and the like.

  The selection button 120 lights a button corresponding to the type and size of the beverage container 3 recognized by the recognition processing device 1, and notifies the user of the type and size of the beverage container 3 placed on the placement table 141. The selection button 121 corresponds to regular size hot coffee RH. The selection button 122 corresponds to large-sized hot coffee LH. The selection button 123 corresponds to the regular size ice coffee RC. The selection button 124 corresponds to the large size ice coffee LC.

  The selection button 120 includes a touch panel, and may accept the type and size of the placed beverage container 3 when a user presses a desired button. In this case, the recognition processing device 1 causes the beverage supply device 100 to supply the beverage of the type and amount corresponding to the type and size selected by the user. In the following description, it is assumed that the selection button 120 executes only notification of the type and size of the beverage container 3 and does not execute reception of the type and size of the beverage container 3.

The opening 130 is a space provided on the front surface of the beverage supply apparatus 100, and a mounting table 141 on which the user places the beverage container 3 is provided.
The mounting table 141 is a black table on which the user mounts the beverage container 3. Note that the color of the mounting table 141 is not limited to black, and for example, the recognition processing device 1 uses the characteristics of the beverage container 3 and the characteristics of the mounting table 141 as the characteristics of at least one of luminance and hue histograms. Any color can be used as long as it can be identified. The histogram shows the luminance or hue characteristics of the beverage container 3 and the mounting table 141 included in the captured image captured by the imaging device 2. In the following description, the characteristics of the beverage container 3, the characteristics of the mounting table 141, and the background characteristics are assumed to be characteristics of a histogram corresponding to each target image. In the following description, the histogram is also simply referred to as a characteristic. Further, when the luminance and hue histograms are not distinguished, they are also simply referred to as histograms.

  The guide 142 is provided on the mounting table 141 in a U-shape, for example, and guides the beverage container 3 to be placed at the mounting position surrounded by the guide 142. Thereby, the user can place the beverage container 3 at the placement position by placing the beverage container 3 on the placement table 141 along the guide 142. In the following description, the mounting position surrounded by the guide 142 is also simply referred to as a mounting position.

  The supply port 150 is a spout used when the beverage supply device 100 supplies a beverage to the beverage container 3. That is, the beverage supply device 100 supplies the beverage container 3 with a type and amount of beverage corresponding to the type and size of the beverage container 3 recognized by the recognition processing device 1 using the supply port 150.

The auxiliary member 161 is a black member installed at a position where a black image is projected in the first region r1 of the photographed image when the beverage container 3a is placed at the placement position. Note that the color of the auxiliary member 161 is not limited to black, and for example, the recognition processing device 1 uses the characteristics of the beverage container 3b and the characteristics of the auxiliary member 161 as the characteristics of at least one of luminance and hue histograms. Any color can be used as long as it can be identified. In the following description, it is assumed that the feature of the auxiliary member 161 is the feature of at least one of the luminance and hue corresponding to each target image.
The imaging hole 170 is a hole provided for the imaging device 2 arranged inside the casing of the beverage supply device 100 to capture an image in the opening 130.

  The recognition processing device 1 acquires a captured image captured by the imaging device 2 and, as shown in FIGS. 2 to 5, when the target is placed at the placement position, regardless of the size of the beverage container 3. A second histogram of the second region r2 included in the photographed image from which the characteristics of the beverage container 3 are extracted is detected. And the recognition processing apparatus 1 determines with the drink container 3 having been mounted in the mounting position, when the characteristic of a 2nd histogram shows the characteristic of the drink container 3. FIG. At this time, the recognition processing apparatus 1 also determines the type of the beverage container 3 from the characteristics of the second histogram of the second region r2. Note that the first region r1 and the second region r2 shown in FIGS. 1 to 5 indicate the positions of the first region r1 and the second region r2 included in the captured image. In the following description, the second histogram of the second region r2 is simply referred to as a second histogram.

  When the recognition processing device 1 determines that the beverage container 3 is placed at the placement position, the recognition processing device 1 sets the position of the first region r1 according to the type of the beverage container 3, and detects the first histogram of the first region r1. . The position of the first region r1 is set according to the type of the beverage container 3 because the size differs depending on the type of the beverage container 3, and the region suitable for determining the difference in characteristics from the background is different. is there.

  And the recognition processing apparatus 1 determines with the size of the drink container 3 being regular size, when the 1st histogram of 1st area | region r1 shows the characteristic of a background. Moreover, the recognition processing apparatus 1 determines with the size of the drink container 3 being large size, when the 1st histogram of 1st area | region r1 shows the characteristic of the drink container 3. FIG. In the following description, the first histogram of the first region r1 is also simply referred to as a first histogram.

  When the beverage container 3a is placed at the placement position, the recognition processing device 1 detects a first histogram indicating the image characteristics of the auxiliary member 161 included in the first region r1 as the background. Moreover, the recognition processing apparatus 1 is the 1st histogram which shows the characteristic of the image of the edge of the opening part of the drink container 3b contained in 1st area | region r1 as a drink container when the drink container 3b is mounted in the mounting position. Is detected. Thereby, the recognition processing apparatus 1 clarifies the difference of the characteristic of the 1st histogram detected with the drink container 3a and the drink container 3b, and recognizes the drink container 3a and the drink container 3b with high precision.

  The recognition processing device 1 selects a type of beverage corresponding to the recognized type of the beverage container 3. Furthermore, the recognition processing apparatus 1 selects an amount corresponding to the recognized size of the beverage container 3. The recognition processing device 1 causes the beverage of the selected type and amount to be supplied from the beverage supply device 100 to the beverage container 3.

  As described above, the recognition processing device 1 recognizes the type and size of the target placed at the placement position by extracting the feature of the designated area included in the captured image captured by the imaging device 2. Therefore, the recognition accuracy of the target can be improved.

The process of the recognition processing apparatus 1 will be described in more detail.
FIG. 6 is a functional block diagram showing an embodiment of the recognition processing apparatus.
The recognition processing device 1 includes a control unit 10 and a storage unit 20. The recognition processing device 1 is connected to the imaging device 2. Moreover, the recognition processing apparatus 1 is connected with the control apparatus which controls operation | movement of the drink supply apparatus 100 which is not shown in figure. In the following description, the recognition processing device 1 will be described as supplying the beverage of the selected type and quantity from the beverage supply device 100 to the beverage container 3 by issuing an instruction to the control device. Note that the recognition processing device 1 may notify the control device of the type and size of the beverage container 3. In this case, the control device causes the beverage container 3 to supply the beverage of the type and amount selected according to the notification content from the beverage supply device 100.

  The control unit 10 includes an acquisition unit 11, a detection unit 12, a determination unit 13, a setting unit 14, and a selection unit 15. The storage unit 20 stores feature information 21, region information 22, type information 23, and size information 24. A plurality of pieces of information may be stored in the storage unit 20.

The feature information 21 stores a feature when the beverage container 3 is reflected and a feature when the beverage container 3 is not reflected in the first histogram of the first region r1 included in the photographed image. In addition, the feature information 21 stores, for the first histogram of the beverage container 3, a feature when an iced plastic cup is reflected and a feature when a white paper cup is reflected.
Further, the feature information 21 stores a feature when the beverage container 3 is reflected and a feature when the beverage container 3 is not reflected in the second histogram of the second region r2 included in the photographed image. In addition, the feature information 21 stores, for the second histogram of the beverage container 3, a feature when an iced plastic cup is reflected and a feature when a white paper cup is reflected.
The feature information 21 may store features experimentally obtained by the user. For example, the feature information 21 may extract features of the first region r1 and the second region r2 a plurality of times from a captured image captured a plurality of times, and store an average value of the extracted features. The feature information 21 may store the features of the first region r1 and the second region r2 that are considered to be theoretically obtained.

FIG. 7 is a diagram (part 1) illustrating a luminance frequency distribution in a detection region.
With reference to FIG. 7, an example of the feature information 21 when a luminance histogram is used for recognition of the beverage container 3 will be described.
The characteristic information 21 is a histogram of the luminance of the first region r1 and the second region r2 included in the photographed image for the cupless beverage container 3a, the beverage container 3b, the beverage container 3c, and the beverage container 3d shown in FIG. Stores features.

  The feature information 21 stores, for example, a feature in which pixels having luminance in the first region r1 and the second region r2 are equal to or less than the first threshold as features of the first histogram and the second histogram without a cup. This is because when the beverage container 3 is not placed at the placement position, as shown in the placement state without the cup, the images of the first region r1 and the second region r2 are each black as a background. This is a characteristic that the auxiliary member 161 and the black mounting table 141 are projected. When the beverage container 3 is not placed, the second region r2 is set so that an image of the black placement table 141 is projected as a background, so that the beverage container 3 is placed and the beverage The difference in the characteristics of the second histogram from when the container 3 is not placed is clarified. Thereby, the recognition processing apparatus 1 can recognize the beverage container 3a and the beverage container 3b with high accuracy.

  The feature information 21 stores, for example, a feature that a pixel having luminance in the first region r1 is equal to or less than a first threshold value as a feature of the first histogram of the beverage container 3a. In addition, the feature information 21 stores, for example, a feature that the luminance of the second region r2 has a spread equal to or larger than the second threshold value as a feature of the second histogram of the beverage container 3a.

  This is because when the beverage container 3a is placed at the placement position, as shown in the placement state of the beverage container 3a, the black auxiliary member 161 is projected as a background in the image of the first region r1. It is the characteristic which has. Further, when the beverage container 3a is placed at the placement position, as shown in the placement state of the beverage container 3a, the lower portion of the beverage container 3a is projected on the image of the second region r2, so that the ice This is because the luminance varies due to the reflection of light. The fact that the brightness of the second region r2 has a spread greater than or equal to the second threshold is, for example, that the variation in the brightness of the second threshold r2 is greater than or equal to a predetermined threshold, or the maximum and minimum values of the brightness of the second region r2 It may be determined on the basis that the difference is equal to or greater than a predetermined threshold.

  The feature information 21 stores, for example, a feature that a pixel having luminance corresponding to the edge of the opening of the beverage container 3b is included in the first region r1 as the feature of the first histogram of the beverage container 3b. In addition, the feature information 21 stores, for example, a feature that the luminance of the second region r2 has a spread equal to or larger than the second threshold value as a feature of the second histogram of the beverage container 3b.

  This is because when the beverage container 3b is placed at the placement position, as shown in the placement state of the beverage container 3b, the edge of the opening of the beverage container 3b is projected in the image of the first region r1. It is a feature to have. Further, when the beverage container 3b is placed at the placement position, as shown in the placement state of the beverage container 3b, the lower portion of the beverage container 3b is projected on the image of the second region r2, so that the ice This is because the luminance varies due to the reflection of light.

  The feature information 21 stores, for example, features corresponding to the background histogram in the first region r1 as features of the histogram of the beverage container 3c. The feature information 21 stores, for example, a feature that the brightness of the second region r2 has a spread less than the second threshold value as a feature of the histogram of the beverage container 3c.

  This is because when the beverage container 3c is placed at the placement position, as shown in the placement state of the beverage container 3c, the background of the first region r1 is projected in the image of the first region r1. It is the characteristic which has. Further, when the beverage container 3c is placed at the placement position, as shown in the placement state of the beverage container 3c, the lower region of the beverage container 3c is projected on the image of the second region r2, so that white This is a characteristic that the brightness corresponding to the appears.

  The feature information 21 stores, for example, a feature that the brightness of the second region r2 has a spread less than the second threshold value as a feature of the histogram of the beverage container 3d. This is because when the beverage container 3d is placed at the placement position, as shown in the placement state of the beverage container 3d, the images of the first region r1 and the second region r2 are respectively above the beverage container 3d. Since the portion and the lower portion are projected, this is a feature that the luminance corresponding to white appears.

FIG. 8 is a diagram (part 1) illustrating a hue frequency distribution in a detection region.
With reference to FIG. 8, an example of the feature information 21 when the hue histogram is used for the recognition of the beverage container 3 will be described.
The feature information 21 includes the hue of the captured image when the imaging device 2 captures the image of the cupless beverage container 3a, the beverage container 3b, the beverage container 3c, and the beverage container 3d shown in FIG. The features corresponding to the histograms are stored.

  The feature information 21 stores, for example, a feature that the type of color included in the first region r1 and the second region r2 is equal to or less than a fourth threshold value as the feature of the first histogram and the second histogram without the cup. This is because when the beverage container 3 is not placed at the placement position, as shown in the placement state without the cup, the images of the first region r1 and the second region r2 are each black auxiliary members. 161 and the black mounting table 141 are projected. When the beverage container 3 is not placed, the second region r2 is set so that an image of the black placement table 141 is projected as a background, so that the beverage container 3 is placed and the beverage The difference in the characteristics of the second histogram from when the container 3 is not placed is clarified. Thereby, the recognition processing apparatus 1 can recognize the beverage container 3a and the beverage container 3b with high accuracy.

  The feature information 21 stores, for example, a feature that the color type included in the first region r1 is the fourth threshold value or less as the feature of the first histogram of the beverage container 3a. In addition, the feature information 21 stores, for example, a feature that the color type included in the second region r2 is greater than or equal to the fifth threshold as a feature of the second histogram of the beverage container 3a.

  This is because when the beverage container 3a is placed at the placement position, as shown in the placement state of the beverage container 3a, the black auxiliary member 161 is projected in the image of the first region r1. It is a feature. Further, when the beverage container 3a is placed at the placement position, as shown in the placement state of the beverage container 3a, the lower portion of the beverage container 3b is projected on the image of the second region r2, so that the ice This is a feature possessed by the fact that there are a plurality of types of colors due to reflection.

  The feature information 21 stores, for example, a feature that the color type included in the first region r1 is equal to or greater than the sixth threshold value as a feature of the histogram of the beverage container 3b. In addition, the feature information 21 stores, for example, a feature that the color type included in the second region r2 is equal to or more than the fifth threshold value as a feature of the histogram of the beverage container 3b.

  This is because when the beverage container 3b is placed at the placement position, as shown in the placement state of the beverage container 3b, the edge of the opening of the beverage container 3b is projected in the image of the first region r1. It is a feature to have. Further, when the beverage container 3a is placed at the placement position, as shown in the placement state of the beverage container 3a, the lower portion of the beverage container 3b is projected on the image of the second region r2, so that the ice This is a feature possessed by the fact that there are a plurality of types of colors due to reflection.

  The feature information 21 stores, for example, a color feature corresponding to the background histogram in the first region r1 as the histogram feature of the beverage container 3c. In addition, the feature information 21 stores, for example, a feature that the color type included in the second region r2 is less than the fifth threshold value as a feature of the histogram of the beverage container 3c.

  This is because when the beverage container 3c is placed at the placement position, as shown in the placement state of the beverage container 3c, the background of the first region r1 is projected in the image of the first region r1. It is the characteristic which has. When the beverage container 3c is placed at the placement position, as shown in the placement state of the beverage container 3c, the image of the second region r2 includes an upper portion and a lower portion of the beverage container 3d, respectively. Since it is copied, it is a feature that the color corresponding to white appears.

  The feature information 21 stores, for example, a feature that the color type included in the first region r1 and the second region r2 is less than the fifth threshold value as a feature of the histogram of the beverage container 3d. When the beverage container 3d is placed at the placement position, as shown in the placement state of the beverage container 3d, the lower part of the beverage container 3d is projected on the images of the first region r1 and the second region r2. Therefore, this is a feature that the color corresponding to white appears.

This will be described with reference to FIG.
The area information 22 stores the positions of the first area r1 and the second area r2 corresponding to the type of the beverage container 3. More specifically, the region information 22 stores the position of the first region r1 that can determine whether the size of the beverage container 3 is a regular size or a large size from the characteristics of the first histogram. In addition, the area information 22 stores the position of the second area r2 where it can be determined from the characteristics of the second histogram whether or not the beverage container 3 is placed at the placement position.

  The area information 22 may store the positions of the first area r1 and the second area r2 obtained experimentally by the user. The region information 22 may store, for example, the positions of the first region r1 and the second region r2 that have been able to accurately determine the type and size of the beverage container 3 from the result of extracting the features a plurality of times. Further, the feature information 21 may store the features of the first region r1 and the second region r2 that can be theoretically obtained when the type and size of the beverage container 3 can be accurately determined. .

  For example, as shown in the first region r1 of the beverage container 3a and the beverage container 3b in FIG. 7, the region information 22 is the first region r1 of the transparent plastic cup containing ice, and the edge of the opening of the beverage container 3b. The area information indicating the position of is stored. Moreover, the area | region information 22 is above the opening part of the drink container 3c as the 1st area | region r1 of a white paper cup, for example, as shown to the 1st area | region r1 of the drink container 3c and the drink container 3d of FIG. The area information indicating the position below the opening of the beverage container 3d is stored.

  The region information 22 is, for example, region information indicating the position below the beverage container 3 as shown in the second region r2 of the cupless, beverage container 3a, beverage container 3b, beverage container 3c, and beverage container 3d in FIG. Is stored. The same position is set in the second region r2 regardless of the type of the beverage container 3.

  The type information 23 stores the container identifier for identifying the type of the beverage container 3, the type of beverage to be supplied corresponding to the type of the beverage container 3, and the temperature of the beverage to be supplied corresponding to the type of the beverage container 3. To do.

FIG. 9 is a diagram illustrating an example of type information.
An example of the type information 23 will be described with reference to FIG.
The type information 23 stores a container identifier, a beverage type, and a beverage temperature in association with each other.

  In the type information 23, the record associated with the beverage container 3a indicates that iced coffee having a temperature of 20 degrees is supplied to the beverage container 3a. The record associated with the beverage container 3b indicates that iced coffee at a temperature of 20 degrees is supplied to the beverage container 3b. The record associated with the beverage container 3c indicates that hot coffee having a temperature of 55 degrees is supplied to the beverage container 3c. The record associated with the beverage container 3d indicates that the beverage container 3d is supplied with hot coffee having a temperature of 55 degrees.

This will be described with reference to FIG.
The size information 24 stores the container identifier for identifying the type of the beverage container 3, the size of the beverage container 3, and the amount of the beverage corresponding to the size of the beverage container 3 in association with each other.

FIG. 10 is a diagram illustrating an example of the size information.
An example of the size information 24 will be described with reference to FIG.
The size information 24 stores a container identifier, a container size, and a beverage quantity in association with each other.

  In the size information 24, the record associated with the beverage container 3a indicates that the beverage container 3a is a regular size and 150 ml of coffee is supplied to the beverage container 3a. The record associated with the beverage container 3b indicates that the beverage container 3b is large and 300 ml of coffee is supplied to the beverage container 3b. Also, the record associated with the beverage container 3c indicates that the beverage container 3c is of regular size and 120 ml of coffee is supplied to the beverage container 3c. The record associated with the beverage container 3d indicates that the beverage container 3d is large and that the beverage container 3d is supplied with 240 ml of coffee.

This will be described with reference to FIG.
The acquisition unit 11 acquires a captured image of the placement position imaged by the imaging device 3. Moreover, the acquisition part 11 acquires the picked-up image containing the drink container 3 imaged with the imaging device 3, when the drink container 3 is mounted in the mounting position.

The detection unit 12 detects the characteristics of the first histogram of the first region r1 included in the captured image. More specifically, when the regular-sized beverage container 3 is placed at the placement position, the detection unit 12 detects a first histogram from which the background feature is extracted by the determination unit 13. Moreover, the detection part 12 detects the 1st histogram from which the characteristic of the drink container 3 is extracted by the determination part 13 when the large sized drink container 3 is mounted in the mounting position.
The beverage supply device 100 includes an auxiliary member 161 that makes the image of the first region r1 corresponding to the beverage container 3a opaque. Thereby, the detection part 12 makes clear the difference in the characteristic of the 1st histogram detected with the drink container 3a and the drink container 3b, and makes it possible to recognize the drink container 3a and the drink container 3b with high precision.

  The detection unit 12 detects the characteristics of the second histogram of the second region included in the captured image. More specifically, the detection unit 12 includes the first feature included in the captured image in which the feature of the beverage container 3 is extracted regardless of the size of the beverage container 3 when the beverage container 3 is placed at the placement position. A second histogram of two regions r2 is detected.

  When the beverage container 3 is placed at the placement position, the determination unit 13 determines that the size of the beverage container 3 is a regular size when the first histogram indicates a background feature, and the first histogram is the beverage container 3. When the feature is indicated, the size of the beverage container 3 is determined as the large size. Specifically, the determination unit 13 extracts the feature of the first histogram detected by the detection unit 12, and determines that the size of the beverage container 3 is a regular size when the extracted feature indicates a background feature. . Moreover, the determination part 13 extracts the characteristic of the 1st histogram detected by the detection part 12, and when the extracted characteristic shows the characteristic of the drink container 3, it determines with the size of the drink container 3 being large size.

The determination unit 13 determines that the beverage container 3 is placed at the placement position when the second histogram shows the characteristics of the beverage container 3. The second histogram is, for example, a histogram indicating the frequency distribution of luminance or hue of the pixels included in the second region r2.
The determination unit 13 determines the type of the beverage container 3 using the second histogram. When the second histogram indicates a frequency distribution of luminance, the determination unit 13 determines that the beverage container 3 is a transparent beverage container with ice when the spread of the second histogram is equal to or greater than the second threshold value. When the second histogram shows the frequency distribution of the hue, the determination unit 13 determines that the beverage container 3 is a transparent beverage container with ice when the color type included in the second region r2 is equal to or greater than the fifth threshold. Is determined.

  As shown in the placement state of FIG. 7, when the beverage container 3 is placed at the placement position according to the type of the beverage container 3, the setting unit 14, when the size of the beverage container 3 is a regular size, The position of the first region r1 is set at the position where the background feature is extracted. Further, as shown in the placement state of FIG. 7, the setting unit 14 has a large size when the beverage container 3 is placed at the placement position according to the type of the beverage container 3. In this case, the position of the first region r1 is set to a position where the characteristics of the beverage container 3 are extracted.

  When the beverage container 3 is a transparent beverage container 3 with ice, the setting unit 14 has a background feature when the size of the beverage container 3 is a regular size as shown in the placement state of the beverage container 3a in FIG. The position of the first region r1 is set to the position extracted by the determination unit 13. Moreover, when the drink container 3 is the transparent drink container 3 containing ice, as shown in the mounting state of the drink container 3b of FIG. 7, the setting part 14 is a drink container when the size of the drink container 3 is large size. The position of the first region r <b> 1 is set at a position where the edge feature of the three openings is extracted by the determination unit 13.

  When the beverage container 3 is not a transparent beverage container 3 with ice, the setting unit 14 determines the background characteristics when the size of the beverage container 3 is a regular size, as shown in the placement state of the beverage container 3c in FIG. The position of the first region is set to the position extracted by the unit 13. When the beverage container 3 is an opaque beverage container 3, the setting unit 14 is located at an upper position of the beverage container 3 when the size of the beverage container 3 is a large size, as shown in the placement state of the beverage container 3 d in FIG. 7. The position of the first region r1 is set at the position where the feature is extracted.

  The selection unit 15 selects the type of beverage to be provided according to the type of the beverage container 3. Specifically, when the type of the beverage container 3 is recognized by the determination process of the determination unit 13, the selection unit 13 refers to the type information 23, and the type of beverage associated with the recognized type of the beverage container 3. The beverage to be supplied to the beverage container 3 is selected according to the beverage temperature.

  The selection unit 15 selects the amount of beverage to be provided according to the size of the beverage container 3. Specifically, when the type of the beverage container 3 is recognized by the determination process of the determination unit 13, the selection unit 13 refers to the size information 24 and the size of the container associated with the recognized type of the beverage container 3. The amount of the beverage supplied to the beverage container 3 is selected according to the amount of the beverage.

  11 and 12 are flowcharts illustrating an example of the recognition process. A recognition process performed by the recognition processing apparatus 1 will be described with reference to FIGS. 11 and 12. In the following description, feature information 21, region information 22, type information 23, and size information 24 corresponding to beverage containers 3a to 3d are stored in advance in the storage unit 20 of the recognition processing device 1. Shall.

This will be described with reference to FIG.
The acquisition part 11 acquires the picked-up image of the area | region containing the drink container 3 mounted in the mounting position imaged with the imaging device 2 (S101).
The detection part 12 will detect the 2nd histogram which shows the characteristic of the pixel contained in the 2nd area | region r2 of the lower part of the drink container 3, if a picked-up image is acquired by the acquisition part 11 (S102).

  The determination unit 13 extracts the feature of the second histogram and determines whether or not the second histogram shows the feature of the beverage container 3 (S103). When the second histogram indicates the background feature (No in S103), the determination unit 13 determines that the beverage container 3 is not placed at the placement position, and executes the process of S101. When the second histogram shows the characteristics of the beverage container 3 (Yes in S103), the determination unit 13 determines that the beverage container 3 is placed at the placement position, and executes the process of S104.

The determination unit 13 determines whether or not the beverage container 3 is a plastic cup (S104).
As an example, when the second histogram is a luminance histogram, the determination unit 13 determines that the beverage container 3 is a plastic cup when the spread of the second histogram is equal to or greater than the second threshold. Further, when the second histogram is a luminance histogram, the determination unit 13 determines that the beverage container 3 is a paper cup when the spread of the second histogram is less than the second threshold.
As another example, when the second histogram is a hue histogram, the determination unit 13 determines that the beverage container 3 is a plastic cup when the color type included in the second region r2 is greater than or equal to the fifth threshold. . In addition, when the second histogram is a hue histogram, the determination unit 13 determines that the beverage container 3 is a paper cup when the color type included in the second region r2 is less than the fifth threshold.

  When the determination unit 13 determines that the beverage container 3 is a plastic cup (Yes in S104), the setting unit 13 refers to the region information 22 and sets the first region r1 at a position corresponding to the plastic cup (S105). . Specifically, when the beverage container 3 is a plastic cup, the setting unit 13 extracts the background characteristics when the beverage container 3 is a regular size, and the edge of the opening of the beverage container 3 when the beverage container 3 is a large size. The position of the first region r1 is set to the position where the features of.

When the setting unit 13 sets the first region r1, the detection unit 12 detects the first histogram (S106).
The determination unit 13 extracts the feature of the first histogram detected by the detection unit 12, and determines whether the beverage container 3 is a regular size or a large size (S106). Specifically, the determination unit 13 determines the size of the beverage container 3 as a regular size when the first histogram indicates a background feature, and determines the size of the beverage container 3 when the first histogram indicates the feature of the beverage container 3. The size is determined to be a large size.

  When the determination unit 13 determines that the beverage container 3 has a regular size (Yes in S107), the selection unit 13 supplies the beverage container 3 with a cold beverage having a regular size amount. (S108). Then, the recognition processing apparatus 1 ends the recognition process. Specifically, when the determination unit 13 determines that the beverage container 3 is the regular size of the plastic cup, the selection unit 13 refers to the type information 23 and specifies the type of beverage and the beverage associated with the beverage container 3a. A beverage to be supplied to the beverage container 3 is selected according to the temperature. Further, when the determination unit 13 determines that the beverage container 3 is the regular size of the plastic cup, the selection unit 13 refers to the size information 24 and determines the size of the container and the amount of the beverage associated with the beverage container 3a. In response, the amount of beverage to be supplied to the beverage container 3 is selected.

  When the determination unit 13 determines that the beverage container 3 is the large size (No in S107), the selection unit 13 supplies the beverage container 3 with a large amount of cold beverage supplied to the beverage container 3. (S108). Then, the recognition processing apparatus 1 ends the recognition process. Specifically, when the determination unit 13 determines that the beverage container 3 is a plastic cup large size, the selection unit 13 refers to the type information 23 and specifies the type of beverage and the beverage associated with the beverage container 3b. A beverage to be supplied to the beverage container 3 is selected according to the temperature. Further, when the determination unit 13 determines that the beverage container 3 is the plastic cup large size, the selection unit 13 refers to the size information 24 and determines the size of the container and the amount of the beverage associated with the beverage container 3b. In response, the amount of beverage to be supplied to the beverage container 3 is selected.

This will be described with reference to FIG.
When the determination unit 13 determines that the beverage container 3 is a paper cup (No in S104), the setting unit 13 refers to the region information 22 and sets the first region r1 at a position corresponding to the paper cup (S201). . Specifically, when the beverage container 3 is a paper cup, the setting unit 13 extracts a background feature when the beverage container 3 is a regular size, and a feature of an upper portion of the beverage container 3 when the beverage container 3 is a large size. The position of the first region r1 is set to the position where is extracted.
The detection unit 12 detects the first histogram when the setting unit 13 sets the first region r1 (S202).

  The determination unit 13 extracts the feature of the first histogram detected by the detection unit 12, and determines whether or not the beverage container 3 has a regular size (S203). Specifically, the determination unit 13 determines the size of the beverage container 3 as a regular size when the first histogram indicates a background feature, and determines the size of the beverage container 3 when the first histogram indicates the feature of the beverage container 3. The size is determined to be a large size.

  When the determination unit 13 determines that the beverage container 3 has a regular size (Yes in S204), the selection unit 13 supplies the beverage container 3 with a warm beverage having a regular size amount. (S205). Then, the recognition processing apparatus 1 ends the recognition process. Specifically, when the determination unit 13 determines that the beverage container 3 is the regular size of the paper cup, the selection unit 13 refers to the type information 23 and determines the type of beverage and the beverage associated with the beverage container 3c. A beverage to be supplied to the beverage container 3 is selected according to the temperature. Furthermore, specifically, when the determination unit 13 determines that the beverage container 3 is the regular size of the paper cup, the selection unit 13 refers to the size information 24 and the size of the container associated with the beverage container 3c. The amount of the beverage to be supplied to the beverage container 3 is selected according to the amount of the beverage.

  When the determination unit 13 determines that the beverage container 3 is the large size (No in S204), the selection unit 13 supplies the beverage container 3 with a large amount of cold beverage. (S206). Then, the recognition processing apparatus 1 ends the recognition process. Specifically, when the determination unit 13 determines that the beverage container 3 is the large size of the paper cup, the selection unit 13 refers to the type information 23 and determines the type of beverage and the beverage associated with the beverage container 3d. A beverage to be supplied to the beverage container 3 is selected according to the temperature. More specifically, when the determination unit 13 determines that the beverage container 3 is the large size of the paper cup, the selection unit 13 refers to the size information 24 and the size of the container associated with the beverage container 3d. The amount of the beverage to be supplied to the beverage container 3 is selected according to the amount of the beverage.

FIG. 13 is a block diagram illustrating an embodiment of a computer device.
The configuration of the computer apparatus 200 will be described with reference to FIG.

  In FIG. 13, a computer device 200 includes a control circuit 201, a storage device 202, a reading device 203, a recording medium 204, a communication interface 205, an input / output interface 206, an input device 207, a display device 208, An imaging device 209. In addition, the communication interface 205 is connected to the network 210. Each component is connected by a bus 211. The recognition processing device 1 can be configured by appropriately selecting some or all of the components described in the computer device 200.

  The control circuit 201 controls the entire computer device 200. The control circuit 201 is, for example, a processor such as a CPU (Central Processing Unit), a multi-core CPU, an FPGA (Field Programmable Gate Array), and a PLD (Programmable Logic Device). The control circuit 201 functions as the control unit 10 in FIG. 6, for example.

  The storage device 202 stores various data. The storage device 202 is, for example, a memory such as a ROM (Read Only Memory) and a RAM (Random Access Memory), or an HD (Hard Disk). For example, the storage device 202 functions as the storage unit 20 in FIG. 6.

  The ROM stores a program such as a boot program. The RAM is used as a work area for the control circuit 201. The HD stores an OS, an application program, a program such as firmware, and various data. The storage device 202 may store a recognition processing program that causes the control circuit 201 to function as the control unit 10.

  When performing the recognition process, the recognition processing device 1 reads the program stored in the storage device 202 into the RAM. By executing the program read into the RAM with the control circuit 201, the recognition processing apparatus 1 recognizes including at least one of an acquisition process, a detection process, a determination process, a setting process, and a selection process. Execute the process.

  Each program described above may be stored in a storage device included in a server on the network 210 as long as the control circuit 201 is accessible via the communication interface 205.

  The reading device 203 is controlled by the control circuit 201 and reads / writes data on the removable recording medium 204. The reading device 203 is, for example, an FDD (Floppy Disk Drive), a CDD (Compact Disc Drive), a DVDD (Digital Versatile Disk Drive), a BDD (Blu-ray Disk Drive), or a BDD (Blu-ray Disk Drive).

  The recording medium 204 stores various data. The recording medium 204 stores a recognition processing program, for example. Furthermore, the recording medium 204 may store the feature information 21, the region information 22, the type information 23, and the size information 24 illustrated in FIG. The recording medium 204 is connected to the bus 211 via the reading device 203, and the control circuit 201 controls the reading device 203 to read / write data.

  The recording medium 204 is, for example, an SD memory card (SD Memory Card), an FD (Floppy Disk), a CD (Compact Disc), a DVD (Digital Versatile Disk), a BD (Blu-ray Disk: registered trademark), and a flash. A non-temporary recording medium such as a memory.

  The communication interface 205 connects the computer device 200 and other devices via the network 210 so that they can communicate with each other. The communication interface 205 may include an interface having a wireless LAN function and an interface having a short-range wireless communication function. The wireless LAN interface may support Wi-Fi (registered trademark) as a wireless LAN standard, for example. For example, the short-range wireless interface may support Bluetooth (registered trademark) as a short-range wireless communication standard. LAN is an abbreviation for Local Area Network.

The input / output interface 206 is connected to an input device 207 such as a keyboard, a mouse, and a touch panel, for example. When a signal indicating various information is input from the connected input device 207, the signal input via the bus 211 is input. Is output to the control circuit 201. In addition, when a signal indicating various information output from the control circuit 201 is input via the bus 211, the input / output interface 206 outputs the signal to various connected devices.
Further, the input / output interface 206 is connected to the display device 208, for example, and outputs an image drawn by the control circuit 201 to the display device.
The input / output interface 206 is connected to the imaging device 209 and accepts input of a captured image captured by the imaging device 209.

The input device 207 accepts, for example, adjustment of the amount of sugar and milk to be added to the coffee, selection of the coffee strength, fine adjustment of each threshold value of the feature information 21 and the position of each region stored in the region information 22, etc. May be. Further, the input device 207 may accept input of the feature information 21, the region information 22, the type information 23, and the size information 24.
The display device 208 displays various information. The display device 208 may display information for accepting input on the touch panel.

The input / output interface 206, the input device 207, and the display device 208 may function as a GUI (Graphical User Interface). Thereby, the computer apparatus 200 accepts an intuitive operation using a touch panel, a mouse, or the like. At this time, the GUI may accept, for example, adjustment of the amount of sugar and milk to be added to the coffee or selection of selection of the coffee strength.
The imaging device 209 is, for example, a camera. The imaging device 209 functions as, for example, the imaging device 2 in FIG.
The network 210 is, for example, a LAN, wireless communication, the Internet, or the like, and communicatively connects the computer apparatus 200 to another apparatus.

  As described above, when the beverage container 3 is placed at the placement position, the recognition processing device 1 according to the first embodiment has the size of the beverage container 3 when the first histogram of the first region r1 indicates the background feature. Is determined as a regular size. Moreover, the recognition processing apparatus 1 determines the size of a drink container as large size, when a 1st histogram shows the characteristic of the drink container 3. FIG. Thereby, since the recognition processing apparatus 1 recognizes the size of the beverage container 3 placed at the placement position by extracting the feature of the first region r1, the recognition accuracy is based on how the beverage container 3 is imaged. , And the recognition accuracy of the beverage container 3 can be improved.

  When the size of the beverage container 3 is a regular size, the recognition processing device 1 extracts a background feature, and when the size of the beverage container 3 is a large size, the feature of the beverage container 3 is The position of the first region r1 is set as the extracted position. Thereby, since the recognition processing apparatus 1 automatically determines the sizes of the plurality of types of beverage containers 3, the recognition processing of the beverage containers 3 can be facilitated.

  When the spread of the second histogram of the ice image is equal to or larger than the second threshold, the recognition processing device 1 utilizes the fact that the ice portion diffuses light and diffuses the frequency distribution of the second histogram. It is determined that the beverage container 3 is a transparent beverage container 3 containing ice. As a result, the recognition processing apparatus 1 can improve the recognition accuracy of the transparent beverage container 3 with ice, which is difficult to recognize by image recognition because the way of entering ice is different for each beverage container 3.

  When the beverage container 3 is a transparent beverage container with ice, the recognition processing device 1 is located at a position where the feature of the opening of the beverage container 3 is extracted when the size of the beverage container 3 is large. The position of the area r1 is set. Thus, when the beverage container 3 is a plastic cup, the recognition processing device 1 determines the size of the beverage container 3 by extracting features that are clearly different from the edge having a specific reflection and the background. Recognition accuracy can be improved.

  When the beverage container 3 is not a transparent beverage container containing ice, the recognition processing device 1 extracts the background features when the size of the beverage container 3 is a regular size, and the beverage container 3 when the size of the beverage container 3 is a large size. The position of the first region r1 is set to a position where the upper position feature is extracted. Thereby, when the beverage container 3 is a paper cup, the recognition processing apparatus 1 determines the size of the beverage container 3 by extracting features that are clearly different between the background and the paper cup, so that the recognition accuracy of the beverage container 3 is improved. can do.

  The recognition processing device 1 selects the amount of beverage to be provided according to the size of the beverage container 3. Accordingly, the recognition processing device 1 can automatically supply a selected amount of beverage from the beverage supply device 100 to the beverage container 3.

  The recognition processing device 1 selects the type of beverage to be provided according to the type of the beverage container 3. Thereby, the recognition processing apparatus 1 can automatically supply the selected type of beverage from the beverage supply device 100 to the beverage container 3.

  The recognition processing device 1 determines that the target is placed at the placement position when the second histogram indicates the characteristics of the beverage container 3. Thereby, the recognition processing apparatus 1 can automatically start the process which determines the kind and size of the drink container 3, when the drink container 3 is mounted.

[Embodiment 2]
A recognition processing apparatus according to the second embodiment will be described.
FIG. 14 is a view (No. 6) showing an embodiment of the beverage supply apparatus. FIG. 15 is a diagram for explaining recognition of a cup on which characters are described.
The recognition processing device 6 according to the second embodiment has two types of drink containers 3 containing ice, on which at least one of characters and drawings is written on the side, and drink containers 3 containing ice, which are not shown on the side. The beverage container 3 containing is identified. The functional block of the recognition processing device 6 has the same configuration as the functional block of the recognition processing device 1 of the first embodiment, but the processing of the detection unit 12 and the determination unit 13 is different from the information stored in the region information 22. Processing of the detection unit 12 and the determination unit 13 and the region information 22 will be described.

Processing of the recognition processing device 6 will be described with reference to FIGS. 14 and 15. In the following description, as shown in FIGS. 14 and 15, the beverage container 3e is described on the assumption that the letters “ABCDEFG...” Are written at a position surrounding the lower portion.
The area information 22 stores the position of the third area r3 that includes characters and the position of the fourth area r4 that does not include characters in the captured image of the beverage container 3e in which the characters captured by the imaging device 2 are described. . In the third region r3 and the fourth region r4, a position where the photographed image includes the image of ice contained in the beverage container 3 is set.

When the beverage container 3 is placed at the placement position, the detection unit 12 detects the third histogram of the luminance of the third region r3 and the fourth histogram of the luminance of the fourth region r4.
When the difference dif between the third histogram and the fourth histogram indicates the characteristics of the fifth histogram according to the brightness of the characters, the determination unit 13 determines that the beverage container 3 is a beverage container 3e with the description of characters.

The case where the recognition processing device 6 recognizes whether or not a graphic is described will be described.
The area information 22 is the position of the third area r3 including the graphic and the position of the fourth area r4 not including the graphic in the captured image of the beverage container 3d (not shown) in which the graphic imaged by the imaging device 2 is described. And store. In the third region r3 and the fourth region r4, a region including the position of ice contained in the beverage container 3 in the photographed image is set.

When the beverage container 3 is placed at the placement position, the detection unit 12 detects the third histogram of the luminance of the third region r3 and the fourth histogram of the luminance of the fourth region r4.
When the difference dif between the third histogram and the fourth histogram indicates the characteristics of the fifth histogram according to the luminance of the graphic, the determination unit 13 determines that the beverage container 3 is a beverage container 3e having a graphic description. .

  In the above description, the beverage container 3 is described as being an iced beverage container. However, by executing the same process, at least one of characters and figures is described in the other beverage containers 3 as well. It can be determined whether or not. In addition, the beverage container 3e does not necessarily have characters and figures written in the entire position surrounding the lower part, and at least characters and figures are included in the third region r3. It only has to be a graphic.

  As described above, the recognition processing device 6 according to the second embodiment determines whether or not at least one of a character and a graphic is described by looking at the difference between features. Therefore, the recognition accuracy of the image of the beverage container 3 can be improved. A fall can be suppressed and the recognition accuracy of the beverage container 3 can be improved.

[Embodiment 3]
A beverage supply apparatus 300 according to Embodiment 3 will be described. The recognition processing device 1 according to the third embodiment has the same configuration as the recognition processing device 1 according to the first embodiment.
FIG. 16 is a view (No. 7) showing an embodiment of the beverage supply apparatus. FIG. 17 is a diagram (part 2) illustrating the luminance frequency distribution in the detection region. FIG. 18 is a diagram (part 2) illustrating a hue frequency distribution in a detection region.

  As shown in FIG. 16, the beverage supply device 300 according to the third embodiment includes an auxiliary member 162. The auxiliary member 161 is a black member installed at a position where a black image is projected in the first region r1 of the photographed image when the beverage container 3c is placed at the placement position. Note that the color of the auxiliary member 161 is not limited to black. For example, the recognition processing device 1 uses the characteristics of the beverage container 3d and the characteristics of the auxiliary member 162 as the characteristics of at least one of luminance and hue histograms. Any color can be used as long as it can be identified.

  When the beverage container 3c is placed at the placement position, the detection unit 12 detects a first histogram indicating the characteristics of the image of the auxiliary member 162 included in the first region r1 as the background. Moreover, the detection part 12 detects the 1st histogram which shows the characteristic of the upper part of the drink container 3d contained in 1st area | region r1 as the drink container 3 when the drink container 3d is mounted in the mounting position. Thereby, the recognition processing apparatus 1 clarifies the difference in the characteristics of the first histogram detected between the beverage container 3c and the beverage container 3d, and recognizes the beverage container 3c and the beverage container 3d with high accuracy.

  As shown in FIGS. 17 and 18, when the beverage container 3 c is placed at the placement position of the beverage supply device 300 according to the third embodiment, the first histogram detected by the recognition processing device 1 shows the luminance and color. The characteristic of the auxiliary member 162 having few pixels is shown. Thereby, the difference between the first histogram when the beverage container 3c is placed and the first histogram when the beverage container 3d is placed is clarified, and the beverage container 3c and the beverage container 3d are accurately obtained. The recognition processing apparatus 1 can be made to recognize.

[Embodiment 4]
The recognition processing device 7 and the beverage supply device 400 according to the fourth embodiment will be described. The functional blocks of the recognition processing device 7 of the fourth embodiment have the same configuration as the functional blocks of the recognition processing device 1 of the first embodiment, but the processing of the determination unit 13 is different, so the processing of the determination unit 13 will be described.

FIG. 19 is a view (No. 8) showing one embodiment of the beverage supply apparatus.
The beverage supply device 400 includes a sensor 4 in addition to the configuration of the beverage supply device 100. For example, the sensor 4 emits light by the light emitting unit and receives light reflected by the beverage container 3 by the light receiving unit. The light reflective sensor detects that the beverage container 3 is on the mounting table, or the beverage container 3. Is a pressure-sensitive sensor that detects the pressure when the is mounted on the mounting table 141.
The sensor 4 detects whether or not the beverage container 3 is placed at the placement position, and notifies the recognition processing device 7 of it. When it is detected that the beverage container 3 is placed at the placement position, the determination unit 13 determines that the beverage container 3 is placed at the placement position.

As described above, the recognition processing device 7 can reduce the processing load for detecting that the beverage container 3 is placed at the placement position.
In addition, this embodiment is not limited to embodiment described above, A various structure or embodiment can be taken in the range which does not deviate from the summary of this embodiment.

DESCRIPTION OF SYMBOLS 1 Recognition processing apparatus 2 Imaging device 10 Control part 20 Storage part 100 Beverage supply apparatus 200 Computer apparatus 201 Control circuit 202 Storage apparatus 203 Reading apparatus 204 Recording medium 205 Communication I / F
206 I / O I / F
207 Input device 208 Display device 209 Imaging device 210 Network 211 Bus

Claims (15)

An acquisition unit for acquiring a captured image of an area including a target placed at the placement position;
When the target is placed at the placement position, if the size of the target is the first size, a background feature is extracted, and if the size of the target is a second size larger than the first size, A detection unit for detecting a first characteristic of a first region included in the captured image from which the feature of the target is extracted;
When the target is placed at the placement position, when the first characteristic indicates a background feature, the size of the target is determined as the first size, and the first characteristic indicates the target feature A determination unit that determines the size of the target as a second size;
A recognition processing apparatus comprising: The detector is
When the object is placed at the placement position, the second characteristic of the second region included in the captured image from which the feature of the object is extracted regardless of the size of the object;
The determination unit
Using the second characteristic to determine the type of the object;
The recognition processing device further includes:
According to the type of the target, when the target is placed at the placement position, if the size of the target is the first size, the background feature is extracted, and the size of the target is the second size. The recognition processing apparatus according to claim 2, further comprising: a setting unit that sets a position of the first region at a position where the feature of the target is extracted in the case of size. The subject is
A cup-shaped beverage container,
The second characteristic is:
Including a second frequency distribution of luminance of the second region;
The determination unit
The recognition processing device according to claim 1, wherein when the spread of the second frequency distribution is equal to or greater than a second threshold value, the beverage container is determined to be a transparent beverage container containing ice. The setting unit
When the beverage container is a transparent beverage container with ice, a background feature is extracted when the size of the beverage container is the first size, and when the size of the beverage container is the second size, The recognition processing apparatus according to claim 3, wherein the position of the first region is set to a position where the feature of the edge of the opening is extracted. The setting unit
When the beverage container is an opaque beverage container, background features are extracted when the size of the beverage container is the first size, and when the size of the beverage container is the second size, The recognition processing apparatus according to claim 3, further comprising: setting a position of the first region at a position where the feature is extracted. The type of the beverage container is
Including two types, a first beverage container without description of characters and graphics, and a second beverage container with description of at least one of the characters and graphics,
The detector is
When the beverage container is placed at the placement position, in the captured image of the second beverage container, the third frequency distribution of the brightness of the third region including at least one of the characters and graphics, the characters and Detecting a fourth frequency distribution of luminance in the fourth region, which does not include a figure,
The determination unit
When the difference between the third frequency distribution and the fourth frequency distribution indicates the characteristics of the fifth frequency distribution according to the luminance of at least one of the characters and figures, the beverage container is the second beverage container. The recognition processing device according to claim 3, wherein the recognition processing device is determined. When the second beverage container is the transparent beverage container containing ice, and at least one of the letters and figures is written at the position containing the ice,
The third region and the fourth region are:
The recognition processing device according to claim 6, wherein the recognition processing device is a region including a position where the ice is contained. The recognition processing device further includes:
The recognition processing apparatus according to any one of claims 3 to 7, further comprising a selection unit that selects an amount of a beverage to be provided according to a size of the beverage container. The recognition processing device further includes:
The recognition processing apparatus according to any one of claims 3 to 8, further comprising a selection unit that selects a type of beverage to be provided according to a type of the beverage container. The detector is
When the object is placed at the placement position, the second characteristic of the second region included in the captured image from which the feature of the object is extracted regardless of the size of the object;
The determination unit
The recognition processing apparatus according to any one of claims 1 to 9, wherein when the second characteristic indicates the feature of the target, it is determined that the target is placed at the placement position. The determination unit
When it is detected that the target is placed by a sensor that detects that the target is placed at the placement position, it is determined that the subject is placed at the placement position. The recognition processing apparatus according to any one of claims 1 to 10. The first region is
The recognition processing device according to any one of claims 1 to 11, wherein the recognition processing device is a region that becomes an opaque image when a second-size target is not placed at the placement position. An acquisition unit for acquiring a captured image of an area including a target placed at the placement position;
When the target is placed at the placement position, if the size of the target is the first size, a background feature is extracted, and if the size of the target is a second size larger than the first size, A detection unit for detecting a first characteristic of a first region included in the captured image from which the feature of the target is extracted;
When the target is placed at the placement position, when the first characteristic indicates a background feature, the size of the target is determined as the first size, and the first characteristic indicates the target feature A determination unit that determines the size of the target as a second size;
An auxiliary unit that makes an image projected in the first region opaque when a second size target is not set at the placement position;
A beverage supply device comprising: An information processing support method executed by a processor of a recognition processing device,
The processor is
Acquire a captured image of the area including the target placed at the placement position,
When the target is placed at the placement position, if the size of the target is the first size, a background feature is extracted, and if the size of the target is a second size larger than the first size, Detecting a first characteristic of a first region included in the captured image from which the feature of the object is extracted;
When the target is placed at the placement position, when the first characteristic indicates a background feature, the size of the target is determined as the first size, and the first characteristic indicates the target feature And determining the size of the object as a second size. An information processing program executed by a processor of a recognition processing device,
Acquire a captured image of the area including the target placed at the placement position,
When the target is placed at the placement position, if the size of the target is the first size, a background feature is extracted, and if the size of the target is a second size larger than the first size, Detecting a first characteristic of a first region included in the captured image from which the feature of the object is extracted;
When the target is placed at the placement position, when the first characteristic indicates a background feature, the size of the target is determined as the first size, and the first characteristic indicates the target feature A recognition processing program for causing a processor to execute a process of determining the target size as a second size.