JPH09264715A - Fruit and vegetable shape determination method - Google Patents

Abstract

In the conventional shape determination method, the distance is detected by counting the number of pixels. Therefore, the number of diagonally positioned pixels and the number of horizontally positioned pixels are the same. Sometimes it was not possible to obtain an accurate distance. SOLUTION: A lighting device 11 for illuminating the upper surface of a tomato 13 (fruit and fruit), and a tomato 13 illuminated by this lighting device 11.
An image pickup device 12 for picking up an upper surface of the image pickup device, an image processing device 15 for extracting the contour and the center of gravity position of the tomato 13 from the image picked up by the image pickup device 12, and a predetermined angle around the center of gravity position extracted by the image processing device 15. Obtained the coordinates of the points of the contour located for each, calculates the distance between the point of each contour and the position of the center of gravity from these coordinates, configured with a shape determination device 16 to determine the shape of fruits and vegetables by the distance thus obtained To do.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fruit and vegetable shape determining method for determining the shape of fruits and vegetables in a fruit and vegetable sorting field.

[0002]

2. Description of the Related Art When selecting fruits and vegetables, factors such as color, shape, and degree of scratches are used to classify and classify them.
In particular, the quality of the shape of fruits and vegetables is an important deciding factor for grade and classification.

An example of a conventional method for determining the shape of fruits and vegetables will be described. As shown in FIG. 11, fruits and vegetables such as tomato 1
The upper part 1A of the tomato 1 is illuminated from above by a lighting device 2, the illuminated tomato 1 is imaged by an imaging device 3 installed above the center of the tomato 1, and the captured image (captured image) is binarized. The image of the tomato 1 and the background image are separated by to form the binarized screen 4 shown in FIG. 13 (b), and the center of gravity g thereof is obtained.

Further, as shown in FIG. 13 (a), a mask screen 5 having a straight line provided at a predetermined angle Θ counterclockwise with respect to one straight line provided outside the center is prepared, As shown in FIG. 13 (c), the center of the mask screen 5 and the center of gravity g of the binarized screen 4 are made to coincide with each other, and the logical product (AND) is performed between each pixel, and the result is shown in FIG. As described above, the number of pixels of the tomato 1 appearing in the straight line "" ... Is counted to obtain the distance from the center of gravity g to the contour at each of the predetermined angles Θ, and the shape is determined based on this distance.

[0005]

However, in the conventional shape determination method, the distance is detected by counting the number of pixels. Therefore, as shown in FIG. The number of pixels located may be the same,
Sometimes it was not possible to obtain an accurate distance. Therefore, the shape judgment was unreliable, and it was not possible to make a clear grade or classification. In addition, because the grades and ranks were unclear, there was a sense of distrust between producers and consumers.

Therefore, the invention according to claim 1 of the present invention has an object to provide a method of determining the shape of fruits and vegetables which can clearly support the fruits and vegetables without depending on the resolution of the number of pixels. .

[0007]

In order to achieve the above-mentioned object, the method of determining the shape of fruits and vegetables according to the first aspect of the present invention takes an image of the upper surface of the fruits and vegetables, and uses this picked-up image as the image of the fruits and vegetables and the background image. Separated, the contour and center of gravity of the fruit and vegetables are extracted from the separated image of fruits and vegetables, and the coordinates of the points of the contour located at each predetermined angle around the center of gravity of the extracted fruit are obtained, and the points of each contour from these coordinates. Is calculated, and the shape of the fruits and vegetables is determined based on the calculated distance.

According to the above method, the contours and the center of gravity of the fruits and vegetables are extracted from the picked-up image of the fruits and vegetables, and the coordinates of the points of the contours that are located at predetermined angles centering on the extracted center of gravity are obtained. Thus, the distance between each contour point and the position of the center of gravity is calculated, and the shape of the fruit or vegetable is determined based on these distances. The quality of the fruits and vegetables, or the grade, is determined by the determination of this shape.

The method of determining the shape of fruits and vegetables according to the second aspect of the present invention is
In the fruit and vegetable shape determining method according to the first aspect of the present invention, the position of the contour point at each predetermined angle is arranged on the x-axis, the distance obtained by calculation is arranged on the y-axis, and these distances are connected in order to form a distance pattern. However, the shape of fruits and vegetables is determined based on this distance pattern.

According to the above method, the distances between the points of the respective contours centered on the position of the center of gravity and the positions of the center of gravity and the position of the center of gravity are connected in sequence to form a distance pattern, and the shape of fruits and vegetables is determined by this distance pattern. It

The method of determining the shape of fruits and vegetables according to the third aspect of the present invention is
The fruit and fruit shape determining method according to the first or second invention, wherein the shape of the fruit and vegetables is determined to be circular when the maximum value and the minimum value of the calculated distances are within a predetermined value. It is characterized by.

According to the above method, when the maximum value and the minimum value of the distance are within the predetermined value, it is determined that the shape of the fruits and vegetables is circular.

[0013]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a configuration diagram of equipment using the fruit and vegetable shape determination method according to the embodiment of the present invention.

In FIG. 1, 11 is a lighting device and 12 is a tomato.
The lighting device 11 is an imaging device provided above the upper surface of the tomato 13, and uniformly illuminates the upper surface of the tomato 13. In addition, the lighting device 11
The shielding plate 14 and the shielding curtain 14A are provided so that the light other than the light of No. 2 does not enter the imaging device 12 and the tomato 13 is not irradiated with the light other than the light of the illumination device 11.

A tomato 13 imaged by the imaging device 12
The imaged image signal a of the tomato 13 is input to the image processing device 15 and further processed by the image processing device 15 from the imaged image of the tomato 13.
(Details will be described later) is input to the shape determination device 16 including a microcomputer for determining the shape of the tomato 13,
The shape determination signal u and the quality determination signal v (details will be described later) output from the shape determination device 16 are the evaluation determination device 17
Is input to the grade of tomato 13 based on this judgment signal,
Or the classification is done. Further, the tomatoes 13 are successively conveyed by the conveyor device 18 into the shielding plate 14 and the shielding curtain 14A, and are imaged by the imaging device 12.

The image processing apparatus 15 will be described in detail with reference to the block diagram of FIG. Reference numeral 21 is a threshold value setting unit for setting a threshold value x for binarizing the picked-up image signal a picked up by the image pickup device 12, and the binarization processing unit 22 is set by the threshold value setting unit 21. The captured image signal a is binarized based on the threshold value x (separated into the image of the tomato 13 and the background image).
The threshold value x is set to a value that can separate the tomato 13 and the background (the surface of the conveyor device 18).

The tomato image signal b obtained by processing in the binarization processing unit 22 is input to the contour detecting unit 23, and this contour detecting unit 23 detects the "0" and "1" of the tomato image signal b. By tracing the boundary, as shown in FIG. 3 (b), tomato 13
Contour 18 is obtained (only the pixels in the contour portion are “1”), and the contour image signal c is output to the shape determination device 16. Further, the tomato image signal b is input to the center-of-gravity position detection unit 24, and the center-of-gravity position detection unit 24 obtains the center of gravity g of the tomato 13 by obtaining the center of the "1" pixel of the tomato image signal b. The d is output to the shape determination device 16.

The shape determining device 16 will be described in detail with reference to the configuration diagram of FIG. Reference numeral 31 denotes a mask memory section. As in the conventional case, the mask memory section 31 is counterclockwise with respect to one straight line provided outside the center z as shown in FIG. A mask image in which a straight line is provided for each predetermined angle Θ is stored.

Reference numeral 32 denotes an image synthesizing unit, which is a mask memory unit.
The center z of the 31 mask images is detected as the center of gravity position of the image processing device 15.
Matches with the center of gravity g of the center of gravity position signal d output from the output unit 24.
The contour image signal c output from the contour detection unit 23.
A logical product (AND) is performed between each pixel, and the result is shown in FIG.
As shown in (c), the contour that appears in a straight line
Xy coordinates {(x₁, Y₁),
(X_Two, Y_Two), (X_Three, Y _Three) ...} is shown in FIG.
Ask.

The coordinates (x _g , y _g ) of the center of gravity g and the contour point '
The xy coordinates of '' ... Are input to the distance calculation unit 33,
In the distance calculation unit 33, the center of gravity g and each contour point "'
The distance J (j ₁ , j ₂ , j ₃ ...) With respect to is calculated and output to the distance pattern unit 34.

The distance pattern unit 34 inputs the distance J (j ₁ , j ₂ , j ₃ ...) From the distance calculation unit 33, forms the pattern of FIG. 5 and outputs it to the shape determination unit 35. The distance pattern diagram of FIG. 5 is formed by connecting each point of the distance with the position of the contour point '''... (the position number of the straight line of the mask memory) as the x-axis and the distance J as the y-axis. Is. In FIG. 5, the predetermined angle .THETA. = 10.degree.

The shape determination unit 35 determines the shape of the tomato 13 based on this distance pattern, and outputs the determination signal u to the evaluation determination device.
Output to 17. Examples of distance patterns depending on the shape of the tomato 13 are shown in FIGS.

FIG. 6 shows a distance pattern when the shape is "circular", and its features are the average value JH and the maximum value J of the distance J.
There is almost no difference between MAX and the minimum value JMIN.
FIG. 7 shows a distance pattern when the shape is “square”,
The feature is that the convex portion P and the concave portion Q are separated by an average value JH of the distance J.
, And there is almost no difference in the maximum value PMAX of each convex portion P, and there is almost no difference in the minimum value QMIN of each concave portion Q.

FIG. 8 shows a distance pattern when the shape is "elliptical", and its characteristic is that the convex portion P and the concave portion Q appear in order with the average value JH of the distance J as a boundary, and each convex portion P has The maximum value PMAX is such that the maximum value is large and the maximum value is small, and there is almost no difference in the minimum value QMIN of each recess Q.

The shape determination unit 35 determines "circular", "square", "elliptical", or "distorted" based on the characteristics of the distance pattern. The procedure of this determination will be described with reference to the flowchart of FIG.

First, the average value JH, the maximum value JMAX and the minimum value JMIN of the distance J are obtained (step -1), then the difference between the average value JH and the maximum value JMAX, and the average value JH and the minimum value J.
It is confirmed whether the difference of MIN is within a predetermined value β (see FIG. 6) (steps-2 and 3). If both are confirmed to be within β, it is determined that the shape is “circular” (step-4), the determination result is output as the shape determination signal u (step-5), and the processing is ended.

In steps -2 and 3, when the difference between the average value JH and the maximum value JMAX and the difference between the average value JH and the minimum value JMIN are not within the predetermined value β, the average value JH is used as a boundary, and Area P (p ₁ , p ₂ , p ₃ ...) And area Q (q ₁ , q ₂ , q ₃ ...) Below average value JH are obtained (step -6), and each area P, Q Maximum value PMAX (p
max1, pmax2, pmax3 ...) and the minimum value QMIX (qmix
1, qmix2, qmix3 ...) is calculated (step −
7).

Next, it is confirmed whether or not the minimum value QMIX of each area Q is within a range of a predetermined value γ (see FIGS. 7 and 8) (step-8). When it is confirmed that it is within the range of γ, it is confirmed whether all the maximum values PMAX of the region P are within the range of the predetermined value γ (step-9). Is determined (step-10), step-5 is executed and the process ends.

In step-9, all maximum values PMA
When X is not within the range of the predetermined value γ, the maximum value PMA of the area P
In X (pmax1, pmax2, pmax3 ...) Every other, that is, pmax1, pmax3 and pmax5, or pmax2
And pmax4 are confirmed to be within the range of the predetermined value γ (step-11), and if it is confirmed to be within the range of γ, it is determined that the shape is "elliptical" (step-12). ), Execute step-5 and end.

If it is not within the range of γ in step-11, and if QMIX is not within the range of the predetermined value γ in step-8, it is judged as "distorted" (step-13), and step-5 is executed. finish.

The shape determination unit 36 determines whether the shape of the tomato 13 is "circular", "square", "elliptical", or "distorted". It should be noted that it is also possible to determine distance patterns of other shapes.

With the above configuration, the image processing apparatus 15 obtains the upper contour 18 of the tomato 13 shown in FIG.
In addition, the center of gravity g is obtained, and the shape determination device 16 uses
Using the mask image shown in (a), the xy coordinates of the contour points''' ... For each predetermined angle Θ shown in FIG. 4 are obtained, and the distance J to the center of gravity g is calculated from these coordinates, and this distance is calculated. A distance pattern is formed based on J, and it is determined based on this distance pattern whether the shape of the tomato 13 is “circular”, “quadrangular”, “elliptical”, or “distorted”.

In this way, the xy coordinates of the contour points''' ... Are obtained, and the distance J from the center of gravity g is obtained from these coordinates, so that the distance J is not significantly affected by the number of divided pixels. Can be obtained accurately, and by further forming a distance pattern with this distance J, the shape of the tomato 13 can be judged from this pattern, and based on this judgment of shape, accurate and reliable evaluation of fruits and vegetables, that is, You can classify and grade. Also the grade,
By clarifying the grounds for classification, it is possible to improve the trusting relationship between producers and consumers.

Although the shape determining device 16 determines the shape of the fruits and vegetables itself, it is also possible to determine the quality of the shape of the fruits and vegetables. At this time, the distance J (j ₁ , j ₂ , j ₃ ...) Calculated by the distance calculation unit 33 is input to the pass / fail judgment unit 36.

A quality determination method for the shape of the quality determination unit 36 will be described with reference to the flowchart of FIG. 10 and the pattern diagram of the distance J in FIG. First, the distance J (j ₁ , j ₂ , j ₃ ...
)) Average value JH is calculated (step -1), and the average value is added with the allowable value α (threshold value) J + (see FIG. 5).
Check if there is a contour point whose distance J exceeds (Step-
2) If yes, it is determined that there is a protrusion A (step-3), the quality determination signal v of the shape "defective" is output (step-4), and the process ends.

In step-2, when there is no contour point whose distance J exceeds the threshold value J +, the average value JH is set to the allowable value α.
It is confirmed whether or not there is a contour point with a distance J less than the value (threshold value) J- (see FIG. 5) obtained by subtracting (step-5), and if there is, it is judged that there is a depression B (step-6). ), The above step-4 is executed, and if there is not, a quality determination signal v of the shape “good” is output (step-7), and the process is ended.

With the above structure, the protrusion A or the depression B of the tomato 13 is determined based on the distance J between each contour point and the center of gravity g.
If the presence or absence of the projection A or the depression B is present, the quality determination signal v of "defective" is evaluated. If the projection A or the depression B is not present, the quality determination signal v of "good" is evaluated. Output to the device 17.

Thus, the accurate distance J between each contour point and the center of gravity g is obtained, and the tomato 13 is calculated based on these distances J.
By determining whether the shape of the tomato is good or bad, it is possible to accurately and reliably evaluate the tomato 13, that is, to classify or classify the tomatoes. In addition, by clarifying the grounds for classifying and classifying, it is possible to improve the relationship of trust between the producer side and the consumer side.

In the above embodiment, only the quality of the tomato 13 is judged, but the invention is not limited to the tomato 13 and can be applied to other fruits and vegetables.

[0040]

As described above, according to the first aspect of the present invention, the contour and the center of gravity of the fruit and vegetables are extracted from the picked-up image of the fruits and vegetables, and the contours of the contours that are positioned at predetermined angles with the extracted center of gravity as the center are extracted. By obtaining the coordinates of the points and obtaining the distance between the point of each contour and the position of the center of gravity from these coordinates, it is possible to obtain the accurate distance without being affected by the resolution of the number of pixels. By determining the shape, accurate and reliable evaluation of fruits and vegetables, that is, grading and ranking can be performed. In addition, by clarifying the grounds for classifying and classifying, it is possible to improve the relationship of trust between the producer side and the consumer side.

According to the second aspect of the invention, the distance pattern is formed by sequentially connecting the distances between the points of each contour located at each angle centered on the center of gravity position and the center of gravity position.
The shape of the fruits and vegetables can be determined, and the results of this determination enable reliable evaluation of the fruits and vegetables, that is, classification and classification. In addition, by clarifying the grounds for classifying and classifying, it is possible to improve the relationship of trust between the producer side and the consumer side.

Further, according to the third aspect of the present invention, when the maximum value and the minimum value of the distance are within the predetermined value, it is possible to determine the shape of the fruit or vegetables as a circle. Based on this determination result, it is possible to perform reliable evaluation of fruits and vegetables, that is, to classify and classify them.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a fruit and vegetable shape determining facility using a fruit and vegetable shape determining method according to an embodiment of the present invention.

FIG. 2 is a block diagram of the same fruit and vegetable shape determination equipment.

FIG. 3 is a picked-up image diagram for explaining the operation of the same fruit and vegetable shape determination equipment.

FIG. 4 is a coordinate diagram for explaining the operation of the fruit and vegetable shape determining equipment.

FIG. 5 is a distance pattern diagram formed in the fruit and vegetable shape determining equipment.

FIG. 6 is a distance pattern diagram formed when the shape of fruits and vegetables is “circular” in the fruit and vegetables shape determining device.

FIG. 7 is a distance pattern diagram formed when the shape of fruits and vegetables is “square” in the same fruit and fruits shape determining device.

FIG. 8 is a distance pattern diagram formed when the shape of fruits and vegetables is “elliptical” in the same fruit and fruits shape determining device.

FIG. 9 is a flowchart of a shape determination device shape determination unit of the same fruit and vegetable shape determination facility.

FIG. 10 is a flowchart of a shape determination device quality determination unit of the same fruit and vegetable shape determination facility.

[Fig. 11] Fig. 11 is a configuration diagram of a conventional fruit and vegetable shape determination facility.

[Fig. 12] Fig. 12 is a captured image diagram of a main part for explaining a problem of a conventional fruit and vegetable shape determination facility.

[Fig. 13] Fig. 13 is a captured image diagram for explaining the operation of the conventional fruit and vegetable shape determination facility.

[Explanation of symbols]

 11 Illumination device (illumination device) 12 Imaging device 13 Tomato (fruit and fruit) 14 Shield 15 Image processing device 16 Shape determination device 17 Evaluation determination device 21 Threshold setting unit 22 Binarization processing unit 23 Contour detection unit 24 Center of gravity position detection Part 31 Mask memory part 32 Image synthesizing part 33 Distance calculating part 34 Distance pattern part 35 Shape judging part 36 Good / bad judgment part a Captured image signal c Contour detection signal d Center of gravity position signal u Shape judgment signal v Good / bad judgment signal

Claims (3)

[Claims] 1. An upper surface of a fruit or vegetable is imaged, the imaged image is separated into an image of the fruit and vegetable and an image of a background, and the contour and barycentric position of the fruit or vegetable are extracted from the separated image of the fruit or vegetable. Obtaining the coordinates of the points of the contour located at each predetermined angle centered on the position, calculating the distance between the point of each contour and the position of the center of gravity from these coordinates, and determining the shape of fruits and vegetables based on these obtained distances. A shape determination method for fruits and vegetables characterized by: 2. The fruit and vegetable shape determining method according to claim 1, wherein the position of the contour point at each predetermined angle is arranged on the x-axis, and the distance calculated is arranged on the y-axis, and these distances are connected in order. A feature is that a distance pattern is formed, and the shape of the fruit and vegetables is determined based on this distance pattern. 3. The fruit and fruit shape determination method according to claim 1, wherein when the maximum value and the minimum value of the distance calculated by the calculation are within a predetermined value, the shape of the fruit and vegetables is a circle. It is characterized by judging.