JP2001116534A - Circularity deciding method, circularity arithmetic unit, and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a circularity deciding method for deciding the shape of a spherical object more accurately without making variation in the circumferential length of the object greatly affect a value of circularity, a circularity arithmetic unit which is used to implement its circularity operation, and a computer- readable recording medium where a program for the circularity operation is recorded. SOLUTION: An object image is extracted (S2) from a picked-up image of an object S by binarizing the picked-up image and the gravity center position of the extracted object image is calculated (S3); and an outline image is extracted (S4) from the extracted object image, circularity is calculated (S5) from the distance between the gravity center position and outline, and the shape of the object S is decided according to the arithmetic result.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention calculates the roundness of an object based on an image obtained by imaging the object such as a spherical fruit and vegetable, and determines the shape of the object based on the calculation result. The present invention relates to a roundness determination method, a roundness calculation device used for performing the roundness calculation, and a computer-readable recording medium storing a roundness calculation program.

[0002]

2. Description of the Related Art For example, in a sorting place where a plurality of fruits and vegetables are sorted according to a predetermined grade, the roundness which is one element of the grade determination is calculated as follows. A plurality of spherical fruits and vegetables such as tangerines or persimmons are placed on a conveyor in a row at a predetermined interval.
At an appropriate position in the conveying direction of the conveyor, an imaging device for imaging fruits and vegetables, and a light emitting device including a straight tube lamp, a reflector, and the like are provided facing the conveying region of the conveyor, and the imaging device performs the imaging of the imaging device. Each time a fruit or vegetable is transported to the area, it is imaged and the resulting image is provided to a shape determination device via an analog / digital converter.

[0003] The shape judging device takes in and stores an image given from an image pickup device, and binarizes the image to extract a binarized image of a target fruit and vegetable. The shape judging device calculates the area and perimeter of the fruits and vegetables from the extracted binarized image and, based on the calculated area and perimeter of the fruits and vegetables, calculates the roundness of the fruits and vegetables by the following equation to obtain the roundness. calculate.

[0004]

(Equation 1)

The shape judging device judges that the closer the calculated circularity value is to "1", the closer the shape of the fruits and vegetables is to a so-called "roundness".

[0006]

By the way, in general,
Since there are relatively many irregularities on the surface of fruits and vegetables, the perimeter is calculated as a large value even if the whole shape is actually close to a perfect circle. .

However, in the roundness calculation of the above-mentioned equation (1), a small change in the perimeter is also squared, so that the resulting roundness value is extremely small. There was a problem.

The present invention has been made in view of such circumstances, and an image of a spherical object is binarized to extract an image of the object from the captured image, and the center of gravity of the extracted image of the object is extracted. By calculating the position, extracting a contour image from the extracted target object image, and calculating the roundness based on the position of the center of gravity and the distance between the contours, the value of the roundness with the fluctuation of the perimeter is calculated. Is not greatly influenced by the roundness determination method that enables more accurate shape determination of the object,
It is an object of the present invention to provide a roundness calculating device used for performing the roundness calculation and a computer-readable recording medium on which a program for the roundness calculation is recorded.

[0009]

According to a first aspect of the present invention, there is provided a method for determining a roundness, wherein a region including a spherical object (S) is imaged, and the image of the object is extracted by binarizing the captured image. (S
2) In the roundness determination method for calculating the roundness of the target object image based on the extraction result and determining the shape of the target object (S) based on the calculation result, the position of the center of gravity of the target image , A second step (S4) of extracting the contour of the object image, and calculating the roundness of the object image based on the position of the center of gravity and the distance between the contours. A third step (S6), and a fourth step (S6) of determining the shape of the object (S) based on the roundness.
7).

The roundness determining method according to the second invention is the roundness determining method according to the first invention, wherein the third step (S
The roundness in 6) is calculated based on the ratio of the difference between the maximum value and the minimum value of the distance to the average value of the distance.

A roundness calculating device according to a third invention (3, 6).
Extracts a target image by binarizing a captured image of a region including a spherical target (S) captured by an imager (14) (S2), and based on the extraction result, extracts the target image. A center-of-gravity position calculating means (6) for calculating a center-of-gravity position of the object image; a contour extracting means (6) for extracting a contour of the object image; Roundness calculating means for calculating the roundness of the object image based on the distance between the position of the center of gravity and the contour (6)
And characterized in that:

A roundness calculating device according to a fourth invention (3, 6).
In the roundness calculating device (3, 6) of the third invention, the roundness calculating means (6) is configured to calculate the roundness based on a ratio of a difference between a maximum value and a minimum value of the distance to an average value of the distance. It is characterized in that the roundness is calculated.

[0013] A recording medium (7, M) according to a fifth aspect of the present invention provides a computer by binarizing a captured image of an area including a spherical object (S) captured by an imager (14). Program code means (S
2) and a computer-readable recording medium storing a program for calculating roundness including program code means for calculating the roundness of the object image based on the extracted result, Computer (3
6) first program code means (S3) for calculating the position of the center of gravity of the object image, and a computer (3, 6)
The second program code means (S4) for extracting the contour of the object image and the computer (3, 6) determine the roundness of the object image based on the position of the center of gravity and the distance between the contours. Third program code means for calculating (S6)
Is recorded.

A recording medium (7, M) according to a sixth aspect of the present invention is the recording medium (7, M) according to the fifth aspect, wherein the third program code means (S6) is arranged so that the third program code means (S6) sets the distance to the average diameter of the distance. The computer (3, 6) is adapted to calculate the roundness based on the ratio of the difference between the maximum value and the minimum value.

According to the present invention, an image of a region including a spherical object is taken, the object image is extracted by binarizing the taken image, and the position of the center of gravity of the extracted object image is calculated. A configuration for extracting the outline of the extracted object image and calculating the roundness of the object based on the distance between the position of the center of gravity and the outline, more specifically, the roundness is calculated based on the average value of the distance. Since the calculation is performed based on the ratio of the difference between the maximum value and the minimum value of the distance, the value of the roundness is not extremely influenced by the fluctuation of the perimeter, and a more accurate Shape determination becomes possible. The calculation of the roundness can be performed using, for example, the following equation.

[0016]

(Equation 2)

Since the maximum and minimum diameters in the equation (2) are the same as the diameters of the circumscribed circle and the inscribed circle having the center of gravity determined from the distance between the center of gravity and the contour, the diameters are the same. It is shown as The average diameter may be, for example, an average value of the number of pixels (distance) between the position of the center of gravity and each of the pixels forming the contour, or may be configured to be calculated using a least square method or the like.

Further, as is clear from equation (2), as the difference between the maximum diameter and the minimum diameter is as small as possible and the shape is closer to a so-called sphere, the value of the roundness is "1". And the value of the roundness decreases as the difference increases.

[0019]

Embodiments of the present invention will be specifically described below with reference to the drawings. FIG. 1 is a block diagram illustrating a configuration of a shape determination device according to the present invention. In FIG. 1, reference numeral 21 denotes a conveyor. On the conveyor 21, a plurality of trays T, T,...
A spherical object S such as tangerine or persimmon is placed on the top, respectively. With such a configuration, the conveyor 21 conveys each object S together with the tray T in the arrow direction.

In the transport area of the objects S, S,..., The first imaging device 11 and the second
The imaging device 12 is arranged at an appropriate distance in the transport direction of the conveyor 21. First imaging device 11 and second imaging device 1
2 includes light-shielding housings 18 and 18 each having an opening through which the conveyor 21 is inserted.

In each housing 18, upper light projectors 15, 15 each having a straight tubular halogen lamp and a reflector for mainly projecting light to the upper part of the object S are separated by a suitable length in the conveying direction of the conveyor 21. And is arranged so as to be orthogonal to the conveying direction of the conveyor 21. In addition, these upper projectors 1
A camera 14 as an image pickup device for picking up an image of the object S is installed in the middle upper part between 5 and 15.

Slightly obliquely above the side edges of the conveyor 21, a spherical halogen lamp and a reflector 4 are provided.
Are provided at positions corresponding to the vertices of a square parallel to the conveyor 21, and are projected from both side projectors 16 to both sides of the object S. Reflecting mirrors 17, 17 for forming respective side images of the object S on the camera 14 are provided between the two pairs of side projectors 16, 16 paired in the transport direction of the conveyor 21. I have.

A rotating device 13 is disposed at a transfer position between the first image pickup device 11 and the second image pickup device 12, and the turn device 13 transfers the tray T to a predetermined position by the conveyor 21. Each time the operation is performed, the object S placed on the tray T together with the tray T is rotated by 90 ° in the direction of the white arrow. Thereby, it is possible to capture an image of a portion other than the portion of the object S facing the tray T.

When the object S is conveyed to an appropriate position in the first imaging device 11 or the second imaging device 12 by the conveyor 21, the camera 14 directly and upwardly reflects the upper portion of the object S together with the tray T. Both sides of the object S are indirectly imaged together with the tray T reflected on the mirrors 17 and 17 and the obtained image signals are supplied to analog / digital (A / D) converters 2 and 2, respectively. Each A / D converter 2 includes a camera 14
From the camera 14 on the side of the first image pickup device 11 to the first frame memory 4 of the shape determination device 3 and to the second frame signal.
The image signal from the camera 14 of the imaging device 12 is given to the second frame memory 5 and stored.

Each time an image is stored in the first frame memory 4 or the second frame memory 5, the image is given to the shape judging section 6. The shape judging section 6 incorporates the given image. An image (object image) of the object S is extracted by binarizing with a threshold value stored in advance in the memory 6a, and an object obtained by directly imaging based on the obtained binarized image. The position of the center of gravity of the object S in the upper image of S and both side images of the object S obtained by indirect imaging is specified, and the contour of the object S is extracted based on the binarized image. The shape determining unit 6 calculates the distance by counting the number of pixels between each pixel constituting the extracted contour and the pixel at the position of the center of gravity, and based on this distance, a perfect circle in each image of the object S. Calculate the degree and store the calculation result in the memory 6a.
Is configured to perform shape determination by comparing with a threshold value stored in. The determination result is output to a sorting device or the like for use.

The processing of the shape judging section 6 as described above is executed based on a computer program stored in the EPROM 7 provided in the shape judging device 3. Further, the computer program stored in the EPROM 7 is read from a recording medium M set in the external storage reading unit 8 via an external storage reading unit 8 such as a ROM drive device included in the shape determination device 3. It is also possible.

Next, the processing contents of the shape judging section 6 based on such a computer program will be described in detail.
FIG. 2 is a flowchart showing the processing content of the shape determination unit 6 of the shape determination device 3 according to the present invention, and FIG. 3 is a schematic diagram showing the content of an image in each processing of the flowchart shown in FIG. Note that the binarized image in FIG.
(B)) In the following images, for convenience of illustration, black and white are reversed from the actual colors.

First, the shape judging section 6 stores the first frame memory 4 or the second frame memory 5 shown in FIG.
An original image as shown in FIG. 3A is fetched (step 1), and the fetched original image is binarized as shown in FIG. 3B to extract a target object image from a background image including the tray T (step 2). ).

Subsequently, the shape judging section 6 sets an XY orthogonal coordinate system for the extracted object image as shown in FIG.
First moment calculation is performed using a known method based on the luminance distributions for the X and Y axes, and the coordinate value of the center of gravity of the object S is calculated from the binarized image (step 3). In addition, the shape determination unit 6 similarly performs a first derivative operation using a spatial filter or the like from the binarized image in which the XY orthogonal coordinate system is set, and obtains a contour image of the object S as shown in FIG. Is extracted (step S4). Step 3
It is needless to say that the processes 4 and 4 may be performed in the reverse order.

Next, based on the number of pixels between the coordinate value of the position of the center of gravity obtained in steps 3 and 4 and the coordinate value of each pixel constituting the outline image, the contour image centered on the position of the center of gravity is determined. Each diameter dimension corresponding to the circumscribed circle and the inscribed circle (maximum diameter,
In addition to calculating the minimum diameter, an average value (average diameter) of the diameters is calculated (step 5). Note that, as described above, the average diameter may be, for example, a mere average value of the number of pixels between the position of the center of gravity and each pixel forming the contour image,
Further, the configuration may be such that the calculation is performed using the least square method or the like.

Then, the roundness is calculated by substituting the calculated maximum diameter, the minimum diameter, and the average diameter into the above-described equation (2) (step 6), and the calculation result is compared with the threshold value stored in the memory 6a. The shape is appropriately determined by comparing (Step 7).

[0032]

FIG. 4 is a schematic diagram showing an upper image (original image) of three peach samples used in the roundness calculation in the embodiment, and FIG. 5 shows a conventional example of the sample shown in FIG. 6 is a table showing calculation results of the roundness of the method and the method of the present invention.

As can be seen from FIG. 4, the three peaches have various shapes as their upper images are shown. Now, for these peaches, the roundness calculation of the conventional method based on the formula (1) and the method of the present invention based on the formula (2) is performed, and the calculation results are calculated for each sample as (a) and (b). , And (c) are shown in FIG. 5 in association with FIG.

In the conventional method, the area is changed from (a) to 4
659 pixels, 4234 pixels, and 4695 pixels,
The perimeters are 263 pixels, 252 pixels, and 265 pixels, and it can be seen that there is no difference in the number of pixels between the original images despite their very different shapes. For this reason, the roundness calculated as a result has the same value as 0.84, respectively.

On the other hand, in the method of the present invention, the maximum diameter and the minimum diameter are 39.5 pixels, 36.5 pixels in (a), and (b)
38.5 pixels, 33.5 pixels, and (c) 42.0 pixels
Pixels are 33.5 pixels, and the value of the flat sample (c) is calculated to be significantly different from other samples. Therefore, the resulting roundness is
(A) becomes 0.92, 0.86, and 0.77, respectively, and the roundness is high in the order of (a), (b), and (c), and the result is close to the actual appearance. I understand.

It should be noted that, although the claims are provided with reference numerals for convenience of comparison with the drawings, the present invention is not limited to the configuration of the accompanying drawings.

[0037]

As described in detail above, the roundness determination method according to the present invention, the roundness calculating device used for performing the roundness calculation, and the computer readout of the roundness calculation program are recorded. In a possible recording medium, a region including a spherical target is imaged, the target image is extracted by binarizing the captured image, and the position of the center of gravity of the extracted target image is calculated. A configuration for extracting the contour of the extracted object image and calculating the roundness of the object based on the center of gravity position and the distance between the contours, more specifically, the roundness is calculated based on an average value of the distance. Since the calculation is based on the ratio of the difference between the maximum value and the minimum value of the distance, the value of the roundness is not extremely affected by the fluctuation of the perimeter, and the shape of the object is more accurate. The present invention has excellent effects, for example, the determination can be made.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a shape determination device according to the present invention.

FIG. 2 is a flowchart illustrating processing performed by a shape determination unit of the shape determination device according to the present invention.

FIG. 3 is a schematic diagram showing the contents of an image in each process of the flowchart shown in FIG. 2;

FIG. 4 is a schematic diagram showing an upper image (original image) of three peach samples used for roundness calculation in the embodiment.

FIG. 5 is a table showing calculation results of roundness of the sample shown in FIG. 4 by the conventional method and the method of the present invention.

[Explanation of symbols]

 Reference Signs List 3 shape determination device 6 shape determination unit 7 EPROM 8 external storage reading unit 14 camera M recording medium S target

Continued on the front page F-term (reference) 2F065 AA17 AA26 AA48 AA52 BB05 BB15 DD03 FF01 GG02 HH02 HH13 JJ03 JJ09 JJ26 LL12 MM02 PP15 QQ04 QQ13 QQ21 QQ23 QQ33 TT01 TT03 2G051 AA05 EA01 DA08 ED23

Claims (6)

[Claims] 1. An image including an area including a spherical object (S) is imaged, an image of the object is extracted by binarizing the captured image (S2), and a perfect circle of the image of the object is extracted based on the extraction result. A roundness determination method for calculating the degree of the object and determining the shape of the object (S) based on the calculation result;
3), a second step (S4) of extracting the contour of the object image, and a third step (S4) of calculating the roundness of the object image based on the position of the center of gravity and the distance between the contours.
6) and a fourth step (S7) of determining the shape of the object (S) based on the roundness. 2. The roundness determination according to claim 1, wherein the roundness in the third step (S6) is calculated based on a ratio of a difference between a maximum value and a minimum value of the distance to an average value of the distance. Method. 3. An object image is extracted by binarizing a captured image of a region including a spherical object (S) captured by an imager (14) (S2), and based on the extraction result, A roundness calculating device for calculating a roundness of an object image, a center-of-gravity position calculating means (6) for calculating a center-of-gravity position of the object image, and a contour extracting means (6) for extracting a contour of the object image. ) And roundness calculating means (6) for calculating the roundness of the object image based on the distance between the position of the center of gravity and the contour.
And a roundness calculation device (3,
6). 4. The roundness calculating means (6) calculates roundness based on a ratio of a difference between a maximum value and a minimum value of the distance to an average value of the distance. The roundness calculation device (3, 6) described. 5. A program code means (S2) for causing a computer to extract a target object image by binarizing a captured image of a region including the spherical target object (S) captured by the image pickup device (14). A program code means for calculating the roundness of the object image based on the extracted result; and a computer-readable recording medium storing a program for calculating the roundness. , 6), the first program code means (S3) for calculating the position of the center of gravity of the object image, and the second program code means (S4) for causing the computer (3, 6) to extract the contour of the object image. Third program code means (S6) for causing a computer (3, 6) to calculate the roundness of the object image based on the distance between the position of the center of gravity and the contour; Recording medium on which the program comprising is characterized in that it is recorded (7, M). 6. The third program code means (S6)
Calculates the circularity based on the ratio of the difference between the maximum value and the minimum value of the distance to the average diameter of the distance, using a computer (3,
6. The recording medium (7, M) according to claim 5, wherein said recording medium is adapted to be operated by (6).