Optical discrimination device and optical sorting apparatus

Abstract

To provide an optical discrimination device and an optical sorting apparatus that execute light emitting control of appropriate illumination light on the basis of a mathematical analysis result without depending on an expert's experience and intuition, precisely discriminate an object, and further can sort it.SOLUTION: An optical discrimination device has: sensor means 10 capable of imaging an object (granular material 20) in a wavelength region of 200 to 2500 nm; multiple illumination means 11 that apply light to the object and have different irradiation wavelength regions; illumination control means 31 capable of controlling a light emitting mode of the illumination means 11; and discrimination means 30 for discriminating the quality of the object on the basis of captured data acquired by the sensor means 10. The illumination control means 31 sets a light quantity balance in the multiple illumination means 11 based on a multivariate analysis according to the quantity of the object that becomes a discriminated object.SELECTED DRAWING: Figure 6

Description

Technical Field

[0001] The present invention relates to an optical discrimination device and an optical sorting device capable of discriminating and sorting the type, state, etc. of an object.

Background Art

[0002] Conventionally, various techniques have been disclosed for identifying and discriminating the material of resins such as plastics. For example, Patent Document 1 discloses a technique of using electromagnetic waves from infrared rays to the X-ray region as a light source used in a sorting device.

[0003] Specifically, a configuration for identifying five types of plastic plates (vinyl chloride, polypropylene, urethane, polyethylene, polystyrene) using infrared light with a wavelength range of 2 to 20 μm, and a configuration for identifying the same five types of plastic plates using a XeCl excimer laser with a wavelength of 308 nm are disclosed.

[0004] Further, Patent Document 2 discloses a configuration including a plurality of illumination sources that output wavelengths in the range of about 400 nm to about 2000 nm as a light source of a classification device. Furthermore, Patent Document 3 discloses an illumination device that directs electromagnetic energy to an object or directs light of a specific wavelength to an object, and discloses that reflected light from at least one illumination device can be detected by an imaging device.

[0005] Also, Patent Document 4 discloses, as a technique for selecting an appropriate wavelength range according to the material to be sorted, an illumination unit that creates illumination light having a color distribution suitable for the type of defect of granular materials so that illumination light in an appropriate frequency band can be used for precise determination of granular materials. Specifically, it includes a multi-color light-emitting package incorporating a plurality of LED elements with different emission wavelengths, and is controlled to a light-emitting pattern that creates strong red-based illumination light when detecting "immature brown rice", and is controlled to a pattern that creates illumination light with strong red-based and green-based light intensities when detecting "stink bug damaged grains", etc. are described.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Summary of the Invention

Problems to be Solved by the Invention

[0007] However, for example, the technology disclosed in Patent Document 4 above selects illumination light in an appropriate wavelength range by a predetermined lighting pattern of a multicolor light-emitting package based on the opinions of a pre-specified skilled person using a manual operating tool, and it is a control by a non-scientific method that relies on the experience and intuition of the skilled person.

[0008]

Means for Solving the Problems

[0009] The invention according to (1) includes sensor means capable of imaging an object in a wavelength range of 200 to 2500 nm, a plurality of illumination means for irradiating the object and having different wavelength ranges for each irradiation, illumination control means capable of controlling the light emission mode of the illumination means, and discrimination means for discriminating the quality of the object based on the imaging data acquired by the sensor means. The illumination control means is an optical discrimination device characterized by setting the light quantity balance in the plurality of the illumination means based on multivariate analysis according to the quality of the object to be discriminated.

[0010] (2) The invention according to (2) has a light quantity balance generation means for generating the light quantity balance, and the light quantity balance generation means acquires high-dimensional sample data of object samples with different qualities in advance under a plurality of types of light quantity balances by the illumination means, and can generate the light quantity balance by reducing the dimension by performing multivariate analysis on the high-dimensional sample data. It is the optical discrimination device according to (1) above.

[0011] (3) The invention according to (3) is the optical discrimination device according to (1) or (2) above, characterized in that the plurality of illumination means includes a red light source, a green light source, a blue light source, and a near-infrared light source.

[0012] (4) The invention according to (4) is the optical discrimination device according to (1) or (2) above, characterized in that the plurality of illumination means are a plurality of near-infrared light sources with different wavelengths.

[0013] (5) to (7) The invention according to (5) to (7) is an optical sorting device characterized by including the optical discrimination device according to (1) to (4) above, and having an ejector means for sorting and removing the object based on the discrimination result of the object by the discrimination means. [Effect of the Invention]

[0014] According to the present invention, it is possible to perform appropriate light emission control of illumination light based on a mathematical analysis method without relying on the experience and intuition of skilled workers as in the prior art, and it is possible to accurately discriminate and further sort objects. [Brief Explanation of Drawings]

[0015]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Mode for Carrying Out the Invention

[0016] Hereinafter, an example of an embodiment of the present invention will be described with reference to the drawings. As shown in the schematic drawing of FIG. 1, the optical discrimination device 1 and the optical sorting device 2 in the present embodiment assume, as an object, for example, a granular object (granular material 20), and have an inspection unit for optically inspecting the quality of the granular material 20.

[0017] And as shown in the figure, the inspection unit has at least sensor means 10 capable of imaging the granular material 20 in a wavelength range of 200 to 2500 nm, and a plurality of lighting means 11 that irradiate the granular material 20 and have different wavelength ranges for each irradiation.

[0018] In addition, as shown in the schematic block diagram of FIG. 2, it has at least lighting control means 31 capable of controlling the light emission mode of the lighting means 11, and discrimination means 30 for discriminating the quality of the granular material 20 based on the imaging data acquired by the sensor means 10. And the lighting control means 31 can set the light quantity balance in the plurality of lighting means 11 based on multivariate analysis according to the quality of the granular material 20 to be discriminated.

[0019] More specifically, the sensor means 10 of the present embodiment can utilize a broadband image sensor. For example, it is possible to use a broadband image sensor (broadband monosensor) having spectral sensitivity from visible light to near-infrared light. By using such a broadband monosensor, it is not necessary to install a plurality of sensor means corresponding to a plurality of illumination means 11 having different wavelength regions, and cost reduction can be achieved.

[0020] Note that the sensor means 10 is not necessarily limited to a broadband monosensor. As the sensor means 10, a plurality of various image sensors capable of imaging in a predetermined wavelength region can be arranged, and the particulate matter 20 can be configured to be imaged in the wavelength region of 200 to 2500 nm described above.

[0021] Also, in the present embodiment, as described above, there are a plurality of illumination means 11 having different wavelength regions to be irradiated. Specifically, as the illumination means 11, at least a red light source (R), a green light source (G), a blue light source (B), and a near-infrared light source (NIR) are provided. And the light quantity balance in each light source is set by the illumination control means 31. That is, the light quantity in each light source is set respectively. Note that the types of the above-described light sources and their numbers are merely examples, and it is possible to arrange two or more arbitrary light sources.

[0022] The above-described light quantity balance is generated by the light quantity balance generation means 32. Specifically, in the light quantity balance generation means 32, high-dimensional sample data of particulate matter samples with different qualities are acquired in advance under a plurality of types of light quantity balances by the illumination means 11, and the light quantity balance is generated by reducing the dimension by performing multivariate analysis on the high-dimensional sample data.

[0023] Regarding the multivariate analysis in the above-described light quantity balance generation means 32, hereinafter, taking principal component analysis as an example, the case of using light sources (illumination means 11A to 11E) having five types of wavelength regions shown in FIG. 3 will be described.

[0024] First, as a pre - treatment, granular samples with different qualities are prepared in advance, and high - dimensional sample data for discriminating the quality of the granular materials are obtained under high - dimensional illumination conditions using the above five types of light sources. That is, as shown in FIG. 4, under a certain high - dimensional illumination condition of light quantities X1 to X5, granular samples with different qualities are imaged to obtain high - dimensional sample data.

[0025] Subsequently, principal component analysis is performed on the above high - dimensional sample data. FIG. 5 shows an example of the contribution rate of each principal component and the basic formula of principal component analysis. As shown in the figure, since the cumulative contribution rate up to the third principal component is about 90%, dimensionality reduction (low - dimensionalization) can be performed from five - dimensional to three - dimensional or lower principal components.

[0026] Then, based on the principal component coefficients obtained by principal component analysis, the light quantity balance (variables X1 to X5) of each lighting means 11A to 11E is changed (generated), and at the same time, by obtaining the imaging information of the granular material 20 to be discriminated, the imaging information can be obtained by reducing the dimension to a one - dimensional space represented by the first principal component as shown in FIG. 6, or a two - dimensional space represented by the first principal component and the second principal component, or a three - dimensional space represented by the first principal component, the second principal component, and the third principal component.

[0027] By the above - described processing, by specially controlling the light quantity balance of each lighting means 11A to 11E (for example, control such as only illumination pattern 1 in FIG. 6, or illumination patterns 1 and 2, or illumination patterns 1, 2, and 3), for example, the same quality discrimination performance as that under high - dimensional illumination conditions composed of all or a plurality of the reflected and transmitted light, reflected light, and transmitted light of a red light source (R), a green light source (G), a blue light source (B), and the reflected and transmitted light, reflected light, and transmitted light of a plurality of near - infrared light sources (NIR) can be obtained. That is, by balancing the light quantities of each lighting means 11A to 11E by multivariate analysis, it becomes possible to accurately discriminate the quality of the granular material 20 (for example, good products, defective products, foreign substances, etc.) under low - dimensional illumination conditions.

[0028] Through the above-described processing, since the imaging information of the granular material 20 can be represented in a space of three dimensions or less, it becomes possible to create a three-dimensional optical correlation diagram of each granular material 20 on the display and plot and display the imaging information. As a result, it becomes possible for an operator to easily input and edit the discrimination threshold value of the granular material 20.

[0029] (Optical sorting device) As described above, an example of the discrimination processing configuration of the optical discrimination device 1 in the present embodiment has been described. However, the optical discrimination device 1 can be provided in a known sorting device for granular materials or the like to form an optical sorting device 2.

[0030] The optical sorting device 2 includes, for example, an inclined chute which is a transfer means for the granular material 20 inside the device body, and an inspection unit which performs an optical inspection by including at least the sensor means 10 and the illumination means 11 described above for the granular material 20 falling from the chute, and an ejector means which is provided below the inspection unit and can blow off defective products of the granular material 20 by air blowing, for example.

[0031] Further, as shown in FIG. 1, the control unit 3 of the optical sorting device 2 is provided with an ejector control unit 33, and the operation of the ejector means 41 is controlled via an ejector drive circuit 40. That is, based on the discrimination result of the granular material 20 by the discrimination means 30 described above, it becomes possible to sort and remove the granular material 20 by the ejector means 41.

[0032] (Other embodiments) As described above, an embodiment of the optical discrimination device and the optical sorting device of the present invention has been described. However, the present invention is not necessarily limited to the above-described embodiment, and includes, for example, the following modification examples.

[0033] For example, it is possible to use rice as an object to be discriminated. That is, it is possible to discriminate and sort good-quality white rice from defective colored rice, lightly roasted rice, and bran among the rice to be discriminated.

[0034] Further, the object to be discriminated is not limited to the above-mentioned rice, and can be other cereal grains such as wheat, beans, and nuts, resin pieces such as pellets and beads, pharmaceuticals, ores, and fine articles such as shirasu. When sorting the raw materials to be discriminated into good products and defective products, or removing foreign substances mixed in the raw materials, etc., the optical discrimination device and the optical sorting device of the present invention can be effectively applied.

[0035] In the above-described embodiment, an embodiment using principal component analysis as an example of multivariate analysis has been described. However, it is not necessarily limited to such an analysis method, and other methods such as linear discriminant analysis (LDA) that can be used as a classification model can be used as long as they can reduce the dimensions.

[0036] As described above, the embodiments and some modified examples of the present invention have been described. However, these descriptions are for facilitating the understanding of the present invention and do not limit the present invention. The present invention can be changed and improved without departing from its gist, and equivalents thereof are included in the present invention. Also, within the range where at least a part of the above-described problems can be solved, or at least a part of the effects can be achieved, combinations or omissions of each component described in the claims and the specification are possible.

Explanation of Reference Numerals

[0037] 1 Optical discrimination device 2 Optical sorting device 3 Control unit 10 Sensor means 11 (11A, 11B, 11C, 11D, 11E) Lighting means 20 Granular material 30 Discrimination means 31 Lighting control means 32 Light quantity balance generation means 33 Ejector control unit 40 Ejector drive circuit 41 Ejector means

Claims

1. sensor means capable of imaging an object in a wavelength range of 200 to 2500 nm, a plurality of illumination means for irradiating the object and having different wavelength ranges for each irradiation, illumination control means capable of controlling the light emission mode of the illumination means, discrimination means for discriminating the quality of the object based on the imaging data acquired by the sensor means, and having the illumination control means is configured to set the light quantity balance in the plurality of illumination means based on multivariate analysis according to the quality of the object to be discriminated An optical discrimination device characterized by this.

2. having light quantity balance generation means for generating the light quantity balance, the light quantity balance generation means is configured to acquire high-dimensional sample data of object samples with different qualities in advance under a plurality of types of light quantity balances by the illumination means, and reduce the dimension by performing multivariate analysis on the high-dimensional sample data to generate the light quantity balance The optical discrimination device according to claim 1, characterized by this.

3. the plurality of illumination means are including a red light source, a green light source, a blue light source, and a near-infrared light source The optical discrimination device according to claim 1 or 2, characterized by this.

4. the plurality of illumination means are configured as a plurality of near-infrared light sources with different wavelengths The optical discrimination device according to claim 1 or 2, characterized by this.

5. equipped with the optical discrimination device according to claim 1 or 2, and having ejector means for sorting and removing the object based on the discrimination result of the object by the discrimination means An optical sorting device characterized by this.

6. equipped with the optical discrimination device according to claim 3, and having ejector means for sorting and removing the object based on the discrimination result of the object by the discrimination means An optical sorting device characterized by this.

7. equipped with the optical discrimination device according to claim 4, and having ejector means for sorting and removing the object based on the discrimination result of the object by the discrimination means An optical sorting device characterized by this.

Images