JPH11101618A - Fruits and vegetables inspection device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a fruits and vegetables inspection device that can effectively discriminate defectiveness such as outline unevenness and distortion of an article to be inspected. SOLUTION: This inspection device is provided with a CCD camera 8 for photographing articles M to be inspected, and a discriminating means 10 for discriminating defectiveness of the article M on the basis of the image information of the CCD camera 8. The discriminating means 10 obtains information indicating the outline of the material M on the basis of the image information of the CCD camera 8, obtains a change state along the above-mentioned outline on the distance from the center-of-gravity position of the outline to a point positioned on the outline and discriminates whether or not the outline shape of the article M to be inspected is abnormal on the basis of the change state of the distance.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention comprises an image pickup means for picking up an object to be inspected such as, for example, oranges and tomatoes, and a judging means for judging a defect of the object to be inspected based on image information of the image pickup means. The present invention relates to an apparatus for inspecting fruit and vegetables.

[0002]

2. Description of the Related Art In the above-mentioned inspection apparatus for fruits and vegetables, conventionally, as shown in, for example, Japanese Patent Application Laid-Open No. 62-102874, an inspection object is inspected based on image information of the inspection object imaged by an imaging means. In some cases, a maximum diameter (width or height) in a predetermined direction, an area, and the like in an external shape of an object are determined. In addition, based on the maximum diameter and area of the test object in the predetermined direction obtained in this way, for example, the data is compared with preset reference data to classify the test object. ing.

[0003]

In the above-mentioned conventional configuration, since the maximum diameter and area in the predetermined direction in the external shape of the inspection object are obtained, the small diameter of the inspection object is less than the reference minimum value. Although it is possible to determine an object as a defect, for example, assuming a tangerine or the like as an example of the inspection object, irregularities on the outer surface due to a defect such as a floating skin that creates a gap between the epidermis and the contents. Or the outer shape may be distorted, but in the above conventional configuration,
No such defect could be determined, and there was still room for improvement in this regard.

The present invention has been made in view of such a point, and an object of the present invention is to provide a fruit and vegetable product that can effectively determine even a defect such as unevenness or distortion of the outer shape of a test object. It is to provide an inspection device.

[0005]

According to a first aspect of the present invention, there is provided an image pickup means for picking up an image of an object to be inspected, and a judging means for judging a defect of the object to be inspected based on image information of the image pickup means. In a device for inspecting fruits and vegetables provided with means,
The determination means is based on image information of the imaging means,
While obtaining information representing the contour of the object to be inspected, a change state in a direction along the contour with respect to the distance from the center of gravity of the contour to a point located on the contour is determined, and based on the change state of the distance. Thus, it is configured to determine whether or not the external shape of the inspection object is abnormal.

First, based on the image information on the object to be inspected by the imaging means, information representing the outline of the object to be inspected is obtained, and the center of gravity of the outline is obtained. Next, a change state in a direction along the contour of the distance from the center of gravity of the contour to a point located on the contour is determined. For example, if the change in the distance from the position of the center of gravity to a point located on the contour is small, it can be determined that there is no abnormality in the outer shape because the outer shape of the inspection object is a smooth arc. When the distance alternates between a long state and a short state, it can be determined that the external shape of the object to be inspected has irregularities and distortion. It can be determined that the shape is defective.

Therefore, it has been possible to provide an apparatus for inspecting fruits and vegetables, which can effectively determine defects such as unevenness and distortion of the outer shape of the inspection object.

According to a second aspect of the present invention, in the first aspect, the discriminating means includes an object to be inspected obtained by a Fourier transform of a function indicating a change state of a distance from the position of the center of gravity to the contour. It is configured to determine whether or not the external shape of the inspection object is abnormal based on the degree of distortion of the external shape.

By the Fourier transform of the function indicating the change state of the distance, for example, the degree of distortion of the external shape of the inspection object can be determined based on which frequency component exists in the Fourier series.

According to a third aspect of the present invention, in claim 1 or 2, a rotation operation means for rotating the inspection object is provided, and the imaging means controls the rotation of the inspection object at a set timing. The image information is obtained by imaging a plurality of times.

Therefore, the object to be inspected is rotated plural times and images are taken a plurality of times at the set timing. The defect of the shape can be properly determined.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, an apparatus for inspecting fruit and vegetables according to the present invention will be described with reference to the drawings. FIG. 1 shows a schematic configuration of a fruit and vegetable inspection device for automatically sorting and sorting by grade while transporting an inspection object such as oranges as an example of fruit and vegetables. The inspection device is configured such that a harvested inspection object M (orange) is appropriately carried in, and is sequentially transported at a predetermined interval by the transport conveyor 1, and is transported by the transport unit 2 and the transport conveyor 1. And a selecting unit 3 that outputs information for selection based on image processing of the image obtained by imaging, and a conveyed object based on the information of the selecting unit 3. A sorting unit 4 for sorting the inspection object M is provided.

The carry-in section 2 allows the operator to manually place, for example, the inspected objects M that have been harvested and loaded on a container on the conveyor 1 by hand. The inspection object M mounted thereon is configured to be conveyed while being conveyed while being conveyed while being conveyed while being lined up in a line along the conveyance direction. Note that a configuration in which the transfer to the transport conveyor 1 is automatically performed by a robot or the like may be employed. Although not described in detail, the sorting unit 4 divides the conveyed inspection object M into a collection unit provided for a plurality of preset grades based on information output from the sorting unit 3. It is configured to sort and collect. As a configuration for sorting, for example, a plurality of sorting conveyors are sequentially arranged below the conveyor 1 in a direction orthogonal to the conveying direction,
The inspection object M of the corresponding grade is configured to be transferred from the transport conveyor 1 onto the sorting conveyor by a push operation tool or a robot hand operated by an actuator.

As shown in FIG. 2, the transport conveyor 1 has endless rotary chains 5 and 6 rotatably wound along the transport direction on both sides in the transport width direction. A roller conveyor system in which a large number of rollers 7 are rotatably and pivotally connected around an axis along the conveyance width direction at predetermined intervals over the rotation chains 5 and 6, and the object M to be inspected It is configured to be conveyed while being placed and supported over the rollers 7.

The sorting section 3 includes a color CCD camera 8 of a three-plate type as a color imaging means for imaging the inspection object M being conveyed by the conveyor 1, and a color CC.
An illuminating unit 9 for illuminating the inspection object M to be imaged by the D camera 8; a determination unit 10 for determining a defective portion of the inspection object M based on image information of the color CCD camera 8; The discriminating means 10 is provided with setting means 11 for artificially setting reference information for selection. The illumination unit 9 includes a color CCD camera 8.
Is configured to be indirect illumination so as not to generate reflected light of strong light in a specific direction with respect to the inspection object M at the time of imaging by. That is, as shown in FIGS. 3, 4, and 5, the color CCD camera 8 forms a reflective wall surface 12 by applying a reflective paint to the inner wall,
A curved reflective plate 15 is attached to the inside of a housing-type reflective cover 13 located above the camera via a support member 14 at a peripheral portion (four locations) at a predetermined distance from the color CCD camera 8. The provided illumination light source 16 is supported via a support arm 17. With this configuration, as shown in FIG. 4, the illumination light from the light source 16 is reflected directly on the reflecting wall surface 12 formed on the upper peripheral wall without directly reaching the inspection object M below. The light reflected by the curved reflective plate 15 and further reflected by the upper peripheral wall is synthesized, and the reflected and diffused light irradiates the inspection object M located below the cover 13, so that the uniform and soft indirect light is applied. Lighting. As a result, there is no disadvantage such as obtaining imaging information as erroneous information due to strong reflected light in a specific direction, and an accurate inspection object M is obtained.
Can be taken.

In the conveyor 1, the rollers 7 rotate in the same direction in the same direction at the same peripheral speed in the area to be photographed by the color CCD camera 8, thereby rotating the object to be inspected. That is, as shown in FIGS. 3 and 4, in the photographing target area, the sliding contact as the rotary operation means for slidingly contacting each roller 7 from below and forcibly rotating each roller 7 by a frictional action. A plate 18 is provided in a fixed position. Thus, the color C
The CD camera 8 can obtain not only the image of the inspection object M from one direction but also imaging information from a plurality of locations in the circumferential direction. In addition, the photographing of the inspection object M by the color CCD camera 8 is performed not only directly from above but also through the reflecting mirrors 19 arranged on both sides in the transport direction as shown in FIG. The imaging information of both sides can also be obtained.

The field of view (imaging area) of the object M to be inspected by the color CCD camera 8 is plural (for example, five).
Are simultaneously imaged, and one inspected object M passes through the range of the imaging region for a set time (for example, about 1 second) and the inspected object M rotates at least about one rotation. The transport speed of the transport conveyor 1, the diameter of the roller 7, the relative positional relationship of the color CCD camera 8, and the like are set. The color CCD camera 8 is provided with a high-speed electronic shutter to obtain a still image of the inspecting object M that is moving continuously. The image capturing process is performed so that a plurality of divided images obtained by dividing the entire circumference of the inspection object M into a plurality can be obtained.

The discriminating means 10 and the setting means 11 are constituted by using a so-called personal computer. As shown in FIG. 6, the three primary color image signals (R, G, B), an image processing unit 20 for performing image processing, a central processing unit (CPU) 21 for processing an image-processed signal, and a control program to be described later for the central processing unit 21 are stored. A memory (ROM) 22, a memory (RAM) 23 for temporarily storing data, a key input device 24 constituting the setting means 11, an input / output device (I / O) for performing signal processing for input / output with respect to the key input device 24 and the sorting unit 4. / F) 25 and the like.

Based on the image captured by the color CCD camera 8, the discriminating means 10 includes saturation information indicating the saturation of the inspection object M, brightness information indicating the brightness of the inspection object M, and lightness information indicating the brightness of the inspection object M. Of the inspection object M based on the chroma information, the lightness information, and the chromaticity information.
Of the color CCD camera 8
Based on the captured image, the outer shape information of the inspection object M, for example, the maximum dimension and the distortion of the outer shape are determined. Then, based on the ratio of the area of the portion determined to be defective to the entire area of the inspected object M on the captured image and the outer shape information and the like, the inspected object M Is configured to determine the grade of When a defective portion of the inspection object M is determined, if the defective portion satisfies a specific condition, the defective portion is excluded from the defective portion.

Next, the control operation of the determination means 10 will be described with reference to the control flowcharts shown in FIGS.

As shown in FIG. 7, the color CCD camera 8
, The three primary color image signals (R, G, B) are input, and based on the three primary color image signals (R, G, B), the image signals are converted into chromaticity (H) and saturation ( S), an HSI conversion process of converting the information into information representing each information of the brightness (I) is executed (steps 1 and 2). Specifically, three primary color image signals (R, G, B) output from the color CCD camera 8
For the chromaticity (H), saturation (S), and lightness (I) by performing arithmetic processing based on the arithmetic expressions of the following [Equation 1] to [Equation 3]. Will be.

[0022]

## EQU1 ## I = 0.3R + 0.59G + 0.11B

[0023]

H = tan ^-1 (C1 / C2) (where C1 = RI, C2 = BI)

[0024]

S = √ (C1 ² + C2 ² ) (where C1 = RI, C2 = BI)

Next, based on the HSI conversion output information, an area of the inspection object M in the image is extracted, for example, by comparing each signal with a preset threshold value (step 3). . Then, the inspection object M to be extracted
Based on the image information, the outer shape determination process of the inspection object M is executed (step 4). That is, as shown in FIGS. 8 and 11, the contour L of the outer shape of the inspection object M is extracted, and the center of gravity G of the contour L is calculated, and the point located on the contour L from the center of gravity G is calculated. A change state in the direction along the contour L with respect to the distance to is determined by Fourier transform, and based on the change state of the distance, whether the external shape of the inspection object M is abnormal (whether there is distortion) or not. Is determined (steps 10 to 13).

To further explain, as shown in FIG. 12, the distance from the position of the center of gravity G to a point located on the contour L in a direction along the contour L is changed as shown in FIG. It can be expressed as a function of the angle θ between the line connecting the contour from the position of the center of gravity and the reference line. At this time,
If the surface of the inspection object has a smooth arc shape, FIG.
As shown in FIG. 12, the above function also becomes a smooth curve, and when its Fourier transform is obtained, only the low frequency component of the spatial frequency is obtained. However, if the surface of the inspection object has irregularities, FIG. ), The function becomes a sharply changing curve, and the Fourier transform contains many high frequency components of the spatial frequency f. Therefore, by determining the state of such a high frequency component, it is possible to determine the distortion of the external shape of the inspection object.

The maximum diameter of the inspection object M is determined based on the information of the contour L (step 14). Although a plurality of images of the inspection object actually exist on the captured image, only one image information is shown in FIG. 11 for easy understanding.

Next, based on the HSI conversion output information, a process of determining a defective portion of the inspection object M is executed (step 5). This failure determination processing will be described in detail based on the flowchart shown in FIG. First, a target pixel is set among a plurality of pixels in the region extracted as the region of the inspection object M (step 20). If the value of the saturation (S) at the pixel is smaller than the preset saturation threshold S0 and the lightness (I) is smaller than the first set value I1, the pixel is defective. Is determined (steps 21, 22, 23). That is,
When the saturation (S) is low and the lightness (I) is considerably low, the surface of the inspection object M is discolored black or stained black due to, for example, being picked up by a bird. This is because it is conceivable.

A setting range in which the value of the saturation (S) is smaller than a preset saturation threshold S0 and the brightness (I) is equal to or more than the first set value I1 and equal to or less than the second set value I2. If so, it is determined that the pixel is defective (steps 24 and 25). That is, when the saturation (S) is low and the lightness (I) is in an intermediate setting range, for example, when the surface is in a “scab” state due to a disease or the like, or water rot This is because it is possible that the state is in the state. Even if the value of the saturation (S) is smaller than a preset saturation threshold S0,
When the lightness (I) is larger than the second set value I2, for example, a medicine or the like may remain on the surface of the inspection object M, or a decrease in the saturation due to a variation in illumination may occur. In such a case, it is not determined that there is a defect.
If it is not determined to be defective, it is determined to be normal,
Such a determination operation is performed for all the pixels in the extracted area (steps 26, 27, 28).

When the location determined to be defective by the above-described processing satisfies a specific condition, specifically, the low-saturation portion is substantially circular and the object to be inspected is When the size is close to a predetermined ratio with respect to the outer diameter of M, such a portion is determined to be a portion referred to as a scab or a navel of the inspection object M (fruits and vegetables). Such points are excluded from the defect (step 29,
30).

Next, the average value of the saturation of all the pixels (excluding the above-mentioned part of the bar and the navel) in the region of the inspection object is obtained by calculation, and the average saturation and the preset standard saturation are calculated. Are compared (step 29). This standard saturation is
It is set as follows. That is, the chromaticity (H) and the saturation (S) generally have a correlation. For example, in the case of chromaticity of a bright color tone, the saturation tends to increase. Therefore, such a general correlation between the chromaticity and the saturation of the inspection object M (for example, orange) to be inspected is measured in advance, and the result obtained by the color CCD camera 8 is used to determine the actual correlation. The saturation having the above correlation is set as the standard saturation corresponding to the measured chromaticity, and this standard saturation and the average value are calculated for all the pixels of the inspected object M of the actually measured saturation. Compare.

If the average saturation is lower than the standard saturation by a set amount or more, the surface of the inspected object M may be, for example, wrinkled, so that it is determined as a defect (step 31, step 31). 32).

Next, on the basis of the results of the above-described defective part determination processing and outer shape determination processing, the inspection object M
Is determined (step 6). Specifically, FIG.
As shown in the figure, while determining the ratio of the total area of the portion determined to be defective with respect to the entire area of the inspection object M on the captured image, such area ratio, information on poor wrinkling, and the outer diameter dimension and The degree of distortion of the outer shape is compared with a separately set reference value to comprehensively determine the grade of the inspection object M (steps 40 and 41). Then, the operation of the sorting unit 44 is controlled so as to correspond to the determined grade, and a sorting process for sorting the inspection object M into a plurality of grades is executed (Step 7).

[Another Embodiment] (1) In the above embodiment, the degree of distortion of the external shape of the object is determined by Fourier transform of a function indicating a change state of the distance from the position of the center of gravity of the object to the contour. However, the present invention is not limited to such a configuration, and may be implemented using other calculation methods.

(2) In the above embodiment, a three-chip color CCD camera 8 is used as an image pickup means. However, a single-chip color CCD camera 8 may be used, or an image pickup tube-type camera may be used. Alternatively, a monochrome image may be obtained.

(3) In the above embodiment, the inspection object is rotated and the inspection object is imaged a plurality of times at the set timing. However, the inspection object is not rotated and the imaging is performed from a predetermined direction. Alternatively, a plurality of cameras may be installed to obtain a plurality of captured images from different directions.

(4) In the above embodiment, the conveyor 1
As described above, the inspection object M is rotated by using the roller 7 conveyor. However, the invention is not limited to such a configuration. The inspection object M may be conveyed without being rotated. Instead of the roller 7 conveyor,
You may implement by various conveyance forms, such as a belt conveyor and a bucket conveyor.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration.

FIG. 2 is a diagram showing a transfer state by a transfer conveyor.

FIG. 3 is a side view showing a configuration of a lighting unit.

FIG. 4 is a diagram showing an illumination state;

FIG. 5 is a plan view showing a configuration of a lighting unit.

FIG. 6 is a block diagram of a sorting unit.

FIG. 7 is a flowchart of a control operation.

FIG. 8 is a flowchart of a control operation.

FIG. 9 is a flowchart of a control operation.

FIG. 10 is a flowchart of a control operation.

FIG. 11 is a diagram illustrating an operation of a determination process.

FIG. 12 is a diagram illustrating an operation of a determination process.

[Explanation of symbols]

 Reference Signs List 8 imaging means 10 discriminating means 18 rotation operation means H chromaticity I lightness S saturation M inspection object

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Yoshiyuki Katayama 64 Ishizukita-cho, Sakai City, Osaka Prefecture Inside the Kubota Sakai Plant (72) Inventor Kenichi Ikeura 64 Ishizukita-cho, Sakai City, Osaka Prefecture Kubota Sakai Manufacturing Co., Ltd. In-house (72) Inventor Kazuyuki Okada 64 Ishizu-Kitacho, Sakai City, Osaka Prefecture Inside Kubota Sakai Plant

Claims (3)

[Claims] 1. An apparatus for inspecting fruits and vegetables, comprising: imaging means for imaging an inspection object; and determination means for determining a defect of the inspection object based on image information of the imaging means. The determining unit obtains information representing the contour of the inspection object based on the image information of the imaging unit, and determines a direction along the contour with respect to a distance from a center of gravity of the contour to a point located on the contour. And an apparatus for inspecting fruit and vegetables configured to determine whether or not the external shape of the object to be inspected is abnormal based on the state of change of the distance. 2. The method according to claim 1, wherein the determining unit is configured to determine a contour shape of the test object based on a degree of distortion of the contour shape of the test object obtained by Fourier transform of a function indicating a change state of a distance from the position of the center of gravity to the contour. The fruit and vegetable inspection apparatus according to claim 1, wherein the apparatus is configured to determine whether or not is abnormal. 3. A rotating operation means for rotating the object to be inspected is provided, and the imaging means is configured to image the object to be rotated a plurality of times at a set timing to obtain the image information. The fruit and vegetable inspection device according to claim 1.