JP7665905B2 - Counting device, counting method and program - Google Patents

Description

This disclosure relates to a counting device, a counting method, and a program.

There is a known device that uses a camera to capture images of the edges of multiple stacked objects to be counted, and then uses the captured images to count the number of objects to be counted.

Patent document 1 discloses a counting method in which multiple folded sheets of paper are imaged with a television camera, and the number of non-shadowed areas at each component point that makes up one horizontal scanning line is added up to count the number of sheets of paper.

Patent document 2 discloses a banknote counting method that includes an imaging step, a removal step, a histogram generation step, and a bundle counting step. Patent document 2 discloses that the bundle counting step counts the number of piles in the histogram as the number of bundles.

JP 2014-002701 A JP 2015-055991 A

When using images of the edges of counting objects such as stacked nylon plastic bags or cards, there is a demand for technology that can count the number of items even when there is dirt or dust on the edges or when the shading of the edges is unclear.

This disclosure provides a technology that allows for more accurate counting when using captured images of the edges of objects to be counted.

The present disclosure is a counting device that includes a processing unit that processes an image of a plurality of counting objects, each of whose ends is aligned to extend in a first direction and arranged in a second direction intersecting the first direction, captured from the side of the ends to count the number of the counting objects, and the processing unit executes an averaging process that calculates an average value of pixel values in an area in the captured image that extends in the first direction and the second direction, across the second direction while moving through the area, and a counting process that counts the number of the counting objects using the average value.

According to the present disclosure, more accurate counting can be achieved when counting using captured images of the edges of the objects to be counted.

FIG. 1 is a diagram showing a schematic configuration of an example of a counting device according to the present embodiment. FIG. 2 is a diagram showing a schematic configuration of a control unit of an example of a counting device according to the present embodiment. FIG. 3 is a functional block diagram of a control unit of an example of a counting device according to the present embodiment. FIG. 4 is a flowchart showing an example of processing of the counting device according to the present embodiment. FIG. 5 is a flowchart showing an example of processing of the counting device according to the present embodiment. FIG. 6 is a diagram illustrating an example of processing of the counting device according to the present embodiment. FIG. 7 is a diagram illustrating an example of processing of the counting device according to the present embodiment. FIG. 8 is a diagram illustrating an example of processing of the counting device according to the present embodiment. FIG. 9 is a diagram illustrating an example of processing of the counting device according to the present embodiment. FIG. 10 is a diagram illustrating an example of processing of the counting device according to the present embodiment.

Each embodiment of the present invention will be described below with reference to the attached drawings. Note that, in the description of the specification and drawings relating to each embodiment, components having substantially the same or corresponding functional configurations may be given the same reference numerals to avoid redundant description. Also, to facilitate understanding, the scale of each part in the drawings may differ from the actual scale.

In directions such as parallel, right-angled, orthogonal, horizontal, vertical, up-down, left-right, etc., deviations are permitted to the extent that they do not impair the action and effect of the embodiment. The shape of the corners is not limited to right angles, and may be rounded in a bow shape. Parallel, right-angled, orthogonal, horizontal, and vertical may include approximately parallel, approximately right-angled, approximately orthogonal, approximately horizontal, and approximately vertical.

<Counting device 1>
1 is a diagram showing a schematic configuration of a counting device 1, which is an example of a counting device according to the present embodiment. The counting device 1 is a device that captures images of a plurality of counting objects T using an imaging unit 11, imports the captured images, and displays the number of counting objects T on a screen by software processing. The counting objects T are, for example, pouch bags, empty envelopes, filled envelopes, folded paper, postcards, cards, etc.

For ease of explanation, an XYZ Cartesian coordinate system is set in FIG. 1. For coordinate axes perpendicular to the paper surface of the drawing, a cross in a circle on the coordinate axis indicates that the direction into the paper surface is positive. However, this coordinate system is defined for the purpose of explanation and does not limit the attitude of the counting device 1, etc.

In this disclosure, unless otherwise specified, the X-axis and Y-axis are parallel to the mounting surface 10b of the glass window 10a of the mounting portion 10c. The X-axis is the axis in the arrangement direction of the counting objects T. The Y-axis is the axis in the spread direction of the counting objects T. The Z-axis is the axis perpendicular to the X-axis and Y-axis.

The counting device 1 includes an imaging unit 11, an illumination unit 12, and a control unit 15 inside the housing 10. The counting device 1 further includes a display operation unit 21, a printing unit 22, an external input/output unit 23, and a power supply unit 25. The imaging unit 11, the illumination unit 12, the display operation unit 21, the printing unit 22, the external input/output unit 23, and the power supply unit 25 are connected to the control unit 15.

The counting device 1 has a glass window 10a on the top of the housing 10. The glass window 10a has a mounting surface 10b on the outside of the housing 10. The counting object T is placed on the mounting surface 10b of the glass window 10a. The counting device 1 has a pressing jig 31 and a pressing jig 32 that sandwich and hold the counting object T placed on the glass window 10a. The portion of the glass window 10a having the mounting surface 10b on which the counting object T is placed is called the mounting portion 10c.

The counting objects T are placed on the placement section 10c in a row in the X-axis direction, i.e., in the arrangement direction. That is, a plurality of counting objects T are placed on the placement surface 10b of the placement section 10c in a row in the X-axis direction (arrangement direction). Each of the counting objects T is aligned so that its end extends in the Y-axis direction and arranged in the X-axis direction. Each of the counting objects T is placed on the placement section 10c so that it spreads in the Y-axis and Z-axis directions. Note that the Y-axis direction, which is parallel to the placement surface 10b and intersects with the X-axis direction, may be referred to as the spread direction of the counting objects T.

[Imaging unit 11]
The imaging unit 11 is a camera that images the end TF where the ends of the multiple counting objects T placed on the glass window 10a are lined up. That is, the imaging unit 11 images the multiple counting objects T from the end side. Specifically, the imaging unit 11 images the multiple counting objects T in the +Z direction of the Z axis direction. The imaging unit 11 is, for example, a monochrome area camera having 5 million pixels, with approximately 2500 pixels in the horizontal direction and 2000 pixels in the vertical direction. The imaging unit 11 images the counting objects T placed on the placement surface 10b of the glass window 10a through the glass window 10a.

A lens 11a is attached to the imaging unit 11. The horizontal field of view of the imaging unit 11 when capturing an image using the lens 11a is, for example, 25 millimeters. Therefore, the length of one pixel is approximately 10 micrometers. Note that the horizontal field of view of the imaging unit 11 is not limited to 25 millimeters, and may be set appropriately between, for example, 20 millimeters and 150 millimeters depending on the object to be measured.

[Lighting unit 12]
The illumination unit 12 illuminates the counting object T when the imaging unit 11 captures an image. The illumination unit 12 is, for example, a light-emitting diode, a light bulb, a discharge tube, a fluorescent lamp, or the like.

[Control unit 15]
The control unit 15 controls the entire counting device 1. The control unit 15, for example, captures imaging data from the imaging unit 11. Then, the control unit 15 performs image processing on the captured imaging data to count the counting objects T. The control unit 15 controls the illumination unit 12. The control unit 15 also controls the display operation unit 21, the printing unit 22, and the external input/output unit 23.

Figure 2 is a diagram showing a schematic configuration of the control unit 15 of the counting device 1, which is an example of a counting device according to this embodiment. The control unit 15 is, for example, a computer. The control unit 15 includes a CPU (Central Processing Unit) 51, a RAM (Random Access Memory) 52, and a ROM (Read Only Memory) 53. The CPU 51 operates based on a program stored in the ROM 53.

The control unit 15 also includes an imaging interface (imaging IF (Interface)) 54. The imaging unit 11 is connected to the imaging interface 54. The control unit 15 also includes an I/O (Input/Output) interface (I/O IF (Interface)) 55. The display operation unit 21, the printing unit 22, and the external input/output unit 23 are connected to the I/O interface 55.

[Display operation unit 21]
The display operation unit 21 accepts input from a user. The display operation unit 21 is used, for example, for setting input. The display operation unit 21 also displays information such as images to the user. The display operation unit 21 is used, for example, for displaying counting results and processed images. The display operation unit 21 is, for example, a display with a touch panel.

[Printing unit 22]
The printing unit 22 prints and outputs the counting results, etc. on paper. The printing unit 22 is, for example, a printer. The printing unit 22 may be omitted.

[External input/output section 23]
The external input/output unit 23 is used when connecting to an external device, such as a display device or a computer.

[Power supply unit 25]
The power supply unit 25 converts the power supplied to the counting device 1 and supplies it to each component of the counting device 1 .

[Pressing jig 31 and pressing jig 32]
The pressing jigs 31 and 32 are jigs for stably holding the counting objects T placed on the glass window 10a. The pressing jigs 31 and 32 hold the counting objects T so that they are lined up in the X-axis direction (line-up direction). The pressing jigs 31 and 32 also hold the counting objects T so that their ends are aligned parallel to the Y-axis direction (spreading direction).

A light-shielding cover may be provided on the housing 10 to cover the holding jig 31, the counting object T, and the holding jig 32.

<Processing of the control unit 15 of the counting device 1>
Next, a description will be given of the processing of the counting device 1, specifically, the processing of the control unit 15 of the counting device 1. Fig. 3 is a functional block diagram showing the processing performed by the control unit 15 of the counting device 1, which is an example of a counting device according to this embodiment.

The control unit 15 has an image acquisition unit 61, a preprocessing unit 62, a region averaging unit 63, a binarization unit 64, and a counting unit 65.

[Image acquisition unit 61]
The image acquiring unit 61 acquires images captured by the imaging unit 11 from the imaging unit 11. The image acquiring unit 61 acquires captured images of a plurality of counting objects aligned so that their ends extend in the Y-axis direction and arranged in the X-axis direction intersecting the first direction, captured from the end side.

[Preprocessing unit 62]
The pre-processing unit 62 processes the image acquired by the image acquisition unit 61 so as to facilitate processing in subsequent steps. The pre-processing unit 62 performs image processing such as gamma correction, brightness correction, contrast correction, filter processing, morphology processing (opening processing, closing processing, etc.), and geometric transformation processing.

The filter process may be, for example, a median filter, a sharpening filter, etc. In addition, a positive-negative conversion process may be performed as the filter process. The filter process is not limited to a single filter process, and multiple filter processes may be combined.

When performing contrast correction, the contrast may be corrected, for example, using a sigmoid function.

If the captured image is tilted, the image may be subjected to a geometric transformation, such as an affine transformation, to correct the tilt.

Furthermore, the pre-processing unit 62 may cut out a part of the image acquired by the image acquisition unit 61 and perform processing. For example, multiple regions, for example, three regions, may be extracted from the image acquired by the image acquisition unit 61, and counting processing of the counting objects T may be performed in each of the multiple regions.

When counting the counting objects T in multiple regions, if the counting results in the multiple regions do not match, for example, an error may be displayed and the counting may be performed again.

[Area averaging section 63]
The region averaging unit 63 moves a rectangular region that is long in the direction in which the end faces extend in the arrangement direction, and calculates the average of the pixel values of the pixels included in the rectangular region.

[Binarization section 64]
The binarization unit 64 binarizes the result of averaging performed by the area averaging unit 63. The binarization process is, for example, adaptive binarization. Note that the binarization process is not limited to adaptive binarization, and binarization may be performed using a predetermined pixel value as a threshold, or other methods may be applied.

[Counting unit 65]
The counting unit 65 uses the binarized result from the binarization unit 64 to count the number of objects T to be counted, for example, by counting the number of places where dark points change to bright points.

Next, the counting process of the counting objects T in the counting device 1, i.e., the counting method, will be described using a flowchart. Figures 4 and 5 are flowcharts showing the process of the counting device 1, which is an example of a counting device according to this embodiment. Figures 4 and 5 show the procedure executed by the control unit 15.

First, the user aligns the objects to be counted T and places them on the glass window 10a of the counting device 1. The user then uses the clamping jigs 31 and 32 to fix the objects to be counted T on the glass window 10a. The user operates the display operation unit 21 to cause the counting device 1 to start the counting process. Note that the counting device 1 may automatically start the counting process, for example, when it detects that the objects to be counted T have been fixed on the glass window 10a.

Next, the imaging unit 11 captures an image of the end TF of the counting object T (step S10, a process of capturing an image of the end of the counting object by the imaging unit (imaging process, imaging process, imaging procedure)). Specifically, the control unit 15 controls the imaging unit 11 to capture an image of the end TF of the counting object T. Then, the image acquisition unit 61 of the control unit 15 acquires an image from the imaging unit 11. The image acquisition unit 61 acquires an image captured from the end side of multiple counting objects that are aligned so that their respective ends extend in the Y-axis direction and are arranged in the X-axis direction intersecting the first direction.

Image IMG1, which is an example of an image acquired by image acquisition unit 61 from imaging unit 11, is shown in FIG. 6. Image IMG1 is an image (captured image) of end TF of counting object T. Arrow DX in FIG. 6 indicates a direction (arrangement direction DX) corresponding to the arrangement direction (X-axis direction in FIG. 1) of counting objects T in the captured image (image IMG1). Arrow DY in FIG. 6 indicates a direction (spread direction DY) corresponding to the spread direction (Y direction in FIG. 1) of counting objects T in the captured image (image IMG1).

In image IMG1, bright pixels are arranged in a line in the direction DY along which the counting objects T extend. Also in image IMG1, rows of bright pixels are arranged in a line at approximately equal intervals in the arrangement direction DX of the counting objects T. Each row of bright pixels corresponds to one edge of the counting objects T. Therefore, the number of counting objects T can be counted by counting the number of rows of bright pixels in the arrangement direction of the counting objects T.

Next, the captured image is preprocessed (step S20, a process of preprocessing the captured image). Specifically, the preprocessing unit 62 of the control unit 15 preprocesses the image acquired by the image acquisition unit 61 from the imaging unit 11. The contents of the preprocessing may be set, for example, using the display operation unit 21.

Next, the pixel values of the pixels contained in the rectangular region in which the preprocessed image is set are averaged (step S30, process of averaging the preprocessed image over a rectangular region (averaging process, averaging process, averaging procedure)). In the image acquired by the image acquisition unit 61 from the imaging unit 11, bright pixels are lined up in the spread direction DY of the counting object T. Therefore, the counting device 1 takes advantage of the fact that bright pixels are lined up in the spread direction DY of the counting object T, and averages over a rectangular region that is long in the spread direction of the counting object T and shorter in the line-up direction than the thickness of the counting object T.

Figure 7 shows image IMG2, which is an example of an image acquired by image acquisition unit 61 from imaging unit 11. When image acquisition unit 61 actually acquires an image, as shown in area A1 of image IMG2, dust attached to the counting objects T may be captured. Also, as shown in area A2 of image IMG2, depending on the stacking state of the counting objects T, the boundaries may not be clearly visible.

By averaging over a rectangular area, the counting device 1 can suppress the effects of dust as shown in area A1 in FIG. 7. In addition, the counting device 1 can emphasize the difference in brightness between pixels in the line-up direction when the boundary is not clearly visible as shown in area A2 in FIG. 7.

Figure 5 is a flowchart showing the details of the process (averaging process) in step S30. When the averaging process of step S30 is started, a rectangular area is set in the preprocessed image at the edge of the target area to be processed, specifically, at the edge of the arrangement direction DX of the counting objects T (step S31, process of setting a rectangular area at the edge of the target area (setting process, setting process, setting procedure)).

Figure 8 is a diagram explaining rectangular areas. For image IMG3, a rectangular area WA1 is set at the end of the alignment direction DX.

It is desirable that the length of the rectangular area in the arrangement direction DX be less than one-third the thickness of the counting object T. For example, if the thickness of the counting object T is 140 micrometers and one pixel is 10 micrometers, the width in the arrangement direction DX is set to four pixels, or a width equivalent to 40 micrometers.

In addition, since the length of the rectangular region in the spreading direction DY is averaged over the spreading direction of the counting objects T, it is desirable for the length to be longer than the arrangement direction DX. For example, it is desirable for the length of the rectangular region in the spreading direction DY to be 10 times or more the length of the arrangement direction DX.

The length of the arrangement direction DX and the spread direction DY of the rectangular regions may be set by the user. The region to be averaged is not limited to a rectangular region. For example, if the counting object T is tilted with respect to the Y-axis direction in FIG. 1, the region may be a parallelogram with one side aligned along the direction in which the end of the counting object T is tilted.

Next, the average value of the pixel values in the rectangular area is calculated (step S32, a process for calculating the average value of the pixel values in a rectangular area (area averaging process, area averaging process, area averaging procedure)). In FIG. 8, the average pixel value is calculated for the pixels of image IMG3 contained in rectangular area WA1.

Next, the rectangular area is moved by one pixel in the arrangement direction DX (step S33, process of moving the rectangular area by one pixel (moving process, moving process, moving procedure)). In FIG. 8, the rectangular area is shifted (moved) in the arrangement direction DX, that is, in the direction of the arrow DA. Note that in FIG. 8, the rectangular area in the nth processing (n is an integer equal to or greater than 1) is represented as rectangular area WAn.

Then, it is determined whether processing has been completed for all of the target regions (step S34, a process for determining whether processing has been completed for all regions). If processing has been completed for all regions (Yes in step S34), the averaging process ends. If processing has not been completed for all regions (No in step S34), the process returns to step S32 and is repeated. By repeating steps S32 to S33, the average value in the rectangular region is calculated across the alignment direction DX.

Returning to FIG. 4, the area averaging unit 63 then forms an average image based on the averaging results from step S30 (step S40, a process of forming an average image using the averaging results). Specifically, the average image (see FIG. 10) is formed by extending the average value obtained in step S30 in the spreading direction DY.

Next, the average image is converted into a binary image (step S50, a process for converting the average image into a binary image (binarization process, binarization process, binarization procedure)). Specifically, the binarization unit 64 binarizes the average image formed by the area averaging unit 63 to form a binary image. The binarization means may be, for example, adaptive binarization or binarization using a specific threshold value.

Next, the number of stripes is counted from the binary image (step S60, process of counting the number of stripes from a binary image (binary counting process, binary counting process, binary counting procedure)). Specifically, the counting unit 65 counts the number of stripes that correspond to the edges of the counting objects T using the binary image formed by the binarization unit 64. The binary image becomes an image with stripes in the arrangement direction DX. The stripes correspond to the edges of the counting objects T. Therefore, the number of counting objects T is counted by counting the number of stripes arranged in the arrangement direction DX.

Next, the number of counting objects T is calculated from the result of counting the number of stripes (step S70, a process of calculating the number of counting objects from the counting result). Specifically, the counting unit 65 regards the counted number of stripes as the number of counting objects T.

In the above process, a binarization process is performed when counting the counting objects T, but the number may be counted by finding the inflection point of the average result after step S30 without performing the binarization process. As shown by way of examples of steps S40, S50, S60, and S70, the process of counting the number of counting objects T using the average value found in the averaging process of step S30 is sometimes called a counting process (counting process, counting procedure).

The control unit 15 may, for example, display the number of counting objects T on the display operation unit 21. The control unit 15 may also, for example, print and output the number of counting objects T on the printing unit 22. Furthermore, the control unit 15 may, for example, output the number of counting objects T from the external input/output unit 23 to an external device.

The results of processing by the counting device 1 are shown in Figures 9 and 10. Figure 9 shows the image IMG4 acquired by the image acquisition unit 61. Figure 10 shows the average image IMG5 generated by the area averaging unit 63.

As shown in FIG. 9, image IMG4 contains linear dust AS1 with high brightness, and linear dust AS2 and dust AS3 with low brightness. The average image IMG5 can remove the effects of dust AS1, dust AS2, and dust AS3 by averaging them over a rectangular area.

Note that the Y-axis direction and the spread direction DY are first directions, the X-axis direction and the alignment direction DX are second directions, and the control unit 15 is an example of a processing unit.

[Action and Effects]
According to the counting device of this embodiment, the number of objects T to be counted can be counted with high accuracy from an image capturing the end portion TF of the objects T to be counted.

For example, in the case of pouch bags equipped with zippers, it is difficult to align the pouch bags neatly due to the zipper. If the pouch bags are not neatly aligned, the color and brightness of the end faces of the pouch bags, which are the counting objects T, will change, making the counting process difficult.

The counting device according to this embodiment can accurately count the counting objects T even if the color or brightness of the edges of the counting objects T changes, by averaging over a wide rectangular area in the direction of the spread of the counting objects T.

Furthermore, counting methods that use wind to turn the counting objects one by one, or that use a feed mechanism to physically turn the counting objects, can cause the counting objects to become damaged or wrinkled, which can lead to the production of defective products. In contrast, the counting device of this embodiment counts the counting objects using a non-contact method without using wind or a feed mechanism, preventing the production of defective products that accompany counting.

The embodiments disclosed herein should be considered in all respects as illustrative and not restrictive. The above-described embodiments may be omitted, substituted, or modified in various ways without departing from the scope and spirit of the appended claims.

1 Counting device 10 Housing 10a Glass window 10b Placement surface 10c Placement section 11 Imaging section 15 Control section 63 Area averaging section 64 Binarization section 65 Counting section DX Arrangement direction DY Spreading direction T Counting object TF Ends WA1, WAn Rectangular area

Claims (8)

a processing unit that processes an image of a plurality of counting objects, each of whose ends is aligned to extend in a first direction and arranged in a second direction intersecting the first direction, captured from the end side, to count the number of the counting objects;
The processing unit includes:
an averaging process of calculating an average value of pixel values in a region extending in the first direction and the second direction in the captured image, while moving the region in the second direction;
a counting process of counting the number of the count objects using the average value;
Execute
Counting device. The processing unit, in the averaging process,
executing a setting process for setting the region to one end of the captured image in the second direction;
A region averaging process for calculating the average value of the pixel values in the set region;
a moving process of moving the region in the second direction in the captured image; and repeatedly executing the moving process to calculate the average value in the second direction.
2. The counting device according to claim 1. A length of the region in the first direction is longer than a length of the region in the second direction.
3. A counting device according to claim 1 or 2. The length of the region in the first direction is 10 times or more the length of the region in the second direction.
4. The counting device according to claim 3. The processing unit, in the counting process,
a binarization process for binarizing the image using the average value;
a binary counting process for counting the number of the count objects using the binarized result;
Execute
A counting device according to any one of claims 1 to 4. a placement section having a placement surface on which the ends of the plurality of counting objects are aligned to extend in the first direction and are placed side by side in the second direction;
An imaging unit that captures images of the plurality of counting objects from the end side to generate the captured images,
6. A counting device according to any one of claims 1 to 5. A counting method for counting a number of count objects, each of whose ends is aligned to extend in a first direction and arranged in a second direction intersecting the first direction, by processing an image captured from a side of the end of the count object, the counting method comprising:
an averaging step of calculating an average value of pixel values in a region extending in the first direction and the second direction in the captured image, while moving the region in the second direction;
A counting step of counting the number of the count objects using the average value.
Counting method. On the computer,
an imaging step of acquiring an image of a plurality of counting objects, each of whose ends is aligned to extend in a first direction and arranged in a second direction intersecting the first direction, by imaging the counting objects from the side of the ends;
an averaging step of calculating an average value of pixel values in a region extending in the first direction and the second direction in the captured image, while moving the region in the second direction;
a counting step of counting the number of the count objects using the average value;
A program for executing the above.

Images