JP2003024875A - Sorting device of material and sorting method thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a sorting device of material having a means which arithmetically processes optical information received by sensors precisely and quickly, and a sorting method thereof. SOLUTION: This sorting device of material is provided with a plurality of sensors 6 which obtain the same information from material 3 individually to each other and an operating means 9 which respectively binarizes the information obtained by the respective sensors 6 and performs logical operation on the basis of the binarized data, and features that the material is sorted on the basis of the operation result.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an object sorting apparatus and a sorting method for sorting heat-resistant glass from a cullet containing heat-resistant glass and soda glass.

[0002]

2. Description of the Related Art In glass manufacturing factories, glass pieces for recycling (cullet) are used as a glass material. However, when heat-resistant glass is mixed with soda glass which constitutes an ordinary bottle, melting treatment is not possible. It becomes an obstacle,
The heat-resistant glass has been separated and eliminated in advance.
Conventionally, as a technique for separating such heat-resistant glass, it is known to use an emission spectrum by irradiation with laser light (for example, Japanese Patent Laid-Open No. 10-34088).

After the heat-resistant glass is removed, the cullet of soda glass is subjected to color discrimination, and the cullets of similar colors are collectively processed. Conventionally, as a color discrimination technique applied in such a case, visible light (wavelength 3
It is known to utilize the transmittance of 80 to 780 nm (Japanese Patent Laid-Open No. 10-19681).

[0004]

The above-mentioned conventional technique has a problem in that it is possible to discriminate an ideal cullet sample having a fixed shape, but it is not possible to discriminate an actual cullet with high accuracy. . Since the actual transmitted or reflected light of the cullet includes disturbance light due to reflection, refraction, etc. of light generated on the crushed surface, it is not possible to expect high-precision discrimination only by simple area irradiation and light reception.

Further, stains including label residue are generated on the cullet, but the stains cannot be excluded from the cullet removal.

A technique for separating glass, ceramics, plastics, and stains by comparing two images using visible light and ultraviolet light (Japanese Patent Laid-Open No. 2000-338046).
However, the types of glass currently produced are diversifying more and more, and there are general glass and heat-resistant glass, etc. that contain pigments that need to be sorted when recycled, and the conventional method does not specify the type of glass. The distinction is difficult.

Therefore, an object of the present invention is to provide an object selecting device and a selecting method that enable high-accuracy selection when selecting objects.

[0008]

The invention according to claim 1 is
A plurality of sensors that individually acquire object information, binarize the information acquired by each sensor, and an arithmetic means that performs a logical operation based on the binarized data, and an object is selected based on the operation result And means for doing so.

According to a second aspect of the present invention, an irradiation means for irradiating a plurality of conveyed objects with light, a plurality of sensors for individually acquiring light transmitted through or reflected by the objects, and each sensor. Each of the optical information obtained in 1. is binarized, and an arithmetic means for performing a logical operation based on the binarized data and a means for selecting an object based on the arithmetic result are provided.

According to a third aspect of the present invention, in the first or second aspect, the sensor is a scanning type sensor or an array type sensor.

According to a fourth aspect of the present invention, a process of individually obtaining object information by a plurality of sensors, and binarizing the information obtained by each sensor, and performing a logical operation based on the binarized data. The method is characterized by including a process and a process of selecting an object based on the calculation result.

According to a fifth aspect of the present invention, a process of irradiating each of a plurality of conveyed objects with light, and a process of individually acquiring information of light transmitted through or reflected from the object by a plurality of sensors. The optical information obtained by each sensor is binarized, and a logical operation is performed based on the binarized data, and an object is selected based on the operation result.

[0013]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing a cullet sorting device. This cullet separating device is provided with a hopper 1 for quantitatively supplying cullet, a conveyor 2 for conveying the supplied cullet to a separating position and dropping it, and a cullet (granular object) 3 conveyed from the conveyor 2 and falling. And a sorting container 4 with a partition for separately storing.

At the position where the cullet 3 is conveyed and dropped, an illuminating device 5 for irradiating visible light is provided as irradiating means for irradiating the cullet 3 which is being conveyed and dropped with light.
CCD cameras (sensors) 6 that receive the transmitted light of
a, 6b, 6c are provided. Reference numerals 7a, 7b and 7c are filters, and by the filters 7a to 7c, for example, only red (R) light is input to the CCD camera 6a,
Only green (G) light is input to the CCD camera 6b, and CC
Only blue (B) light is input to the D camera 6c. The sensor may be a scanning sensor or an array sensor.

The received light signals from each of the CCD cameras 6a to 6c are fetched into the judging circuit (calculating means) 9 via the input signal line 8.

In this embodiment, the judgment circuit 9 executes the arithmetic processing according to FIG. 2, and for example, the cullet is a light-colored general glass (M), a heat-resistant glass (T), a dark-colored general glass (A), It is classified into five types: opaque foreign matter (F) and light-colored general glass with label residue (MZ).
And are eliminated.

Next, the calculation processing procedure in the determination circuit 9 will be described.

In FIG. 2A, CCD cameras 6a-6c correspond to the inputs 1, 2, 3 respectively. In this judgment circuit 9,
In FIG. 2B, input 2 from the raw data obtained from the image of input 1
2C, the raw data obtained from the image is subtracted to obtain the inter-image calculation data (RG), and the inter-image calculation data (RG) and the images of the inputs 1 and 3 are obtained in FIG. 2C. Each raw data is binarized.

For example, when the cullet is a light-colored general glass (M), as shown in FIGS. 3A to 3D, raw data (optical information) of each input 1, 2, 3 and the inter-image calculation data are used. (RG) is required. These raw data are
The brightness of the light captured by each CCD camera 6a, 6b, 6c is shown, and in FIGS. 3A to 3D, (x1
・ A total of 225 addresses from y1) to (x15 ・ y15)
The brightness is displayed by numbers on the screen divided into addresses. In FIGS. 3A to 3D, each address indicates the same position.

When the cullet is heat-resistant glass (T), dark-colored general glass (A), opaque foreign material (F), and light-colored general glass with label residue (MZ), illustration is omitted, but each is shown in FIG. 3A. ~ Each input 1 corresponding to Fig. 3D
Raw data (optical information) of 3 to 3 and calculation data (RG) are obtained.

Binarization operation is performed on these raw data according to (1) to (4).

(1) With respect to the threshold value Th = 230, the raw data of the input 3 is calculated and arranged with 1 above and 0 below.

For example, when the cullet is light-colored general glass (M), the raw data of FIG. 3A is arranged into binarized data 1 and 0 as shown in FIG. 4A. In FIG. 4A, the portion of the binary data 1 is determined as the portion where the light from the illumination device 5 directly enters the camera without passing through the object, and is defined as the background light.

(2) With respect to the threshold value Th = 90 as a reference, the raw data of the input 1 is calculated and arranged with 1 above and 0 as below.

For example, when the cullet is light-colored general glass (M), the raw data of FIG. 3B is arranged into binarized data 1 and 0 as shown in FIG. 4B. In FIG. 4B, the portion of the binary data 1 is defined as light-colored general glass or heat resistant glass.

(3) The raw data of the input 1 is calculated and arranged with the threshold value Th = 5 as a reference, 0 above and 1 below.

For example, when the cullet is light-colored general glass (M), the raw data of FIG. 3C is organized into binarized data 1 and 0 as shown in FIG. 4C. In FIG. 4C, the portion of the binary data 1 is defined as an opaque foreign substance.

(4) The inter-image calculation data (R-G) is calculated by setting the threshold value Th = 40 as a reference and setting the above value as 1 and the below value as 0.

For example, when the cullet is light-colored general glass (M), the raw data of FIG. 3D is arranged into binarized data 1 and 0 as shown in FIG. 4D. In FIG. 4D, the portion of the binary data 1 is defined as additional information regarding the heat resistant glass.

The case where the cullet is a light-colored general glass (M) has been described above, but the cullet is a heat-resistant glass (T), a dark-colored general glass (A), an opaque foreign substance (F), and a light-colored glass. Label residue on general glass (M
In the case of Z) as well, according to the same procedure, each of the inputs 1 to 3 corresponding to FIGS.
5A to 5D, 6A to 6D, and 7A corresponding to the binarized data 1 and 0 based on the raw data (optical information) of No. 3 and the operation data (RG). 7D and 8A to 8D are obtained.

As a result of the binarization calculation according to (1) to (4), the binarized data is 4 in the processing order as shown in FIG. 2D.
It is allocated in the order of the first digit to the first digit, and as shown in FIG. 2E, a binary code having the following regularity is obtained for every pixel of the image.

"1XXX" is a code indicating background light.

"0100" is a code indicating light-colored general glass (M).

"0101" is a code indicating heat resistant glass (T).

"001X" is a code indicating an opaque foreign substance (F).

"000X" is a code indicating the dark general glass (A).

When the cullet is a light-colored general glass (M), for example, a binary code having regularity shown in FIGS. 9A and 9B is obtained for every pixel of the image.
In the case of heat-resistant glass (T), dark-colored general glass (A), opaque foreign matter (F), and light-colored general glass with label residue (MZ), FIGS. 10A, 10B and 11 respectively.
As shown in A and FIG. 11B, FIG. 12A and FIG. 12B, and FIG. 13A and FIG. 13B, a binary code having regularity can be obtained.

For example, in FIG. 9A, "01
There are many binary codes of "00", and it is understood that this cullet is light-colored general glass (M). However, the cullet often has an irregular shape and is often soiled. When the cullet has an irregular shape or stains, ambient light is generated, which makes it difficult to discriminate logic in a region where the cullet exists.

Disturbances caused by irregular shapes and dirt are characterized by a narrow width. In this embodiment, for example, when two or more same binary codes are not consecutive in the X and Y directions,
As shown in FIG. 2F, this is excluded as a disturbance, the background light is naturally excluded, and only the effective cell within the contour is set as the determination target.

From FIG. 9A, for example, two or more binary codes which are not continuous in the X and Y directions, and the background light "1XX
When “X” is excluded, as shown in FIG. 9B, “010
A binary code of "0" and four binary codes of "0101" remain. That is, the code “0100” indicating the light-colored general glass (M) is 67, and the code “0101” indicating the heat resistant glass (T) is 4.

In the case of heat resistant glass (T), as shown in FIG. 10A, there are many binary codes of "0101", two or more binary codes which are not continuous in the X and Y directions, and the background light "1XXX". 10B, for example, as shown in FIG. 10B, for example, a code “0100” indicating four light-colored general glasses (M) and a code “0101” indicating 75 heat-resistant glasses (T) remain.

In the case of dark-colored general glass (A), FIG. 11A
11B, when there are many binary codes of “0001” and two or more binary codes that do not continue in the X and Y directions and the background light “1XXX” are excluded, FIG.
As shown in FIG. 5, for example, a code “0100” indicating four light-colored general glasses (M) and a code “0001” indicating 55 dark-colored general glasses (A) remain.

In the case of an opaque foreign material (F), as shown in FIG. 12A, there are many binary codes of "0010", two or more consecutive binary codes in the X and Y directions, and the background light "1XXX". 12B is eliminated, as shown in FIG. 12B, for example, a code “00” indicating an opaque foreign substance (F) of 57 is displayed.
10 ”remains.

Label residue on light-colored general glass (M
In the case of Z), as shown in FIG.
When there are many binary codes and two or more binary codes that do not continue in the X and Y directions and the background light "1XXX" are excluded, as shown in FIG. 13B, for example, 24 light-colored general glass (M ), A code “0101” indicating four heat resistant glasses (T), and a code “0010” indicating 42 opaque foreign matters (F).

Next, the valid cell is discriminated.

Heat-resistant glass (T) and opaque foreign matter (F) are excluded. On the other hand, label residue (MZ) is not excluded from the light-colored general glass (M), the dark-colored general glass (A), and the light-colored general glass.

In the determination of valid cells, as shown in FIG. 2G, the frequency of appearance of various binary codes in the valid determination cell is calculated, and the occupation rate of various binary codes is calculated. As a result of this calculation, if the degree of occupation of the binary code satisfies the following expression (1), it is determined to be general glass. a is a coefficient.

(M + A) × a> (F + T) (1) Explaining with the example of FIG. 9B, M = 67, A = 0, F = 0,
Since T = 4, these are substituted into the equation (1), and a =
When it is set to 0.2, 13.4> 4 is satisfied, and the formula (1) is satisfied.

In the example of the heat-resistant glass (T) shown in FIG. 10B, when substituting into the equation (1), 0.8> 75 is not satisfied, and in the example of the dark general glass (A) shown in FIG. 11B. ,
When substituting into the formula (1), 11> 0 is satisfied, and in the example of the opaque foreign substance (F) in FIG. 12B, when substituting into the formula (1), 0> 57 is not satisfied and the bright color system is not satisfied. In the example of FIG. 13B in which the label residue (MZ) is attached to the general glass, when substituting in the equation (1), 4.8> 46, which does not hold.

When the expression (1) is not satisfied, the following expression (2) is determined. When the formula (2) is established, the heat resistant glass (T) is checked based on the formula (3), and when the formula is established, it is determined that the heat resistant glass (T). If the formula (2) is not established, the check is made for the label residue (MZ) on the light-colored general glass based on the formula (4), and if it is established, it is determined that the label residue is present (MZ). If that is not the case, it is determined as an opaque foreign substance (F).

(T)> (F) (2) However, when (T) and (M + A) are mixed, (T) is not overlooked, and therefore (T) <(M + A) may be used for the determination.

(T)> (M + A) × c (3) (F) × b <(M + A) (4) where b and c are statistically determined arbitrary threshold values.

The presence of (M + A) highly suggests the possibility of label residue.

Ceramics rarely show the characteristic of (M + A).

In the example of FIG. 10B, as described above,
Equation (1) does not hold because 0.8> 75, and when substituting into equation (2), 75> 0 holds. Therefore,
Substituting into the equation (3), when c = 0.5, 75> 2 holds, and therefore, the glass is determined to be heat resistant glass (T).
In the example of FIG. 12B, as described above, the expression (1) is 0> 5.
7 is not satisfied, and when substituting into the equation (2), 57> 0
And is approved. Then, when substituting into equation (4), b
= 0.5, 10.5 <0.
It is determined to be an opaque foreign substance (F).

When the cullet is judged to be heat-resistant glass (T) or opaque foreign matter (F), as shown in FIG. 2H,
Elimination of the caret is performed.

That is, the caret exclusion command is output to the exclusion device 10 shown in FIG. This exclusion command is sent as a valve drive signal via the command signal line 11 to the drive unit 12
The drive device 12 opens and closes the solenoid valve 13 in response to the command. The solenoid valve 13 is incorporated in the air pipe 15 for supplying the exclusion primary air to the exclusion nozzle 14. When the primary air is supplied to the exclusion nozzle 14 by opening the valve, the air ejected from the exclusion nozzle 14 is excluded. The cullet 3 of the kind to be pressed is pressurized, so that the cullet 3 deviates from the route of spontaneous fall and is accommodated in the separation chamber 4 side of the sorting container 4.

In the example of FIG. 13B, as described above,
The expression (1) is not satisfied with 4.8> 46, and is substituted with the expression (2) with 42> 4. But,
Substituting into equation (4), if b = 0.5, then 21 <24
Therefore, it is determined that the label residue (MZ) is attached to the light-colored general glass. The label residue (MZ) on this general glass is not excluded.

When the conditions are not satisfied in the processing according to the equations (2) to (4), it is determined that the glass is ordinary glass. Coefficient a,
b and c are appropriately set and effective for avoiding erroneous determination for a sample having a label residue or a sample having a lot of disturbance.

In the present embodiment, a binary code having a certain regularity is obtained for every pixel of the image, a specific occupancy in the validity judgment cell of this binary code is calculated, and based on this, the caret is calculated. As the cullet is sorted,
High-precision sorting is possible compared to the conventional one.

In this apparatus, it is desirable to increase the number of digits of the binary code in order to improve the sorting accuracy. The number of digits may be increased by increasing the number of CCD cameras and increasing the number of inputs, or by subtracting or adding each input as shown in FIG. 2B. In FIG. 2H, the stricter the threshold value for the occupation rate is, the more preferable it is to improve the sorting accuracy.

In the above embodiment, the input data is optical information, but the present invention is not limited to this, and may be magnetic information, for example.

In this embodiment, as shown in FIG. 1, the conveyor 2 is installed in a horizontal state at an angle θ1.
In general, in order to make the irradiation light incident on the CCD camera 6 accurately, it is desirable that the irradiation light be transmitted through the flat cullet 3 in the vertical direction and then be incident on the CCD camera 6. In order to achieve this, in the present embodiment, the installation angle θ1 of the conveyor 2 depends on the conveyor speed and CCD.
The parabolic locus L1 of the cullet 3 emitted from the conveyor 2 is set appropriately in relation to the installation angle θ2 of the camera 6, and is set to be substantially orthogonal to the line L2 connecting the camera and the lighting device 5 at the point A. To be done.

Although the present invention has been described based on the embodiment, it is obvious that the present invention is not limited to this.

[0066]

As described above, according to the present invention, an object can be accurately classified based on the optical information received by the sensor.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a cullet sorting device according to an embodiment of the present invention.

2A to 2H are diagrams showing a data processing flow.

3A to 3D are views showing raw data obtained by a sensor.

4A to 4D are diagrams showing binarized data.

5A to 5D are diagrams showing binarized data.

6A to 6D are diagrams showing binarized data.

7A to 7D are diagrams showing binarized data.

8A to 8D are diagrams showing binarized data.

9A and 9B are diagrams showing binary codes.

10A and 10B are diagrams showing binary codes.

11A and 11B are diagrams showing binary codes.

12A and 12B are diagrams showing binary codes.

13A and 13B are diagrams showing a binary code.

[Explanation of symbols]

1 hopper 2 conveyors (granular objects) 3 cullet 4 sorting containers 5 Lighting equipment 6a, 6b, 6c CCD camera 7a, 7b, 7c filters 8 input signal lines 9 Judgment circuit (calculation means) 10 Exclusion device 11 Command signal line 12 Drive 13 Solenoid valve 14 Exclusion nozzle 15 Air piping

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] April 8, 2002 (2002.4.8)

[Procedure Amendment 1]

[Document name to be amended] Statement

[Name of item to be amended] Claims

[Correction method] Change

[Correction content]

[Claims]

[Procedure Amendment 2]

[Document name to be amended] Statement

[Correction target item name] Explanation of code

[Correction method] Change

[Correction content]

[Explanation of Codes] 1 Hopper 2 Conveyor 3 Cullet (granular object) 4 Sorting container 5 Illuminator 6a, 6b, 6c CCD cameras 7a, 7b, 7c Filter 8 Input signal line 9 Judgment circuit (arithmetic means) 10 Exclusion device 11 Command Signal line 12 Drive device 13 Solenoid valve 14 Exclusion nozzle 15 Air piping

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Ikuo Hiraga             2-21-10 Omorinaka, Ota-ku, Tokyo Hiraga             Machine Industry Co., Ltd. (72) Inventor Hiroshi Nobori             2-21-10 Omorinaka, Ota-ku, Tokyo Hiraga             Machine Industry Co., Ltd. F term (reference) 2G051 AA90 AB20 CA03 CA07 CB02                       CC07 DA01 EA08 EA11                 2G059 AA05 BB08 BB15 EE01 EE02                       EE13 GG10 HH02 JJ02 KK03                       KK04 LL04 MM01 MM05 MM09                 3F079 AB00 AD00 CA31 CA44 CB30                       CB34 CB35 CC03 DA12

Claims (5)

[Claims] 1. A plurality of sensors for individually obtaining object information, a computing means for binarizing the information obtained by each sensor, and a logical operation based on the binarized data, and a result of the computation. An object selection device comprising means for selecting an object based on the object selection device. 2. An irradiation unit for irradiating a plurality of conveyed objects with light, a plurality of sensors for individually acquiring light transmitted through or reflected by the object, and optical information acquired by each sensor. An object selecting device comprising: a binarizing unit, a computing unit that performs a logical operation based on the binarized data, and a unit that sorts an object based on the result of the computation. 3. The object selection device according to claim 1, wherein the sensor is a scanning sensor or an array sensor. 4. A process of individually acquiring object information by a plurality of sensors, a process of binarizing the information acquired by each sensor, and a logical operation based on the binarized data, and a result of the operation. And a step of selecting an object based on the above. 5. A process of irradiating a plurality of conveyed objects with light, a process of individually acquiring light information transmitted or reflected by the object by a plurality of sensors, and a light acquired by each sensor. An object selecting method comprising the steps of binarizing information and performing a logical operation based on the binarized data, and a step of selecting an object based on the operation result.