JP2018146251A - Foreign matter detection system, foreign matter detection method and program thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect foreign matters included in an inspection object 1.SOLUTION: A foreign matter detection system 100 comprises: an irradiation unit 10 which irradiates the inspection object 1 with light in a band including near-infrared rays; a first imaging unit 21 which acquires first intensity value image data by receiving reflected light and/or transmitted light from the inspection object 1 due to light having a first center wavelength included in the band; a second imaging unit 22 which acquires second intensity value image data by receiving reflected light and/or transmitted light from the inspection object 1 due to light having a second center wavelength included in the band; a relative intensity calculation unit 71 which calculates relative intensities which correspond to respective measurement points on a surface of the inspection object 1 and are ratios of intensity values of pixels in the first intensity value image data and intensity values of pixels in the second intensity value image data; and a detection unit 72 which detects foreign matters included in the inspection object 1 on the basis of the relative intensities calculated by the relative intensity calculation unit 71.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a foreign matter detection system, a foreign matter detection method, and a program for detecting foreign matter mixed in a plant or a processed plant product in a nondestructive manner, easily, reliably, and efficiently.

For thorough food hygiene, for example, in a food processing factory such as a cut vegetable processing factory, a technique for nondestructively detecting and separating and removing vegetables and the like to which foreign matters such as insects are attached is required. For this reason, it is common to place inspection personnel on the processing line, detect whether or not foreign substances such as insects are attached with the naked eye, and separate and remove vegetables and the like to which foreign substances such as insects have adhered. .
For cut vegetables, it is not easy to detect the attachment site of insects or the like with the naked eye. In the future, consumption of cut vegetables is expected to increase with the increase in double-income households, single households and elderly households.

In Patent Document 1, two pixels included in a hyperspectral image obtained by imaging an inspection object by the imaging means 20 as a hyperspectral sensor that acquires a hyperspectral image are selected and compared with each other. Describes a foreign matter detection method for detecting foreign matter.
Patent Document 2 discloses that a first inspection light having a central wavelength in the near infrared region and a second inspection light having a central wavelength different from the first inspection light in the near infrared region from a surface light source to food. The first inspection light and the second inspection light are transmitted through the inspection object, and the food image is picked up by the difference image between the first image and the second image. It is described that a foreign object is acquired.

JP 2013-164338 A JP 2014-044070 A

However, the method described in Patent Document 1 is based on a hyperspectral camera, and the system itself may be expensive.
In the method described in the cited document 2, since the detection is performed using the difference in intensity, it is necessary to irradiate the sample with uniform light and alternately switch to irradiate light of different wavelengths, which is expensive in mounting. There is a fear. In Patent Document 1 and Patent Document 2, it is unclear how to separate and remove the inspection object in which foreign matter is detected.

  The present invention enables non-destructive, easy, reliable, and efficient detection of foreign substances present in plants or processed plant products, and enables easy separation and removal of plants or processed plant products containing foreign objects. It is an object of the present invention to provide a foreign object detection system, a foreign object detection method, and a program therefor.

  (1) This invention has the irradiation part (for example, the below-mentioned "irradiation part 10") which irradiates the light of the band containing near infrared rays with respect to a test object, and the 1st center wavelength contained in the said band. A first imaging unit (for example, described later) that captures the inspection object by receiving reflected light and / or transmitted light from the inspection object by the first light and obtains first intensity value image data “The first imaging unit 21”) and the reflected light and / or transmitted light from the inspection object by the second light having the second central wavelength included in the band, thereby receiving the inspection object. A second imaging unit (for example, “second imaging unit 22” described later) that captures an image to acquire second intensity value image data, and a coordinate value of the pixel coordinate system of the first imaging unit (coordinates of the image coordinate system) Value) and the coordinate value of the pixel coordinate system of the second imaging unit (the coordinate value of the image coordinate system) Based on the correspondence, the ratio between the intensity value of the pixel of the first intensity value image data and the intensity value of the pixel of the second intensity value image data corresponding to each positioning point on the surface of the inspection object Relative intensity calculating means (for example, “relative intensity calculating unit 71” described later) for calculating the relative intensity, and the relative intensity at each positioning point on the surface of the inspection object calculated by the relative intensity calculating means. The present invention relates to a foreign matter detection system (for example, “foreign matter detection system 100” to be described later) including detection means (for example, “detection unit 72” to be described later) that detects foreign matters mixed in the inspection object.

  (2) The first imaging unit (for example, “first imaging unit 21” described later) is an optical filter (for example, “bandpass optical” described later) for acquiring first light having the first central wavelength. Filter 211 ") (for example," first infrared camera 212 "described later), and the second imaging unit is an optical filter for acquiring second light having the second central wavelength. The foreign object detection system according to (1), which is a camera (for example, a “second infrared camera 222 described later) provided with (for example, a“ bandpass optical filter 221 ”described later).

  (3) The detection unit (for example, “detection unit 72” described later) is configured to measure each position of the surface of the inspection object calculated by the relative intensity calculation unit (for example, “relative intensity calculation unit 71” described later). The foreign matter detection system according to (1) or (2), wherein foreign matter mixed in the inspection object is detected when it is determined that the relative intensity at a point exceeds a preset threshold value.

  (4) The foreign matter detection system further includes a display unit (for example, “display unit 40” described later) and the display unit with the first intensity value image data with a pixel that exceeds the threshold as a predetermined color. Or the foreign substance detection system as described in (3) provided with the display control means (For example, the below-mentioned "display control part 73") superimposed and displayed on a said 2nd intensity | strength value image data.

(5) The foreign object detection system further includes:
A projection unit that projects a preset marker (for example, “projection unit 60” described later), and the marker in the vicinity of each positioning point on the surface of the inspection object that is determined that the relative intensity exceeds the threshold value A foreign matter detection system according to (3) or (4), comprising projection control means (for example, “projection control unit 74” described later).

  (6) The foreign matter detection system further includes a transport line (for example, “transport line 50” described later) for transporting the inspection object, and the projection control means (for example, “projection control unit 74” described later). Project the marker in the vicinity of each positioning point on the surface of the inspection object that is determined to have the relative intensity exceeding the threshold value, which is the inspection object conveyed by the conveyance line (5) Foreign matter detection system described in 1.

  (7) The foreign matter detection system according to any one of (1) to (6), wherein the inspection object is a cut vegetable, and the foreign matter is an insect and / or an egg of the insect attached to the cut vegetable.

  (8) The present invention provides an irradiation unit (for example, “irradiation unit 10” to be described later) that irradiates the inspection object with light in a band including near infrared rays, and a first central wavelength in the band. A first imaging unit (for example, “first” described later) that captures the inspection object by receiving reflected light and / or transmitted light from the inspection object by light of 1 imaging unit 21 ") and the reflected light and / or transmitted light from the inspection object by the second light having the second central wavelength in the band, and the inspection object is imaged to receive the first light. A foreign object detection system including a second imaging unit (for example, “second imaging unit 22” described later) that acquires the intensity value image data of 2 and a control unit (for example, “control unit 70” described later); First intensity value image data acquired by the first imaging unit. Between the coordinate value of the pixel coordinate system (the coordinate value of the image coordinate system) and the coordinate value of the pixel coordinate system of the second intensity value image data acquired by the second imaging unit (the coordinate value of the image coordinate system) On the basis of the correspondence between the intensity value of the pixel of the first intensity value image data and the intensity value of the pixel of the second intensity value image data corresponding to each positioning point on the surface of the inspection object. A relative intensity calculating step (for example, a “relative intensity calculating step” described later) for calculating a relative intensity that is a ratio, and the relative intensity at each positioning point on the surface of the inspection object calculated by the relative intensity calculating step. The present invention relates to a foreign matter detection method including a detection step (for example, a “detection step” to be described later) that detects foreign matters mixed in the inspection object.

  (9) The present invention relates to a computer program for causing a computer to function as each unit of the foreign matter detection system described in any one of (1) to (7).

  According to the foreign matter detection system, foreign matter detection method, and program therefor according to the present invention, foreign matter mixed in a plant or a processed plant product is detected nondestructively, easily, reliably, and efficiently, and foreign matter is mixed. The removed plant or plant processed product can be easily separated and removed.

It is a whole lineblock diagram of a foreign substance detection system concerning this embodiment. It is a functional block diagram of a control part concerning one embodiment of the present invention. It is a figure which shows the plane of the conveyance line. It is a figure which shows the flow of a series of processes in one Embodiment of this invention.

Hereinafter, a preferred embodiment of the foreign object detection system 100 of the present invention will be described with reference to the drawings. Here, all of the following embodiments are examples of the present invention, and the present invention is not limited thereto.
In the present invention, near infrared means light in a band including a wavelength of about 700 nm to 1400 nm. Unless otherwise specified, light having a wavelength in the vicinity of 700 nm (for example, wavelength of 680 nm) is also referred to as near infrared light for convenience.

[First Embodiment]
The inspection object 1 in the present embodiment is a cut vegetable conveyed on the conveyance line, and the foreign matter is an insect and / or an egg of the insect attached to the cut vegetable.
FIG. 1 is an overall configuration diagram of a foreign object detection system 100 according to the present embodiment. As shown in FIG. 1, the foreign object detection system 100 includes at least an irradiation unit 10, an imaging unit 20, a display unit 40, a conveyance line 50, a projection unit 60, a control unit 70, and a storage unit 80. Prepare.

<Irradiation unit 10>
The irradiation unit 10 irradiates the inspection object 1 (cut vegetables) with light in a band including near infrared rays (hereinafter referred to as “near infrared rays”). For example, the irradiating unit 10 may be composed of a plurality of point light sources using a halogen lamp (not shown) that generates broadband light including a near infrared region as a point light source.
More specifically, the irradiation unit 10 irradiates near infrared rays toward a predetermined irradiation region 51 on the transport line 50. The irradiation area 51 is set in advance so as to include a first image area captured by the first imaging unit 21 described later and a second image area captured by the second imaging unit 22. Hereinafter, an area where the first image area and the second image area overlap is referred to as an inspection area 52. As will be described later, all the inspection objects 1 (cut vegetables) are set so as to pass through the inspection region 52.
By doing so, the near infrared ray irradiated by the irradiation unit 10 reaches a substance located in the inspection region 52, for example, the surface of the transport line 50 and the surface of the inspection object 1 (cut vegetable), and reaches the reached region. The scattered light is emitted to the outside due to absorption and scattering caused by the constituent substances.
In addition, when detection using transmitted light is performed, the irradiation unit 10 is installed on the lower side of the conveyance line 50 and irradiated toward the irradiation region 51. By doing so, the near infrared rays irradiated by the irradiation unit 10 reach the substance located in the inspection region 52, for example, the back surface of the transport line 50 and the back surface of the inspection object 1 (cut vegetables), and reach the reached region. The absorption due to the constituent material is received, and the transmitted light is transmitted to the outside.

<Imaging unit 20>
The imaging unit 20 includes at least two imaging units, that is, a first imaging unit 21 and a second imaging unit 22.
For example, the first imaging unit 21 includes a bandpass optical filter 211 for acquiring first light having a first central wavelength included in the near-infrared region (hereinafter also referred to as “first near-infrared”). One infrared camera 212 may be used.
Further, the second imaging unit 22 acquires, for example, second light (hereinafter, also referred to as “second near infrared ray”) having a second center wavelength included in the near infrared region, which is different from the first near infrared ray. The second infrared camera 222 including the bandpass optical filter 221 may be used.
By doing so, the first imaging unit 21 is a substance in which the broadband light including the near infrared region including the first near infrared ray irradiated by the irradiation unit 10 is located in the inspection region 52, for example, the inspection object 1 (cut vegetable). The first intensity value image data including the intensity value of the first near infrared ray at each positioning point is received by receiving reflected light and / or transmitted light reflected and / or transmitted after reaching each positioning point on the surface of get.
Similarly, the 2nd imaging part 22 is the surface of the substance in which the broadband light containing the near infrared region including the 2nd near infrared ray irradiated by the irradiation part 10 is located in the test | inspection area | region 52, for example, the test object 1 (cut vegetable) The second intensity value image data including the intensity value of the second near infrared ray at each positioning point is acquired by receiving the reflected light and / or transmitted light reflected and / or transmitted after reaching each positioning point. .

Here, the imaging unit 20 (the first imaging unit 21 and the second imaging unit 22) is fixed so as to be arranged in parallel, for example, on a gantry (not shown). The components such as the imaging lens and the imaging element of the first imaging unit 21 and the second imaging unit 22 are preferably configured with the same characteristics.
The first imaging unit 21 and the second imaging unit 22 can acquire pixel information of each point on the surface of the inspection object 1 (cut vegetable) at a preset time interval. For example, by setting 1/30 seconds as the time interval, the first imaging unit 21 and the second imaging unit 22 can acquire 30 frames of image data (30 fps) per second. The image data measured by the first imaging unit 21 and the second imaging unit 22 can be attached with a time stamp that is a measurement time.
The time interval is not limited to 1/30 seconds. For example, it may be set arbitrarily such as 1/24 seconds, 1/12 seconds, or the like. Note that it is desirable to control the imaging timing by the control unit 70 described later so that the first imaging unit 21 and the second imaging unit 22 are time-synchronized.
By carrying out like this, in the case of carrying out the processing work by transporting the inspection object (cut vegetable) on the transportation line 50 described later, while photographing the inspection object 1 (cut vegetable) transported on the transportation line 50, The first intensity value image data and the second intensity value image data are acquired in real time, and as will be described later, foreign matter (insects and / or insect eggs, etc.) is mixed or adhered to the inspection object 1 (cut vegetables). It is possible to detect whether or not to do so.
Accordingly, the worker can inspect the inspection object 1 (cut vegetables) in which foreign matters (insects and / or insect eggs, etc.) are mixed or adhered from the inspection objects 1 (cut vegetables) conveyed on the conveyance line 50. Can be easily separated and removed.

<Image coordinate system>
The first imaging unit 21 of the pixel data P ₁ obtained by imaging the substance located in the inspection area 52, for example, each point P on the surface of the transport line 50 and the surface of the inspection object 1 (cut vegetable) by the first imaging unit 21. Coordinate values (u ₁ (P), v ₁ (P)) in the pixel coordinate system (referred to as “first pixel coordinate system”) and the second imaging unit 22 of the pixel data P ₂ imaged by the second imaging unit 22. The coordinate values (u ₂ (P), v ₂ (P)) in the pixel coordinate system (referred to as “second pixel coordinate system”) are known to those skilled in the art (for example, triangulation) This can be achieved by the principle of a stereo camera using
That is, the coordinate values of _{P 1 (u 1 (P)} , v 1 (P)) and the coordinate values of _{P 2 (u 2 (P)} , v 2 (P)) second pixel from the first pixel coordinate system and Correspondence can be made by a transformation matrix to the coordinate system and a transformation matrix (homography transformation matrix) from the second pixel coordinate system to the first pixel coordinate system.

<3D coordinate system>
Similarly, pixel coordinate values (u ₁ (P), v ₁ (P)) of the point P in the first pixel coordinate system are known to those skilled in the art by a known method described in the reference (for example, calibration of a stereo camera). (Or pixel coordinate values (u ₂ (P), v ₂ (P)) in the second pixel coordinate system and coordinate values (X (P), Y (P), Z (P)) in the three-dimensional space) Can be associated.
That is, the coordinate values of _{P 1 (u 1 (P)} , v 1 (P)) and the coordinate values in three-dimensional space first pixel coordinates and (X (P), Y ( P), Z (P)) Correspondence can be made by a transformation matrix from a system to a three-dimensional spatial coordinate system.
If the thickness of the inspection object 1 in the Z direction is sufficiently small, the pixel coordinate value in the first image coordinate system and the pixel coordinate value in the second image coordinate system are associated with each other by a homography transformation matrix, thereby substantially inspecting. A position on the object 1 can be associated. For example, when the cut vegetable mentioned as an example in this embodiment is set as the inspection object 1, the position can be associated with the homography transformation matrix.
<References>
JP 2016-86250 A,
JP 2006-67272 A,
JP-A-8-261719, etc.
http://www.ime.info.hiroshima-cu.ac.jp/~hiura/lec/iip/geometry.pdf

<Display unit 40>
The display unit 40 is configured by a display device such as a liquid crystal display, for example, and displays an image in response to an instruction from the control unit 70. More specifically, the display unit 40 includes image data captured by the first imaging unit 21 and / or the second imaging unit 22 and, as will be described later, a foreign substance (attached to the inspection object 1 (cut vegetable)). Position information of insects etc. is displayed.

<Conveying line 50>
A conveyance line conveys a detection target object (cut vegetable) at the predetermined fixed speed. The transport line 50 includes the irradiation area 51 and the inspection area 52 described above. The irradiation area 51 is set so as to include a first image area captured by the first imaging unit 21 and a second image area captured by the second imaging unit 22. The inspection area 52 is set so as to be included in an area where the first image area and the second image area overlap.
All the inspection objects 1 (cut vegetables) are arranged on the conveyance line 50 so as to pass through the inspection region 52. All inspection objects 1 (cut vegetables) receive near-infrared rays emitted from the irradiation unit 10 when passing through the inspection region 52, and reflect and / or transmit reflected light and / or transmitted light, respectively. To do. Accordingly, it is possible to detect whether or not a foreign substance (insect and / or insect egg) is mixed or adhered by the detection unit 72 described later.
In addition, a projection area 53 is provided on the transport line 50. When the inspection object 1 (cut vegetables) to which foreign matter (insects, etc.) adheres is projected into the projection area 53, a marker set in advance by the projection unit 60 described later with respect to the inspection object 1 (cut vegetables). Projected. Workers can easily discriminate inspection objects 1 (cut vegetables) in which foreign objects (insects and / or insect eggs, etc.) are mixed or adhered from inspection objects 1 (cut vegetables) conveyed on the conveyance line. And can be separated and removed.

<Projector 60>
The projection unit 60 is a projection unit corresponding to a projector, and projects a marker in a state in which a marker generated by a projection control unit 74 (to be described later) is focused on a projection region 53 set in advance on the transport line 50. . More specifically, as described above, since the thickness of the cut vegetables in the Z direction is sufficiently small, on the plane of the projection region 53 where the cut vegetables (three-dimensional objects) to which foreign substances (insects, etc.) are attached are arranged. By projecting the marker toward the position, when the detection target object (cut vegetable) to which a foreign object (such as an insect) is attached is transported to the projection region 53, the marker is projected onto the detection target object (cut vegetable). Will be.
Note that the projection unit 60 may project the marker onto the surface of the cut vegetable (three-dimensional object) to which the foreign object (such as an insect) is attached, which is transported to the projection region 53.
More specifically, the projection unit 60 automatically adjusts the focus position based on the three-dimensional position coordinates of the surface of the cut vegetable to which foreign substances (such as insects) specified by the projection control unit 74 adhere, and then adjusts the marker. Project.

<Control unit 70>
The control unit 70 includes, for example, a processor having a CPU, a RAM, a ROM, an I / O, and the like, and controls each component of the foreign object detection system 100. The CPU executes each program for foreign object detection read from the RAM, ROM, or storage unit 80, and when executing the program, reads information from the RAM, ROM, and storage unit 80, and stores the information in the RAM and storage unit 80. Write information to it. Details will be described later.

  The storage unit 80 includes, for example, a semiconductor memory, a hard disk drive, and the like, and includes an operating system (OS), a foreign object detection program, a conversion matrix from the first pixel coordinate system to the second pixel coordinate system, and second pixel coordinates. A conversion matrix from the system to the first pixel coordinate system, a conversion matrix from the first pixel coordinate system to the three-dimensional space coordinate system, and the like are stored.

  FIG. 2 shows a functional block diagram of the control unit 70. As shown in FIG. 2, the control unit 70 includes a relative intensity calculation unit 71 as a relative intensity calculation unit, a detection unit 72 as a detection unit, a display control unit 73 as a display control unit, and a projection control unit. A projection control unit 74.

<Relative intensity calculation unit 71>
The relative intensity calculation unit 71 is based on the association between the pixel coordinates of the first pixel coordinate system of the first imaging unit 21 and the pixel coordinates of the second pixel coordinate system of the second imaging unit 22. A relative intensity value corresponding to each positioning point P on the surface of (cut vegetable) is calculated as a ratio between the intensity value of the pixel of the first intensity value image data and the intensity value of the pixel of the second intensity value image data. To do.
Specifically, the relative intensity calculator 71 calculates the relative intensity as follows.

The positioning point on the surface of the substance (for example, the inspection object 1 (cut vegetable) or the surface of the conveyance line 50) located in the inspection region 52 irradiated with near infrared rays by the irradiation unit 10 is reflected at P and the point P (and / or Or, the coordinate values in the first pixel coordinate system of the pixel (1 pixel) P ₁ obtained by imaging the first near infrared ray transmitted through the first imaging unit 21 are (u ₁ (P), v ₁ (P)), a point The coordinate values in the second pixel coordinate system of the pixel (1 pixel) P ₂ obtained by imaging the second near-infrared ray reflected at P by the second imaging unit 22 are defined as (u ₂ (P), v ₂ (P)). .
Here, as described above, the coordinate values of P ₁ (u ₁ (P), v ₁ (P)) and the coordinate values of P ₂ (u ₂ (P), v ₂ (P)) are the first pixels. Correspondence can be made by a transformation matrix from the coordinate system to the second pixel coordinate system.

The relative intensity calculation unit 71 includes the pixel P ₁ included in the first intensity value image data captured by the first imaging unit 21 and the second intensity value image data captured by the second imaging unit 22 for all points P. to calculate the relative intensity at all points P on the basis of the pixel P ₂ included in the.
Specifically, the intensity value of the first near infrared ray in the pixel P ₁ in the first pixel coordinate system is n (P ₁ ), and the intensity value of the second near infrared ray in the pixel P ₂ in the second pixel coordinate system is n (P ₁ ). ₂ ) When the relative intensity f (P) at the point P is assumed, the relative intensity calculating unit 71 calculates the relative intensity f (P) at each point P by the following formula 1.
f (P) = n (P ₁ ) / n (P ₂ ) (Formula 1)

<Detection unit 72>
The detection unit 72 is mixed in the inspection object 1 (cut vegetable) based on the relative intensity f (P) at each positioning point P on the surface of the inspection object 1 (cut vegetable) calculated by the relative intensity calculation unit 71. Alternatively, foreign substances (such as insects and / or insect eggs) are detected.
The detection unit 72 determines whether or not the relative intensity f (P) at each positioning point P on the surface of the inspection object 1 (cut vegetable) calculated by the relative intensity calculating unit 71 exceeds a preset threshold value. Thus, it is detected whether or not a foreign object (insect and / or insect egg, etc.) is mixed or adhered to the inspection object 1 (cut vegetable).
More specifically, in the detection unit 72, the relative intensity f (P) at each positioning point P on the surface of the inspection object 1 (cut vegetable) calculated by the relative intensity calculation unit 71 exceeds a preset threshold value. In this case, it is determined that foreign objects (insects and / or insect eggs) are mixed or attached to the inspection object 1 (cut vegetables).

<Near infrared spectroscopy>
Here, the reason why it is possible to detect whether or not foreign substances (insects and / or insect eggs, etc.) are mixed or attached to the inspection object 1 (cut vegetables) based on the relative strength will be described.
Near-infrared spectroscopy is spectroscopy in the near-infrared region, and a component is calculated based on the degree of absorption (absorbance) or change in reflection by irradiating a measurement object with near-infrared rays. Near-infrared light is much less absorbed than mid-infrared light and far-infrared light, so that non-destructive and non-contact measurement is possible without creating a section or the like.
Near-infrared rays have a low absorption rate in biological constituents, and particularly near-infrared rays having a wavelength of 800 nm to 1000 nm are said to have high biological transmittance. Therefore, the inventor finally obtained the intensity value by irradiating the inspection object 1 assuming various infrared rays with a near infrared ray having a wavelength of 680 nm to 990 nm.
Specifically, when the inspection object 1 is only cut vegetables (case 1), the inspection object 1 is attached to the front side of the cut vegetables with insects or insect eggs (case 2), the inspection object 1 The intensity value of each wavelength was determined for each of the case where the insect or insect egg was attached to the back side of the cut vegetables (case 3) and the case where only the insect or insect egg was used (case 4).
By repeating such an experiment, the inventor attached an insect or insect egg to the front side of the cut vegetable (case 2), and placed the test object 1 on the back side of the cut vegetable with the insect or insect egg. It was confirmed that almost the same strength value was obtained in any case (Case 3).
For example, by calculating the relative ratio of the intensity values of two near infrared rays having a wavelength of 680 nm as the first wavelength and a wavelength of 910 nm as the second wavelength, Case 1 and Case 2 to Case 4 are Successfully distinguished. Further, in particular, by calculating the relative ratio of the intensity values of two near infrared rays having a wavelength of 730 nm as the first wavelength and a wavelength of 980 nm as the second wavelength, Case 1 and Case 2 to Case 4 , Succeeded in distinguishing more clearly.
For example, when the wavelength of 730 nm is the first wavelength and the wavelength of 980 nm is the second wavelength, the relative ratio of the intensity values of the two near infrared rays is as follows.
When only cut vegetables are used (Case 1): 0.13
When insects are attached to the front side of cut vegetables (Case 2): 0.22
When insects are attached to the back side of cut vegetables (Case 3): 0.22
When only insects are used (case 4): 0.18
This is presumably due to the fact that the absorbance of cellulose decreases in the wavelength bands of 730 nm and 980 nm, and the absorbance of water starts to increase from around 980 nm. As a result, it was possible to distinguish Case 1 from Case 2 to Case 4 by setting a threshold value according to the light source, the installation environment of the apparatus, and the like.
As described above, the inventor is different from the first light (first near infrared ray) having the first center wavelength and the first near infrared ray included in the near infrared region, and the second light included in the near infrared region. By calculating the relative ratio of the intensity values of the second light having the central wavelength (second near-infrared ray), foreign matter (insect or insect egg) adheres to the front side or back side of the inspection object 1 (cut vegetable). Regardless, the invention has been completed that enables non-destructive and non-contact detection of the inspection object 1 (cut vegetables) in which foreign substances (insects or insect eggs) are mixed or adhered.

<Display control unit 73>
The display control unit 73 sets a predetermined pixel on the display unit 40 corresponding to the position of a foreign object (such as an insect and / or insect egg) mixed or attached to the inspection object 1 (cut vegetable) detected by the detection unit 72. As a color (for example, red), it is displayed so as to overlap the first intensity value image data.
More specifically, the display control unit 73 mixes or adheres to the inspection target 1 (cut vegetables) on the display unit 40 based on the first image coordinate system with the first intensity value image data as the background. A pixel corresponding to the position of a foreign object (such as an insect and / or insect egg) is displayed in the foreground as a predetermined color (for example, red).
Thereby, the operator or the worker can easily confirm foreign matters (insects and / or eggs of insects) mixed or attached to the inspection object 1 (cut vegetables) by referring to the display unit 40. .

  The display control unit 73 causes the display unit 40 to mix or adhere to the inspection object 1 (cut vegetables) with the second intensity value image data as a background based on the second image coordinate system (insects). And / or pixels corresponding to the positions of insect eggs, etc.) may be displayed in the foreground as a predetermined color (for example, red).

<Projection control unit 74>
The projection control unit 74 controls the projection unit 60 to project the marker in a state where the projection area 53 set on the transport line 50 is in focus. More specifically, the projection control unit 74 projects the marker toward the position of the projection area 53 where the cut vegetables (three-dimensional object) to which foreign matters (insects and the like) are attached are arranged. By doing so, when the detection target object (cut vegetable) to which foreign matter (insect or the like) is attached is transported to the projection region 53, the marker is projected onto the detection target object (cut vegetable).
Thereby, the worker can easily confirm and separate and remove foreign matters (insects and / or eggs of insects) mixed or adhering to the inspection object 1 (cut vegetables) by referring to the markers. .
Note that the projection control unit 74 may project the marker onto the surface of the detection target (cut vegetable) to which foreign matter (insects, etc.) is attached, which is transported to the projection region 53.
More specifically, the projection control unit 74 automatically adjusts the focus position with respect to the projection unit 60 based on the three-dimensional position coordinates of the surface of the cut vegetable to which foreign matter (such as insects) adheres, and then sets the marker. You may make it project.
Further, as described above, since the thickness of the cut vegetables in the Z direction is sufficiently small, the marker is projected toward the position on the plane of the projection area 53 where the cut vegetables to which foreign matter (insects, etc.) are attached are arranged. You may do it.

FIG. 3 is a diagram illustrating a plane of the conveyance line 50. The transport direction in the plane of the transport line 50 is defined as the X axis, the direction orthogonal to the transport direction of the transport line (plane) is defined as the Y axis, and the direction perpendicular to the plane of the transport line 50 is defined as the Z axis.
With respect to the inspection object 1 (cut vegetables) conveyed on the inspection area 52 on the conveyance line 50, the detection unit 72 has a foreign object (insect and / or insect egg, etc.) mixed or adhered to the inspection object 1 ( At the time t ₀ when the (cut vegetable) is detected, the three-dimensional position coordinates of the foreign matter (insect and / or insect egg, etc.) are set to (X ₀ , Y ₀ , Z ₀ ).
It is assumed that the speed of the transport line 50 is v (m / s), the start coordinate value in the X-axis direction of the projection area 53 on the transport line is X ₁ , and the end coordinate value in the X-axis direction is X ₂ .
The position coordinates (X, Y, Z) of the foreign object (such as insects and / or insect eggs) at time t (t ≧ t ₀ ) are expressed by (Expression 2).
X = X ₀ + v (t−t ₀ )
Y = Y ₀ (Formula 2)
Z = Z ₀
When the pixel coordinate value in the first image coordinate system and the pixel coordinate value in the second image coordinate system are associated with each other by the homography transformation matrix, Z ₀ may be a distance from the projector to the projection plane.
Then, the time t during which the foreign object (insect and / or insect egg, etc.) is conveyed on the projection area 53 is expressed by (Expression 3).
((X ₁ −X ₀ ) / v) + t ₀ ≦ t ≦ ((X ₂ −X ₀ ) / v) + t ₀ (Formula 3)
The projection control unit 74 projects the marker in the vicinity of the position indicated by (Formula 2) at the time t when the foreign object (insect and / or insect egg, etc.) is conveyed on the projection region 53.
Specifically, the projection control unit 74 controls the projection unit 60 to project the marker toward the planar position (X ₀ + v (t−t ₀ ), Y ₀ ) on the projection region 53.
In this way, the projection control unit 74 projects the marker onto the inspection object 1 (cut vegetable) that is conveyed or conveyed by the conveyance line 50 and in which foreign matters (insects and / or insect eggs, etc.) are mixed or adhered. be able to.
Note that the projection control unit 74 is a three-dimensional inspection object 1 (cut vegetable) in which foreign matter (insects and / or insect eggs, etc.) arranged on the projection region 53 with respect to the projection unit 60 is mixed or adhered. position _{(X 0 + v (t-} t 0), Y 0, Z 0) may be controlled so as to project the marker in a state of in focus towards.

Thereby, the worker removes the inspection object 1 (cut vegetable) in which foreign matters (insects and / or insect eggs, etc.) are mixed or adhered from the inspection object 1 (cut vegetable) conveyed on the conveyance line. It can be easily identified and separated and removed easily. Moreover, a marker is projected on the conveyance line 50 where the inspection object 1 (cut vegetables) to which foreign matters (insects and / or insect eggs etc.) are attached is located, and the positions of the foreign matters (insects and / or insect eggs etc.). By informing the workers, work can be made more efficient.
Further, the foreign object detection system 100 can be attached to an existing carry-in line, and the introduction cost can be reduced.

  FIG. 4 is a flowchart illustrating a flow of a series of processes in the embodiment of the present invention described so far. The processing flow will be described with reference to FIG. The inspection object 1 (cut vegetables) is transported on the transport line 50. In addition, the first imaging unit 21 and the second imaging unit 22 capture the inspection region 52 in synchronization.

  In ST <b> 1, the irradiation unit 10 irradiates near-infrared rays toward the inspection region 52 on the transport line 50.

  In ST <b> 2, the relative intensity calculation unit 71 calculates the relative intensity f (P) at all points P in the inspection area 52.

  In ST3, the detection unit 72 mixes or adheres foreign matter (insects and / or insect eggs, etc.) at the measurement point P where the relative intensity f (P) calculated by the relative intensity calculation unit 71 exceeds a preset threshold. Identify as a place to do.

  In ST4, the display control unit 73 causes the display unit 40 to store foreign substances (such as insects and / or insect eggs) that are mixed or adhered to the inspection object 1 (cut vegetables) with the first intensity value image data as a background. The pixel corresponding to the position is displayed in the foreground as a predetermined color (for example, red).

  In ST5, the projection control unit 74 is positioned on the projection area 53 where the inspection object 1 (cut vegetables) mixed with or attached to foreign matters (insects and / or insect eggs) conveyed by the conveyance line 50 is located. In contrast, a marker is projected.

  In ST6, the operator discriminates and removes the inspection object 1 (cut vegetables) on which the marker is projected.

As described above, the foreign object detection system 100 according to the first embodiment includes the first near infrared ray and the first near infrared ray included in the light in the band including the near infrared ray at each positioning point P on the surface of the inspection object 1 (cut vegetable). (2) Relative intensity calculating means 71 for calculating a relative intensity that is a relative ratio of intensity values by near-infrared rays, and determining whether the relative intensity calculated by the relative intensity calculating means 71 exceeds a preset threshold value. And a detection unit 72 that detects whether or not a foreign object (insect and / or insect egg, etc.) is mixed or adhered to the inspection object 1 (cut vegetable).
As a result, whether foreign matter (insects or eggs of insects) is attached to the front side of the test object 1 (cut vegetables) or attached to the back side, foreign matter (insects or eggs of insects) is attached. It enables non-destructive and non-contact detection of the object 1 (cut vegetables). Further, by calculating the relative intensity in units of pixels, it is possible to detect with high accuracy including the size of a foreign object (insect or insect egg).
Moreover, the foreign object detection system 100 can be attached to an existing line and can be introduced at low cost.

In addition, the foreign object detection system 100 acquires the first near-infrared reflected light and / or transmitted light and the second near-infrared reflected light and / or transmitted light by the first imaging unit 21 and the second imaging unit 22, respectively. To do.
Thereby, the foreign object detection system 100 can be constructed at low cost.

Further, the foreign object detection system 100 sets the inspection object 1 (cut vegetable) to which the foreign object (insect or insect egg) adheres to the display unit 40 as a predetermined color (for example, red) in units of pixels, and the first near infrared ray. The first intensity value image data that is reflected light and / or transmitted light or the second intensity value image data that is reflected light and / or transmitted light of the second near-infrared light is displayed in an overlapping manner.
Thereby, it is possible to display to a user (for example, an inspector, a worker, etc.) with high accuracy including the size of a foreign object (insect or insect egg) in pixel units.

Moreover, the foreign material detection system 100 projects the marker on the conveyance line 50 where the inspection object 1 (cut vegetable) to which the foreign material (insect or insect egg) adheres is located.
Thereby, the worker can easily check the inspection object 1 (cut vegetables) to which the foreign matter (insect or insect egg) adheres, and the inspection object 1 (to which the foreign object (insect or insect egg) adheres) ( The separation work of cut vegetables) can be made efficient.

In addition, the foreign object detection system 100 is configured to convey the inspection object 1 (cut vegetable) to which the foreign object (insect or insect egg) adheres to the inspection object 1 (cut vegetable) conveyed by the conveyance line. A marker is projected at the position of the line 50.
Thereby, the isolation | separation operation | work of the test target object 1 (cut vegetable) to which a foreign material (an insect or an insect egg) adheres can be performed at high speed.

  In addition, the foreign object detection method and the computer program according to the first embodiment have the same effects as the foreign object detection system 100.

  The preferred embodiment of the foreign object detection system 100 of the present invention has been described above, but the present invention is not limited to the above-described embodiment, and can be changed as appropriate.

<Freedom of system configuration>
In the foreign object detection system 100 according to the present invention, the relative intensity calculation unit 71, the detection unit 72, the display control unit 73, and the projection control unit 74 are configured to be included in the control unit 70. It is a design matter that can be appropriately made for the user to concentrate on the computer or to distribute the computer to a plurality of computers.
For example, the relative intensity calculation unit 71 and the detection unit 72 may be used as detection devices, and the detection device may include the irradiation unit 10, the first imaging unit 21, and the second imaging unit 22. At this time, the projection control unit 74 may be a projection device, and the projection device may include the projection unit 60. When the detection device and the projection device are separated, data communication may be performed between the detection device and the projection device using, for example, a communication cable.
In this case, the display unit 40 and the display control unit 73 may be configured to be included in the detection device or the projection device. The display unit 40 and the display control unit 73 may be independent as a display device. In this case, the detection device, the projection device, and the display device may be configured to perform data communication via, for example, a communication cable.

<Modification 1>
In the first embodiment, the imaging unit 20 includes at least a first imaging unit 21 and a second imaging unit 22, and each includes a first infrared camera 212 including a bandpass optical filter 211 for acquiring the first near infrared ray. , And a second infrared camera 222 including a bandpass optical filter 221 for acquiring the second near infrared ray.
On the other hand, instead of including the first imaging unit 21 and the second imaging unit 22 in the imaging unit 20, one multispectral camera or one hyperspectral camera may be applied. The multispectral camera arbitrarily selects a plurality of wavelengths and specializes in measuring the wavelengths, thereby enabling spectrum measurement. The hyperspectral camera can acquire a spectrum split into several tens of bands or more for each pixel of an image. By doing so, the imaging unit 20 can be integrated into one camera. Therefore, the imaging unit 20 includes one image coordinate system.
In this case, the relative intensity calculation unit 71 includes the first pixel that receives the reflected light and / or transmitted light of the first near-infrared light having the first center wavelength and is included in the pixel data obtained by imaging all the points P and the first pixel. The second pixel that receives the reflected light and / or transmitted light of the second near-infrared having a center wavelength of 2 can be selected, and the relative intensity can be calculated based on the intensity values of both.
Further, by a method known to those skilled in the art (for example, camera calibration), the pixel coordinate value (u (P), v (P)) in the pixel coordinate system of the point P and the coordinate value (X (P (P)) in the three-dimensional space are used. ), Y (P), Z (P)).
As described above, since the thickness of the cut vegetables in the Z direction is sufficiently small, the Z coordinate value Z ₀ of the point P on the cut vegetables may be approximated by the distance from the projector to the projection plane. Thereby, the pixel coordinate values (u (P), v (P)) in the pixel coordinate system of the point P and the coordinate values (X (P), Y (P) in the three-dimensional space are calibrated by the homography conversion. , Z ₀ ).
Thereby, the projection control unit 74 can control the projection unit 60 to project the marker toward the position on the projection region 53. Further, the projection control unit 74 can also control the projection unit 60 to project the marker in a focused state toward the surface of the cut vegetables above the projection region 53.

<Modification 2>
In the first embodiment, the inspection area 52 and the projection area 53 on the transport line 50 are set separately, but the projection area 53 may be configured to include the inspection area 52.

<Modification 3>
In the first embodiment, the irradiation unit 10 is composed of a point light source, but may be composed of a surface light source.

<Modification 3>
In the first embodiment, it is desirable to control the imaging timing by the control unit 70 so that the first imaging unit 21 and the second imaging unit 22 are time-synchronized, but the present invention is not limited to this. For example, when the shooting timings of the first imaging unit 21 and the second imaging unit 22 are slightly shifted, linear interpolation may be performed using a time stamp.

<Modification 4>
In the first embodiment, reflected light and / or transmitted light from the inspection object are received, but only reflected light from the inspection object may be received depending on the sensitivity of the camera. In addition, only the transmitted light from the inspection object may be received.

DESCRIPTION OF SYMBOLS 100 Foreign object detection system 1 Inspection object 10 Irradiation part 20 Imaging part 21 First imaging part 22 Second imaging part 40 Display part 50 Conveyance line 51 Irradiation area 52 Inspection area 53 Projection area 60 Projection part 70 Control part 71 Relative intensity calculation part 72 Detection Unit 73 Display Control Unit 74 Projection Control Unit 80 Storage Unit

Claims (9)

An irradiation unit that irradiates light in a band including near-infrared rays to the inspection object;
First intensity value image data obtained by imaging the inspection object by receiving reflected light and / or transmitted light from the inspection object by the first light having the first central wavelength included in the band. A first imaging unit for acquiring
Second intensity value image data obtained by imaging the inspection object by receiving reflected light and / or transmitted light from the inspection object by the second light having the second center wavelength included in the band. A second imaging unit for acquiring
Based on the correspondence between the coordinate values of the pixel coordinate system of the first imaging unit (coordinate values of the image coordinate system) and the coordinate values of the pixel coordinate system of the second imaging unit (coordinate values of the image coordinate system). A relative intensity corresponding to each positioning point on the surface of the inspection object is a ratio of the intensity value of the pixel of the first intensity value image data and the intensity value of the pixel of the second intensity value image data. A relative intensity calculating means for calculating;
Detection means for detecting foreign matter mixed in the inspection object based on the relative intensity at each positioning point on the surface of the inspection object calculated by the relative intensity calculation means;
Foreign object detection system comprising: The first imaging unit is a camera including an optical filter for acquiring first light having the first center wavelength,
The foreign object detection system according to claim 1, wherein the second imaging unit is a camera including an optical filter for acquiring second light having the second center wavelength. The detection means includes
Claims for detecting foreign matter mixed in the inspection object when it is determined that the relative intensity at each positioning point on the surface of the inspection object calculated by the relative intensity calculation means exceeds a preset threshold value. The foreign matter detection system according to claim 1 or claim 2. The foreign object detection system further includes:
A display unit;
4. The display control unit according to claim 3, further comprising: a display control unit configured to display, on the display unit, pixels exceeding the threshold as a predetermined color so as to overlap the first intensity value image data or the second intensity value image data. 5. Foreign object detection system. The foreign object detection system further includes:
A projection unit for projecting a preset marker;
The foreign matter detection system according to claim 3, further comprising: a projection control unit that projects the marker in the vicinity of each positioning point on the surface of the inspection object that has been determined that the relative intensity exceeds the threshold value. . The foreign object detection system further includes:
A transport line for transporting the inspection object;
The projection control means projects the marker in the vicinity of each positioning point on the surface of the inspection object which is the inspection object transported by the transport line and the relative intensity of which is determined to exceed the threshold. The foreign object detection system according to claim 5.   The foreign matter detection system according to any one of claims 1 to 6, wherein the inspection object is a cut vegetable, and the foreign matter is an insect and / or an egg of an insect attached to the cut vegetable. An irradiation unit that irradiates light in a band including near-infrared rays to the inspection object;
By receiving reflected light and / or transmitted light from the inspection object by the first light having the first center wavelength in the band, the inspection object is imaged to obtain first intensity value image data. A first imaging unit that
By receiving reflected light and / or transmitted light from the inspection object by the second light having the second center wavelength in the band, the inspection object is imaged to obtain second intensity value image data. A second imaging unit that
A foreign object detection system comprising a control unit,
A pixel coordinate system coordinate value of the first intensity value image data acquired by the first imaging unit (a coordinate value of the image coordinate system) and a pixel of the second intensity value image data acquired by the second imaging unit Intensity values of pixels of the first intensity value image data corresponding to each positioning point on the surface of the inspection object based on association with coordinate values of the coordinate system (coordinate values of the image coordinate system) And a relative intensity calculating step for calculating a relative intensity that is a ratio of the intensity value of the pixel of the second intensity value image data;
A detection step of detecting foreign matter mixed in the inspection object based on the relative intensity at each positioning point on the surface of the inspection object calculated by the relative intensity calculation step;
A foreign matter detection method comprising:   The computer program for functioning a computer as each means of the foreign material detection system described in any one of Claims 1-7.

Images