JPS6041583A - Method of classifying and deciding class selector - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a grading device for sorting objects to be sorted, such as fruits, by size, and more particularly to a classification determination method for the grading device.

Conventionally, in this type of class sorting device, for example, the conveyor belt is connected in the conveyance direction (X direction), forms a substantially flat front end surface in the conveyance direction, and can be selectively tilted toward the − side of the conveyance direction. The objects to be sorted, such as fruits, which have been conveyed by a conveying member (hereinafter simply referred to as "RA board-shaped conveying member"), are sorted from a specific direction using an appearance detection means such as a COD camera. The appearance of a predetermined part of an object to be sorted, such as fruit, was detected and sorted according to a predetermined class.

Therefore, when detecting the appearance, it is necessary to make sure that the predetermined part of the object to be sorted, such as the fruit, falls within the limited area of the COD camera.
It was necessary to correct the mounting position for each item to be sorted, such as the individual shims, by using 1,000 steps, etc. However, such a method not only requires a considerable number of skilled workers, but even those skilled in the art must use a keyboard-shaped conveyor film that is driven at a considerably high speed.
There is a natural limit to the ability to efficiently supply a considerable amount of materials to be sorted by adjusting the placement posture.

Therefore, the present invention has been made in view of the above-mentioned problems of the prior art, and its purpose is to control the sorting of fruits, etc., by hand on the placing section. It is an object of the present invention to provide a classification determination method for a classification device that does not require modification of the placement and is capable of detecting the appearance of objects to be sorted, such as fruits, from an unspecified direction and performing -CO classification.

A feature of the present invention for achieving the above object is to detect the external appearance of supplied objects to be sorted from a predetermined direction, and to carry out class sorting according to the results of the detection. A method for determining the classification of a device, which uses a predetermined class to determine the crystallization (selected and classified) of the object that can be detected at all times, regardless of the appearance of the object to be sorted. 1) Calculate the value of the selected part of the object to be sorted by calculating the value of the selected part of the object by calculating from the detected appearance of The present invention provides a classification determination method for a roughness class sorting device that compares the set value with a set value to determine which R-class the object to be separated belongs to. The present invention will be explained in detail below with reference to the drawings.

FIG. 1 is a diagram showing an example of a circuit configuration according to an embodiment of the present invention, and FIGS. Figures 2 (C) and (2) are diagrams showing the external appearance of a predetermined portion by calculation from the external appearance of the object to be sorted shown in Figures 2 (A) and (B). FIG.

At J-3 in FIG. 1, reference numeral 18 denotes an illumination device 20 that provides an appropriate amount of light so that the object to be sorted conveyed by the keyboard-shaped conveyance member 4 is captured on the image receiving surface of the CCD camera 1;
is a camera control, which controls the drive of the CCD camera 1. 27 It converts the detection signal given from the CCD camera 1 via the camera control 20 into a digital signal and formats it to the A/D, 〇〇 smoothing. A claim memory 29 stores the above-mentioned detection information stored in the frame memory 28 and outputs the image shown in FIG. 2 (a) or (b), respectively. As shown in the figure, there is a short diameter calculator that calculates the value of a predetermined part by calculation, and 30 is the value of the short diameter of an approximate ellipse with a constant width for so-called mortar-shaped objects to be sorted In other words, the value of the diameter of the approximate circular cross section), and for the so-called ampan-shaped objects to be sorted, the long axis of the approximate ellipse with a certain width b (in other words, the value of the diameter of the approximate circular cross section) are determined for each predetermined class. a classification discrimination memory which stores fixed data in the storage area and retrievably stores any of the fixed data in the type fy of the object to be sorted in response to a command from the classification discrimination device 31;
Reference numeral 31 indicates a computation 1 related to the minor axis or major axis of the approximate ellipse of the image obtained by the diameter/minor axis calculator 29 and compares the information with the fixed data given from the classification discrimination memory 30 to determine the object to be sorted. A classification discriminator that determines to which class the shape dimension belongs to a predetermined portion (minor axis or major axis of the approximate ellipse); The moving distance information of the conveying member 4, etc.
Eat 'I'! The information given from the position detection signal generator 1'12 that detects the position of the object to be sorted in the S part 4-1x is input, and the information given from the position detection signal generator 1'12 for detecting the position of the object to be sorted is inputted to the controller 35 and 71 of the control unit i~.
A timing device 40 inputs a large amount of information to the roll 41, 40 is a pulse [n] - 35, and 35 is a predetermined corner handling gate based on the above information obtained from the timing device 45 and the classification judgment information given from the classification discriminator 31. 8 (a) 8 ([-1)
, 8 (C), <2)... Controllers 1 to 36 control the opening and closing of the gates 11 and 36 for controlling the opening and closing of the gates 11 and 36 for setting the exit gates of the sorted materials into predetermined classes. Setting device, 43a
, /1.3b ・-・・- is class gate 8 (a),
8 (b) Detect the position of the object to be sorted on the transport part shrine 4 with 21 moving keys placed on the transport start end side and send the position detection number 18 to the gate controller 35 respectively. Isa to give!
A position detection signal device 37 amplifies the drive command signal given from the gate controller 35 and supplies it to the solenoid/I4 for a predetermined class group 1; 2 a buffer relay for the A/D;
This is a classification device that includes a C smoothing 27, a diameter/minor diameter sieve 29, a classification discriminator 31, a goo 1 controller 35, and the like.

The operation of the above structure will be explained below using FIGS. 2A to 2D as Aj1.

Figure 2 (a) shows the image carried by the CCD camera 1 and the camera control +
1- to the major axis/minor axis calculator 29 via the A/D, AGC smoothing 27 and frame memory 28! Among the image information that can be obtained, 4 f! There is a type b illustrated in Category 1.

Of these, FIG. 2(A) (2) is a 1hA image of the object to be sorted placed in a situation where the overall shape can be roughly grasped. If the mounting position is as shown in the figure, since the II+3 image is an approximate ellipse, the major axis, zero axis calculator 29 calculates the values from the CCD camera 1 to camera controller 1-roll 20, Δ/D, AGC smoothing 27, and It is possible to calculate the major axis and minor axis of the approximate ellipse based on the image information provided via the frame memory 28, and to derive the values of the major axis and minor axis, respectively. Even when the major axis/minor axis calculator 29 recognizes from the image information provided from the CCD camera 1 via the camera control 20 etc. that the placement status of the objects to be sorted is as shown in FIG. It is possible to perform almost the same arithmetic processing as in FIG. 2(a) (2). Therefore thirteen”;
Also in this case, the classification device 2 can grasp the overall shape of the object to be sorted. However, the material to be sorted is shown in Figure 2 (a) ■
As is clear from the figure, the platform placed in the state shown in Fig. nm is possible because it falls within the detection area of the camera 1, but the major axis of the approximate ellipse is outside the detection area of the CCD camera 1. Therefore, the major axis and minor axis aft O device 29 is
/D, the above-mentioned shadow given through the AGC smoothing 27, etc. (Depending on the image information, it is not possible to calculate the major axis of the approximate ellipse by 0111. However, the major axis/minor axis calculator 29 , ΔGG smoothing 27 and the like, the parts that can always be used for sieving from the image information shown in (a) - () in FIG. It is clear from what has been explained so far that it is the diameter.

Therefore, when classifying the so-called open-type objects to be sorted, such as lemons and cucumbers, the major axis/minor axis calculation unit 29 is used regardless of the orientation in which the objects to be sorted are placed on the keyboard-shaped conveyance part shrine 4. is CCD
From the image information provided by the camera 1, it is sufficient to exclude only the diameter of the approximate circular cross section (minor axis of the approximate ellipse) by a method described later. It is sufficient to set the value of the approximate circular diameter of the object to be sorted having a certain area and store it as fixed data.

Next, in FIG. 2 (b), an image is taken with the CCD camera 1 and the camera control 20. Among the image information given to the major axis/minor axis calculator 29 via the A/D, AGC smoothing 27, and frame memory 28, for example, the keyboard shape of the so-called Anpan-shaped sorted object represented by unshu oranges, persimmons, etc. Conveying member 4
This diagram illustrates four typical placement situations on soil.

Of these, FIG. 2(B) (2) is a photographed image of an object placed in a situation where the overall shape of the object to be sorted can be roughly grasped.

If the mounting position is as shown in the figure, the image is an approximate ellipse, so the major axis/minor axis calculator 29 uses image information given from the CCD camera 1 via the camera control 20, A/D1ΔGC smoothing 27, and frame memory 28. It is possible to calculate the major axis and minor axis of the approximate ellipse based on , and exclude the major axis value and the minor axis value, respectively. The major axis/minor axis calculator 29 recognizes from the image information provided from the CCD camera 1 via the camera controller 1-roll 20 that the loading status of the object to be sorted is as shown in FIG. In this case, it is possible to perform almost the same arithmetic processing as in FIG. 2(b) (2). Therefore, in the above case as well, the classification device 2 can grasp the overall shape of the objects to be sorted. However, when the objects to be sorted are placed in the state shown in Figure 2 (B) ■ and Figure 2 (B) ■, the major axis of the approximate ellipse (that is, the diameter of the approximate ellipse cross section) is As it is clear from the above, it enters the detection area of CCD camera 1, so t)
However, the short axis of the approximate ellipse is outside the detection area of the camera 1. Therefore, the major axis/minor axis calculator 29 calculates Δ/D, A G C from the CCD camera 1.
It is not possible to exclude only one diameter of the approximate ellipse by the above-mentioned irregular image information obtained through the smoothing 27 or the like. However, the major axis/minor axis calculator 29 is CC
Among the image information shown in FIG. It is clear from what I just explained that it is the diameter of the approximate circular cross section of the formula.

Therefore, in the so-called Anpan-type class sorting of unshu mandarin oranges, Yawa 11, etc., the keyboard-shaped conveyance section +4 of the sorted materials is used.
iTi to 40! Irrespective of the same posture, the major axis is 91
The diameter calculator 29 calculates the shadow 191 given by the CCD camera 1.
Diameter of approximate circular cross section (long axis of approximate ellipse) is always selected from information
On the other hand, the classification discrimination memory 30 stores the approximation circle no.
All you need to do is set i++ and store it as fixed data (8).

Figure 2 (c) can be used, for example, in Figure 2 (a) or 1. JFigure 2 (
The so-called egg-shaped or ampanned material to be sorted as shown in (b) is passed through the major axis/minor axis generator 29 provided in the classification device 2.
/l< This is an example of the Buno method for sieving of the approximate circle arbitrary value used when classifying and sorting by predetermined grade. Major axis and minor axis 1ti: The O device 29 detects an arbitrary point PX+ on the outer periphery of the image detected by the COD camera '1, and another arbitrary point Pu on the outer periphery of the image other than this point PX+,
Connect PI2, PI3...pan with a straight line. After the above work is completed, the major axis and minor axis director 29 is moved to PX +
Pu, PX + PI2. PX + PI3.・
-... px I PI n are each calculated, and each calculation result is temporarily stored. Next, the major axis/minor axis calculator 29 calculates any point PX2 on the outer periphery -L of the image other than the above 1]×1, and calculates a point other than the point PX2 on the outer periphery of the image.
Any other point P2 + , P22 , P23
....P2n are connected with straight lines. After the above-mentioned work is completed, the major axis/minor axis calculator 29 stores PX2P21゜西了■■, 臺了TING富...PX2P2n, respectively, and temporarily stores the respective calculation results. Major axis minor axis calculator 29
In this way, for each point on the outer periphery of the image detected by the CCD camera 1, the quality of the line segment connecting the point and the other points on the outer periphery is calculated in order. At this point, all the calculation data stored up to that point are retrieved and compared, and the values of Pxn, pHlnMaX, that is, the value of the major axis of the approximate ellipse are determined. Figure 2 (2) is Figure 2 (
This is an example of a method used when calculating the short axis of the approximate ellipse based on the long axis value of the approximate ellipse calculated in step c). The major axis/minor axis calculator 29 calculates the shadow of another object (an arbitrary point Py1 on the outer circumference of this image and this point based on the image information), which is given from the CCD camera 1 via A/θ, AGC smoothing 27, etc. Find a point Pl+ on the outer periphery of the image where the line segment connecting the point yl with a point other than PV+ is approximately perpendicular to the major axis, and calculate PVf1]+
Play the length of +14-J. Similarly, the major axis/minor axis calculator 29 calculates Py 2 PI2, Py 3 PI3-...what'
Vn P n+nc)) Each value is extracted and temporarily stored. The major axis/minor axis calculator 29 takes out all the calculation data stored in J3 up to that point, performs a comparison operation, and calculates and reads the value of Pyn Pmn Max, that is, the value of the minor axis of the approximate ellipse.

The major axis/minor axis calculator 29 is an embodiment of the present invention, and calculates the value of a predetermined portion of the object to be sorted using a wire method such as that shown in (1) above, and outputs information on these calculated values to the classification discriminator 31. do.

When the classification discriminator 31 recognizes that the information given from the major axis/minor axis calculation unit 29 is related to so-called egg-shaped materials to be sorted, such as lemons and Japanese cucumbers, the classification discriminator 31 selects the so-called egg-shaped materials from the classification discrimination memory 30. Fixed data regarding the object to be sorted is called up, and the fixed data and the calculated value data are compared to determine which class the object to be sorted belongs to. The classification discriminator 31 outputs the judgment result to the controller 35, and the gate controller 35 controls the class 1 to size the object to be sorted into a predetermined class gate. Become. 4. The method for using -15 dot lines is just one example of the present invention, and it goes without saying that the present invention is not limited to this example.

As explained above, according to the present invention, a part to be detected that can be detected at any time regardless of the appearance of the object to be sorted is selected according to the type of the object to be sorted, and a predetermined class of the object to be sorted is selected. Set the value of the selected part for each case, calculate the value of the selected part of the object by calculating from the detected appearance of the object described in 61. It was decided to compare the set value with the above-mentioned value to determine to which class the object to be sorted (1) f1 belongs. No need to correct the mounting position.

Also, since the detection of the evening view is limited to detection from a specific direction, it is difficult to detect the appearance detection means. The selection becomes easier.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an example of a circuit configuration according to an embodiment of the present invention, and FIGS. Figures 2 (c) and (d) showing the external appearance are similar to figures 2 (a) and (
This figure shows a method (1) for deriving the shape and size of a predetermined portion by deriving from the object to be sorted 9111 shown in (b). 1... Appearance detection means (CCD camera) 2... Classification device (control) placement) Figure 2 (A) Figure 2 (C) Figure 2 (Port 2 Figure 2 (D)

Claims (1)

[Scope of Claims] A classification determination method for a class sorting device that detects the external appearance of supplied objects to be sorted from a predetermined direction and classifies them according to predetermined classes according to the detection results, the method comprising: Select a part that can be detected at any time regardless of the appearance of the object to be sorted according to the type of the object to be sorted, set the value of the selected part for each predetermined class to be classified, and From the detected appearance of the object, calculate the value of the selected part of the object to be sorted, and combine this calculation result with the set (l[1
1. A classification determination method for a class sorting device, characterized in that it is determined which class the object to be sorted belongs to by comparison.