JPH11194021A - Measuring device for fruits and vegetables - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a measuring device for fruits and vegetables for measuring the bending of an object to be measured in a state close to a human feeling due to the amount of recess on the outer surface of fruits and vegetables and for appropriately measuring the bending regardless of the posture of the object. SOLUTION: An image-processing means for evaluating a long object K to be measured by picking up the image of the object K using a CCD camera and performing the image processing of the image pick-up information obtains a reference line 50 in the longitudinal direction of the object K based on the image pick-up information of the CCD camera, obtains reference points P3 and P4 on a contour where the distance from the reference line 50 can be maximized at each setting region being located at both edge sides in the longitudinal direction of the object K, and evaluates, as the bending Q of the object K, the maximum distance between a virtual line 51 for connecting the reference points P3 and P3 and a contour part located at the virtual line side out of the contours of the object K.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image pickup means for picking up an image of a long object to be measured and outputting image pickup information for measurement, and processing the image pickup information of the image pickup means to image the object to be measured. The present invention relates to a fruit and vegetable measuring device provided with an image processing means for evaluating.

[0002]

2. Description of the Related Art Conventionally, the above-mentioned fruit and vegetable measuring apparatus has a structure as disclosed in Japanese Patent Application Laid-Open No. 61-174980, for example. In other words, for the object to be measured moving in the predetermined direction, a detection line connecting the most bulging portions in the moving direction on both sides in the longitudinal direction of the object to be measured is obtained, and among the perpendicular lines provided for this detection line, The maximum length from the detection line to the object to be measured is determined, and the length is measured as the bend of the object to be measured. The bending of the object to be measured is measured in a state close to the sense.

[0003]

However, in the above-mentioned conventional configuration, even if the longitudinal direction of the object to be measured is slightly inclined with respect to the moving direction from the aligned posture perpendicular to the moving direction, the bending is not required. Can be measured, but if the posture of the measured object changes greatly,
There is a possibility that the bending cannot be measured properly.
For example, if the longitudinal direction of the object to be measured is in a posture along the moving direction, it is not possible to obtain the most bulged portions and the detection lines as described above. You can't.

The present invention has been made in view of such a point, and an object of the present invention is to measure a bent as an amount of dent on an outer surface of a fruit and vegetable so that an object to be measured can be approximated to a human sense. It is an object of the present invention to provide a fruit and vegetable measuring device that can measure a bend properly, regardless of the posture of an object to be measured, while measuring the bend.

[0005]

According to a feature configuration of the present invention, the image processing means obtains a reference line along the longitudinal direction of the object to be measured, based on the imaging information of the imaging means. In each of the setting areas located at both ends in the longitudinal direction of the object to be measured, a reference point on the contour where the distance from the reference line is the largest is obtained, and each reference point on each of both ends in the longitudinal direction is connected. A configuration in which the maximum distance between a virtual line and a contour portion located between the reference points and on the virtual line side among the contours of the measured object is evaluated as a curve of the measured object. Have been.

Therefore, the image processing means first obtains a reference line along the longitudinal direction of the object to be measured based on the imaging information of the imaging means. As the reference line along the longitudinal direction, there can be considered a method of using a principal axis obtained by calculating a moment feature amount as described later, or a method of using an approximate straight line to a middle point sequence of the contour of the image of the object to be measured. This reference line is determined so that the object to be measured is always in the longitudinal direction regardless of the posture. Each reference point (see symbols P3 and P4 in FIG. 12) on the contour where the distance from the reference line is the maximum in each of the setting regions located at both ends in the longitudinal direction of the measured object is the reference line ( Reference numeral 50 in FIG.
Is the point on the contour where the separation distance from the
A virtual line (see reference numeral 51 in FIG. 12) corresponding to the extreme end position in the bending direction (a direction orthogonal to the longitudinal direction) of the measured object and connecting each reference point regardless of the posture of the measured object; The maximum separation distance of the contour portion of the measurement object located on the side corresponds to the amount of depression in the outer peripheral portion of the measurement object, and this maximum separation distance is evaluated as the bending of the measurement object.

As a result, it is possible to measure the bending of the object to be measured in a state close to the human sense of visually determining the bending. And a fruit and vegetable measuring device capable of measuring the amount of fruit and vegetables.

According to a second aspect of the present invention, in the first aspect, the image processing means calculates a moment feature amount of the image of the object, calculates a center of gravity of the image of the object, and Determining a tilt angle of the object to be measured with respect to a reference direction on the captured image, obtaining a straight line passing through the position of the center of gravity and having the tilt angle with respect to the reference direction, and using the straight line as the reference line, It is configured to evaluate things.

The image processing means extracts an image area of the object to be measured based on the image information of the imaging means, and calculates a moment feature value for the image. Hereinafter, this will be described in detail with reference to equations and FIG.

First, the above image is expressed by the following [Equation 1].
Moment M around origin defined _0,0, M_1,0, M
_0,1Ask for.

[0011]

(Equation 1)

Here, f (x, y) is the value of the pixel at the coordinates (x, y) on the binary image. Then, the center of gravity position (x ₀ , y ₀ ) of the image of the cucumber is obtained from the following [Equation 2].

[0013]

X ₀ = M _1,0 / M _0,0 , y ₀ = M _0,1 / M _0,0

Next, the position around the center of gravity on the image of the object to be measured is
Second moment M_1,1, M_2,0, M _0,2To [Equation 3]
And the image of the measured object expands based on [Equation 4].
Direction (main axis direction) and the reference direction on the captured image (Fig.
12 along the x-axis) and the x-axis
And a line 50 passing through the position G of the center of gravity at an angle θ
Then, the object to be measured is evaluated.

[0015]

(Equation 3)

[0016]

[Equation 4] θ = 1/2 · tan ⁻¹ ｛2M _1,1 / (M _2,0 −
M _0,2 )｝

Accordingly, a reference line along the longitudinal direction of the measured object can be reliably obtained regardless of the posture of the measured object.

According to a third aspect of the present invention, in the first or second aspect, the image processing means is located on a contour of an image of the object to be measured and in a direction along the reference line. The system is configured such that two points that are most separated from each other are obtained, and a distance between the two points in a direction along the reference line is evaluated as a length of the object to be measured.

Therefore, the distance between the two most distant points in the direction along the reference line in the contour of the object to be measured is evaluated as the length, so that the object to be measured is always in the longitudinal direction regardless of its posture. The length can be evaluated in an appropriate condition along the length.

According to a fourth aspect of the present invention, in any one of the first to third aspects, there is provided a transport conveyor for transporting the object to be measured from a supply location to an extraction location through an imaging location. The imaging unit is configured to image the object to be measured at the imaging location.

The object to be measured supplied to the supply location of the transport conveyor is conveyed to the takeout location through the imaging location, and is imaged by the imaging means at the imaging location. Since the supply and unloading of the object to be measured are performed at different locations, the work efficiency is high, and the measurement of the object to be measured can be performed regardless of the mounting posture when the object is supplied on the conveyor. Is performed properly, so that it is possible to provide a highly efficient measuring device without troublesome supply operation such as supply in an aligned state.

[0022]

BEST MODE FOR CARRYING OUT THE INVENTION The fruit and vegetable measuring apparatus according to the present invention will be described below in the case where it is applied to a cucumber sorting apparatus as an example of an object to be measured. FIG. 1 shows a cucumber sorting apparatus. This cucumber sorting apparatus takes out cucumber K one by one from a container 1 in which a large number of cucumbers K are loaded in a state where the longitudinal directions thereof are aligned in the same direction, and feeds the cucumber K one by one to a conveyor conveyor. When the cucumber K is located at a predetermined imaging location on the conveying path of the belt conveyor 3, the cucumber K is supplied to a supply location located on the upstream side of the belt conveyor 3.
A measuring unit 4 is provided for measuring the length, bending and the like of the cucumber K based on the imaging information of the cucumber K from above, and obtaining the grade and class of the cucumber K from the result. Then, cucumbers K are taken out one by one from a take-out location on the lower side in the transport direction from the imaging location, and based on the evaluation results of the measuring unit 4, the cucumbers K are set in the storage boxes 5 as a plurality of storage units in advance. An unloading device 6 is provided as unloading means for sorting and loading based on the sorting reference information.
Further, the supply device 2, the unloading device 6, and the belt conveyor 3
Is provided with a control device 7 as control means for controlling the operation of.

As shown in FIG. 2, the supply device 2 sucks and holds the cucumbers K one by one and takes them out of the container 1 by means of suction nozzles 8 provided so as to be vertically movable and reciprocally movable in the horizontal direction. It is configured to be transferred onto the belt conveyor 3. That is, the support mechanism 10 is provided so as to be reciprocally movable with respect to the machine frame 9 in a horizontal direction (conveyor horizontal width direction) orthogonal to the conveying direction of the belt conveyor 3. 8 is supported so as to be able to move up and down.

More specifically, as shown in FIG. 5, the supporting mechanism 10 includes a pipe frame 11 extending in the lateral direction, a rectangular frame portion 12 connected to the pipe frame 11 and extending downward, and a pipe frame 11. A plurality of guide rollers 14 attached to both ends of the pipe frame 11 are provided along the machine frame 9. It is configured to be slidably engaged with and fixed to the left and right fixed rails 15, 15, and the entire support mechanism 10 is provided slidably in the horizontal direction. In addition, one for sucking and taking out cucumber K into the rectangular frame portion 12.
A support cylinder 16 provided with a plurality of suction nozzles 8 at the lower end is provided so as to be able to move up and down.

The support mechanism 10 is connected to a rubber belt 17 supported by the machine frame 9 while being rotatably wound along the fixed rails 15, 15, and is provided at one end of the machine frame 9. The rubber belt 17 is driven to rotate forward and backward by the first electric motor M1 so as to be reciprocated in the horizontal direction. A rubber belt 18 rotatably wound in the up-down direction is provided inside the rectangular frame portion 12 so as to be connected to the support cylinder 16.
The support cylinder 16, that is, the suction nozzle 8 is moved up and down by driving the rubber belt 18 to rotate forward and backward by a second electric motor M <b> 2 provided in the inside. An air suction passage 19 for performing a suction operation from the suction nozzle 8 by an air compressor (not shown) is formed inside the support cylinder 16. A laser type distance sensor 20 for detecting a target to be taken out of the plurality of cucumbers K loaded in the container 1 is provided on the lateral outer side of the rectangular frame portion 12.

[0026] After the harvest, the appearance defect such as color unevenness and scratches is artificially selected by an operator, and the defective appearance product is distinguished and stored in another storage device, for example. In the container 1, non-defective cucumber K having no defective appearance is loaded and stored in a state where the cucumber K is aligned in the longitudinal direction. Further, the container 1 is placed so that the longitudinal direction of the loaded cucumbers K is the same as the transport direction of the belt conveyor 3.

As shown in FIG. 3 and FIG.
Is provided with a support mechanism 21 similar to the supply device 2,
One suction nozzle 22 for sucking and taking out cucumber K
Is configured to be movable up and down and movable in the horizontal direction. However, in the unloading device 6, the support mechanism 21 supporting the suction nozzles 22 is in the same direction as the conveyor conveyance direction and in a direction perpendicular to the conveyor conveyance direction ( (In the width direction of the conveyor).
Is provided so as to be rotatable around the vertical axis. The supply device 2 is provided inside the support cylinder 23.
An air suction passage 24 for suction operation from the suction nozzle 22 by an air compressor (not shown) is formed in the same manner as in the above. That is, as shown in FIG. 6, the support mechanism 21
5, a rectangular frame portion 26 connected to the pipe frame 25 and extending downward, a cover body 27 which is located above the pipe frame 25 and covers an electric motor and the like described later, and the like.
A plurality of guide rollers 28 attached to both ends of the pipe frame 25 are slidably engaged with fixed rails 30 on both left and right sides provided along a support rail portion 29. Support mechanism 21
The whole is slidably provided in the horizontal direction along the support rail portion 29. The support rail portion 29 is supported by the machine frame 31 so as to be reciprocally movable along the width direction of the conveyor, and a rubber belt 32 is rotatably wound along the longitudinal direction of the support rail portion 29. The support mechanism 21 is connected to the support mechanism 21, and the rubber belt 32 is driven forward and reverse by the third electric motor M3, so that the support mechanism 21 is configured to be reciprocated along the conveyor conveyance direction, as shown in FIG. A rubber belt 33, which is rotatably wound along the machine frame 31, is connected to the support rail portion 29, and the rubber belt 33 is driven forward and reverse by the fourth electric motor M4, so that the support rail portion 29 is conveyed horizontally. It is configured to be operated to reciprocate along the direction. Further, the support cylinder 23 is provided in the same manner as in the case of the supply device 2,
Rubber belt 34 that is rotatably wound along the vertical direction
And the rubber belt 3 is driven by the fifth electric motor M5.
4 is driven up and down in the up and down direction by rotating the head forward and backward.

In the unloading device 6, the support cylinder 23 is placed in the sixth electric motor M
6 is provided so as to be rotatable around the vertical axis X1 by the drive of 6, so that the rotation posture can be corrected while the cucumber K is being sucked. That is, a drive shaft 35 having a hexagonal cross section is disposed inside the rectangular frame portion 26 in a state parallel to the support cylinder 23 so as to be able to be driven and operated by the sixth electric motor M6 so as to cover almost the entire vertical direction. A drive gear 36 is externally fitted so as to be rotatable integrally with the shaft 35 and slidable in the vertical direction, and is integrally formed with the support cylinder 23 in a state where the driven gear 37 provided on the support cylinder 23 meshes with the drive gear 36. It is provided so that it can be moved up and down,
The support cylinder 23 is configured to be rotatable regardless of the elevating operation.

The belt conveyor 3 is driven to rotate by a seventh electric motor M7, as shown in FIG. 3, and is controlled so as to intermittently repeat a transfer operation and a stop as described later. It is configured to be.

Each of the electric motors M1 to M7 as described above
Are constituted by pulse motors, respectively, and as shown in FIG. 8, the operation state is controlled by a control device 7 as control means. The control device 7 includes a microcomputer, and is configured to control the operation of each electric motor according to a control procedure preset by a control program stored in a storage means (ROM).

As shown in FIG. 4, the plurality of storage boxes 5 are distributed and arranged on both sides in the width direction of the belt conveyor.
The cucumber receiving surface of the storage box 5 is set in a horizontal posture. still,
A plurality of cucumbers K for storing cucumber K, which is supplied through a chute 39b from a passage hole 39a formed between the belt conveyor 3 and the storage box 5, is provided below the mounting table 38 on which the storage box 5 is mounted. A collection unit 40 is provided.

As shown in FIG. 7, the measuring section 4 is provided with a box-shaped light-shielding cover 41 forming a shielding space for photographing, and the cucumber K ( CCD as an imaging means for photographing an object to be measured
The camera 42 is installed in a fixed position. The opening in the light-shielding cover 41 that allows the passage of the cucumber K is shielded so that external light does not enter the shielded space, and is cut at appropriate intervals so as to allow the cucumber K to pass through. A droop 43 is provided.

The CCD camera 42 has a CCD image pickup device (charge-coupled device) 42a as an image forming portion having a two-dimensional resolution, and an image pickup range for the cucumber K located in front of the CCD image pickup device 42a. Image sensor 42a
The image of cucumber K in a state where it is wider than the light receiving area of
And a lens mechanism 42b as an optical unit for forming an image on the D image pickup device 42a.
Is configured to sequentially output the imaging information of the imaging location in each set cycle, and is configured to output the imaging information as brightness (luminance) information, so-called monochrome image information. In addition, a light source (incandescent lamp) 44 for irradiating from above is provided in the light-shielding cover 41 so as to make the amount of irradiation of the cucumber K, which is the subject, uniform over substantially the entire surface.
Reflecting mirrors 45 are provided on both sides in the width direction of the belt conveyor 3 so as to irradiate the cucumber K from the lateral direction. C
The CD image pickup device 42a has a size of 256 × 2
A device having a resolution of 56 pixels is used, and the region to be imaged is a square region with a side of about 30 cm.

Further, an image processing means 100 for evaluating the cucumber K by performing image processing on the image pickup information output by the CCD camera 42 is provided. This image processing means 100
Specifically, as the evaluation processing of the cucumber K, specifically, the length, bending, thickness, etc. of the cucumber K are measured, and based on the measurement result and the sorting reference information set and stored in advance, It is configured to determine the class / grade for sorting the cucumbers K.

As shown in FIG. 8, the image processing means 10
Numeral 0 denotes an analog / digital conversion process for converting analog image information output from the CCD camera 42 into digital information, and at the same time, a binarization process for distinguishing the image of the cucumber K from other areas. A / D converter 46 as value processing means, image memory 47 as storage means for sequentially storing output data of A / D converter 46
And a data processing section 48 as data processing means for determining the class / grade for sorting the cucumbers K based on the data stored in the image memory 47.

The A / D converter 46 is a CCD camera 4
Since the information output from 2 is converted to binary information in real time and processed, it is processed at high speed. or,
The data processing unit 48 is constituted by a microcomputer, reads the image data stored in the image memory 47 every set time (1/30 second), and stores the read data in a storage unit (RO).
The image processing is executed in accordance with the processing procedure described later by the control program set and stored in M).

The conveying mounting surface of the belt conveyor 3 has a different brightness (lower) than that of the cucumber K.
The color of the cucumber K in the captured image obtained by the CCD camera 42 is more reliable than that of the other area (background, that is, the conveyor mounting surface). The image is configured to be bright. Thus, the image of the cucumber K can be extracted as accurately as possible.

Next, a processing procedure of the evaluation processing for the cucumber in the image processing means 100 will be described based on a control flowchart shown in FIG. First, an image is fetched from the image memory, and the image is stored in the A /
Binarization processing is performed by real-time processing using the D converter 46 (step 1). That is, binarization is performed by setting an area having a luminance equal to or higher than the set threshold to “1” and setting the other area to “0”. Next, while searching for the contour of the image area, a noise removal process for removing a minute area is performed (step 3). More specifically, the image data is sequentially scanned in the horizontal direction from the upper left to search for a pixel of data “1”. When the pixel of the data "1" is found, the scanning of the image data is interrupted, and from that point, the contour point of the image of the data "1" is searched, and the peripheral length of the contour point is simultaneously measured. . If the perimeter is less than a preset value, it is ignored as a noise component, and if it is more than the set value, the image area is a cucumber image area.

The outline 4 of the image area thus obtained
Using the information of No. 9, a moment feature value (center of gravity position, main axis) is calculated (step 4). More specifically, moments M _0,0 , around the origin defined by the following [Equation 5]
M _1,0 and M _0,1 are obtained.

[0040]

(Equation 5)

Here, f (x, y) is the value of the pixel at the coordinates (x, y) on the binary image. Then, the barycentric position (x ₀ , y ₀ ) of the image of the cucumber can be obtained from the following [Equation 6].

[0042]

X ₀ = M _1,0 / M _0,0 , y ₀ = M _0,1 / M _0,0

Next, the secondary moments M _1,1 , M _2,0 , and M _0,2 around the position of the center of gravity on the image of the cucumber are obtained from the following [ _Equation 7], and based on [Equation 8], the cucumber is obtained. An angle θ between the direction in which the image is extended (principal axis direction) and the x-axis is determined, and a line 50 passing through the center of gravity G at an angle θ with the x-axis becomes a principal axis as a reference line. Note that the main axis may be parallel to the x-axis in the image (θ = 0 °).

[0044]

(Equation 7)

[0045]

[Equation 8] θ = 1/2 · tan ⁻¹ ｛2M _1,1 / (M _2,0 −
M _0,2 )｝

Next, the length L of the cucumber is measured based on the outline 49 (step 5). That is, as shown in FIG. 12, each point sequence on the contour 49 is projected on the main axis 50 in a direction orthogonal to the main axis 50, and the distance between the two points where the distance is the largest among the projected points is determined. Let the length of the cucumbers be L.

The bending of the cucumber is measured based on the outline 49 (step 6). More specifically, as shown in FIG. 12, each point P1,
The distance (the distance in the direction perpendicular to the main axis) between the outline 49 and the main axis 50 is determined in order from each of the P2 to the set range toward the center in the length direction, and the distance on the outline 49 corresponding to the largest one is obtained. The reference points P3 and P4 are obtained. Next, the distance between a straight line 51 (virtual line) connecting the respective reference points and the contour data between the respective reference points P3 and P4 is sequentially obtained, and the maximum distance between the two points is defined as a cucumber curve Q.

An appropriate position Hi for suction (holding) by the suction nozzle 22 in the unloading device 6 is determined (step 7). That is, two points where a straight line passing through the center of gravity position G and perpendicular to the main axis 50 intersects the 49 contours are obtained, and the center position between these two points is determined as the proper position Hi.

The thickness F of the cucumber is measured based on the outline 49.
(Step 8). More specifically, as shown in FIG.
Along the length direction of the cucumbers (the direction along the main axis)
Width on the contour 49 so that it is equally divided into N (for example, 12 equally)
Each point a along one side of _1,a_Two... a_nIs specified.
In addition, the end of the cucumber in the longitudinal direction has "heap" etc.
The image is removed from the small area set at both ends.
I have. Then, the outline 49 opposing each point in the width direction.
Each point b above_1,b_Two... b_nTo its longitudinal setting
Find distances at multiple points up to contour 49 over a fixed range
Therefore, the minimum value is used as the measurement value at that location. That
Such a measurement is performed at each of the above points, and the
The average value is defined as the thickness F of the cucumbers.

Of the plurality of regions in which the length of the cucumber is divided into N equal parts for the thickness measurement, the outermost regions 53R and 53L located at both ends in the longitudinal direction are located adjacent to the central region. For each of the set setting areas 54R and 54L and the intermediate area 55 positioned in the middle in the longitudinal direction, the area of each area is determined by counting the number of pixels in the image (step 9).

Next, as shown in FIG. 14, the rectangular data 56 whose long side direction is parallel to the main axis 50 and circumscribes the image of the cucumber is obtained, and the distance between the long side of the rectangular data 56 and the contour 49 is obtained. The contour data is created by determining the distance at each set interval (step 10). This contour data is used at the time of carrying out operation.

Finally, the class of the cucumber is determined based on the measurement information (step 11). This class determination process will be described in detail with reference to FIG. That is, basically, based on the measured value of the length L and the preset class reference information (for example, a plurality of classes such as SS, S, M, L, and LL), the length of the cucumbers is determined. Is determined (step 101), and based on the measured value of the bending Q and predetermined grade reference information (for example, a plurality of grades such as A, B, and C). It is determined which grade the bend of the cucumber corresponds to (step 102). Then, the rank of the class (for example, AM, ALL, BS, etc.) is determined based on the combination of the class and the class (step 103).

Next, it is determined whether the cucumber is thick or fine based on a criterion set in advance corresponding to the relationship between the length L and the thickness F of the cucumber (step 10).
4, 105). Then, when it is determined that the class is extremely thick, the class determined based on the length L is changed to one rank higher (step 106). If it is determined that the class is extremely fine, the class determined based on the length L is changed to one rank lower (step 107).

Further, based on the area information measured in the step 9, it is determined whether or not the body is thick or narrow and whether or not the central part is constricted (step 10).
8). That is, as shown in FIG.
(53L) and a setting area 54R (54L) adjacent thereto
, That is, whether the ratio of the area of 53R to the area of 54R, and the ratio of the area of 53L to the area of 54L is larger than the judgment value for thick tail or smaller than the judgment value for thin tail. Is used to determine whether or not it is thick or thin, and the intermediate area 55 and the setting area 54R (54
L) to determine whether or not a constricted state. When such a thick or narrow or narrow neck is determined, processing such as changing the rank based on a preset reference value or determining that the product is out of grade is performed (step 109). Then, based on the measurement items as described above, it is finally determined which of the equal ranks set in advance for the cucumber K for each storage box 5 ( Step 110).

Next, the control operation of the control device 7 will be described with reference to a control flowchart shown in FIG. When the operation is started, the first supply operation by the supply device 2 is executed (step 21). That is, the supply device 2
With the suction nozzle 8 raised,
The distance sensor 20 measures the distance to the upper surface of the cucumber K while reciprocating horizontally in the entire range of the container 1 and divides the entire width of the container 1 into five zones. The position of the cucumber K at the highest position (the position where the distance is the minimum) is detected. Then, the cucumber in the highest zone among the highest positions in each zone is sucked and placed and supplied to the supply position on the belt conveyor 3. In the supply operation from the next time, the cucumber K is taken out from the zone where the cucumber is at the highest position at that time. When the extraction is completed for all five zones, the distance sensor 20 is moved again to detect the height, and the cucumbers are sequentially extracted in the same order.

Next, conveyance of the belt conveyor 3 is started,
Based on the image information in the image processing means 100, when it is determined that the lower end of the cucumber K in the transport direction has reached the predetermined area U on the upper side in the transport direction in the imaging area of the CCD camera 42, from that point in time After the conveyance by the set distance, the conveyance of the belt conveyor 3 is stopped (steps 22 to 2).
5). The set distance is set to a value close to the maximum value expected as the length of the cucumber K (for example, about 25 cm) (see FIG. 15).

In the state where the conveyance of the belt conveyor 3 is stopped, the evaluation processing of the cucumber K by the image processing means 100 is executed based on the image of the cucumber K taken by the CCD camera 42, and at the same time, this conveyance is performed. In the stopped state,
The supply operation of the cucumber K by the supply device 2 and the sorting and unloading operation by the unloading device 6 are executed (steps 26 and 2).
7, 28). When the supply operation and the unloading operation are completed, the conveyance by the belt conveyor 3 is restarted, and
3, the above-described evaluation processing is repeated until the processing is completed for all the cucumbers K in the container 1 (steps 29, 30, 31).

The unloading operation will be described. At the start of control (when power is turned on), the suction nozzle 22 is located at a preset origin (reference position) (see FIG. 9). This origin is a position where the cucumber K to be carried out is expected to be present when the conveyance of the belt conveyor 3 is stopped in a plan view, and as shown in FIG.
The position is set to a position corresponding to twice the distance α between the supply position and the imaging position by the CCD camera 42. FIG.
As shown in FIG. 7, in the unloading operation, the image processing means 1
The suction position (holding position) Hp is determined as coordinate data in plan view from the information on the appropriate position Hi for suction of the cucumber K determined in 00 and the information on the transport distance by the belt conveyor 3 (step 40). . In addition, the conveyance distance of the belt conveyor 3 can be determined from the count value of the pulse applied to the seventh electric motor M7 (pulse motor). The storage box 5 to be sorted is determined based on the evaluation result by the image processing means 100, and the target carry-in position for the storage box 5 is determined (step 41). More specifically, as shown in FIG. 16, cucumbers have already been carried into the storage box 5,
If the previous carry-in position is stored, a position where the rectangular data 56 of the cucumber K at the previous carry-in position and the rectangular data 56 of the cucumber to be carried in this time are closely arranged as a temporary position (target reference position). Initial settings (Fig. 1
6 (a)), the target carry-in position is determined based on the contour data determined by the image processing means 100 such that the gap between the cucumbers is reduced. That is, as shown in FIG. 16 (b), the minimum value of the sum of the gaps between the previous contour 49 of the cucumbers and the contour 49 of the present cucumbers (the distance between the contour 49 and the rectangle at a plurality of locations) is calculated. In other words, the carry-in target position is obtained by shifting the position from the tentative position to the lower side of the inclination so that each cucumber comes close to the extent of contact with each other. If the previous carry-in position is not stored, the carry-in target position is set so as to approach the inner edge at one end of the storage box.

By moving the suction nozzle 22 to the suction position,
The suction nozzle 22 is lowered to the location where the cucumber K is present, and the suction operation is performed by sucking air (step 42). It is determined whether or not suction has succeeded by checking if the pressure in the air suction passage has become negative by a negative pressure sensor. If suction is successful, the cucumber K slightly rises while sucking Then, the proper posture for storage, that is, the rectangular data of the cucumber K at the previous loading position and the cucumber K to be loaded this time
When the cucumber K is bent, the rotation phase of the cucumber K is corrected by the rotation operation of the support tube 23 so that when the cucumber K is bent, the posture is such that both ends are located on the lower side. (Steps 43, 44, 4
5). In other words, the cucumber K is conveyed to the target carry-in position while being sucked, and is carried into the storage box 5 (step 46). When the loading operation is completed, the loading position with respect to the storage box 5 is stored, and the suction nozzle 22 is returned to the origin position (steps 47 and 48). When the cucumber K is loaded from one end side to the other end side of one storage box 5, the descending amount of the suction nozzle 12 is reduced by a set amount, and the cucumber K in the lower row of the previously loaded lower row is reduced. It will be loaded sequentially on top. In step 43, if the suction operation has not been properly performed even after the set number of times (about three times), the unloading operation is terminated to prepare for the subsequent cucumbers K. (Steps 49 and 50).

The cucumber K whose suction operation by the unloading device 6 could not be properly performed, or the cucumber K or the like other than the rank set for each storage box 5 is disposed on the lower side in the transport direction of the belt conveyor 3. It is collected by the collection unit 57 and is processed manually.

[Another Embodiment] (1) In the above embodiment, the main shaft 50 is obtained by a moment calculation as a reference line along the longitudinal direction of the cucumbers. May be obtained along the outline, and an approximate straight line of the middle point sequence may be obtained, and this approximate straight line may be used as a reference line.

(2) In the above embodiment, the CCD camera 42 having a two-dimensional resolution is used as the imaging means. However, a CCD line sensor having a one-dimensional resolution only is arranged along the conveyor width direction. The image information of the object to be measured may be obtained by sequentially reading the output. Further, instead of the CCD camera 42, a MOS type imaging device or other imaging device may be used.

(3) In the above embodiment, the object is placed and transported in a posture in which the longitudinal direction of the object is along the conveyor transport direction. However, the present invention is not limited to such a configuration, and the object can be placed in any direction. It may be one that is transported.

(4) In the above embodiment, the supply device 2 automatically supplies the object (cucumber K) onto the belt conveyor 3;
An unloading device 6 for automatically taking out an object (cucumber K) from the belt conveyor 3 and storing it in the storage box 5 is provided, but instead of such a configuration, the following a, b , C. a. A configuration in which the supply operation and the unloading operation of the object are manually performed manually. b. A configuration in which the unloading device 6 is provided, the unloading operation is performed automatically, and the supply operation of the target object is performed manually. c. A configuration in which the supply device 2 is provided, the supply operation is performed automatically, and the operation of unloading the target object is performed manually. In addition, a. And c. In the configuration described above, the evaluation result of the object by the image processing means may be displayed on the display means so that the worker can recognize the evaluation result.

(5) In the above embodiment, the case where cucumber K is used as an object to be measured is exemplified. However, the present invention is not limited to cucumber K, but can be applied to various fruits and vegetables such as eggplant, tomato, and persimmon.

[Brief description of the drawings]

FIG. 1 is an overall perspective view of a cucumber sorting device.

FIG. 2 is a front view of a cucumber sorting device.

FIG. 3 is a side view of a cucumber sorting device.

FIG. 4 is a front view of the unloading device.

FIG. 5 is a sectional view of a supply device.

FIG. 6 is a sectional view of an unloading device.

FIG. 7 is a sectional view of a measuring unit.

FIG. 8 is a control block diagram.

FIG. 9 is a plan view of an unloading device.

FIG. 10 is a control flowchart of an evaluation process.

FIG. 11 is a flowchart of a control operation.

FIG. 12 is an explanatory diagram of an operation of an evaluation process.

FIG. 13 is an explanatory diagram of the operation of the evaluation processing.

FIG. 14 is an explanatory diagram of the operation of the evaluation processing.

FIG. 15 is a plan view showing an operation state at an imaging location.

FIG. 16 is an explanatory diagram of an operation for obtaining a carry-in target position.

FIG. 17 is a control flowchart of an unloading operation.

FIG. 18 is a control flowchart of equal class determination processing.

[Explanation of symbols]

 3 Conveyor 42 Imaging means 50 Reference line 51 Virtual line 100 Image processing means P3, P4 Reference point Q Bend G Center of gravity θ Tilt angle

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Atsushi Kitamachi 64 Ishizu-Kitacho, Sakai City, Osaka Prefecture Inside Kubota Sakai Factory

Claims (4)

[Claims] 1. An image pickup means for picking up an image of a long object to be measured and outputting image pickup information for measurement, and an image processing means for image-processing the image pickup information of the image pickup means to evaluate the object to be measured. Wherein the image processing means obtains a reference line along a longitudinal direction of the measured object based on imaging information of the imaging means, and a longitudinal direction of the measured object. In each of the setting regions located at both ends, a reference point on the contour at which the separation distance from the reference line is the maximum is obtained, and a virtual line connecting each of the reference points at both ends in the longitudinal direction, and the object to be measured. Among the contours, is a device for measuring fruit and vegetables configured to evaluate the maximum separation distance between the reference point and the contour located on the imaginary line side as a curve of the object to be measured. . 2. The method according to claim 1, wherein the image processing unit calculates a moment feature amount of the image of the measured object, and calculates a center of gravity of the image of the measured object, and a tilt angle of the measured object with respect to a reference direction on the captured image. The method according to claim 1, wherein a straight line passing through the position of the center of gravity and having the inclination angle with respect to the reference direction is obtained, and the straight line is used as the reference line to evaluate the object to be measured. Measuring device for fruits and vegetables. 3. The image processing means determines two points that are located on the contour of the image of the object to be measured and that are most separated in a direction along the reference line, The fruit and vegetable measuring device according to claim 1 or 2, wherein the measuring device is configured to evaluate a separation distance in a direction along the length as a length of the object to be measured. 4. A transport conveyor for transporting the object to be measured from a supply location through an imaging location to a take-out location, wherein the imaging means captures an image of the workpiece at the imaging location. The fruit and vegetable measuring device according to any one of claims 1 to 3, which is configured.