Recognition device

Abstract

The present application provides an identification device. The identification device includes: a plurality of collection units configured to collect scattered light from a plurality of test objects; a first light splitting unit configured to disperse light from part of the plurality of collection units; a second light splitting unit configured to disperse light from the remaining part of the plurality of collection units; an imaging unit configured to acquire a first spectrum projected from the first light splitting unit and a second spectrum projected from the second light splitting unit, the imaging unit including a plurality of light receiving elements arranged in at least a first direction; and an acquisition unit configured to acquire information about the test objects based on output signals from the imaging unit. One of the wave numbers of the first spectrum and the wave numbers of the second spectrum increases in the first direction, and the other of the wave numbers of the first spectrum and the wave numbers of the second spectrum decreases in the first direction.

Description

Technical Field

[0001] One aspect of the embodiments relates to an identification device configured to identify properties (such as composition and ingredients) of a test object based on scattered light from the test object. Background Technology

[0002] A recognition device is known that uses spectral analysis to optically identify the properties of test objects. The recognition device is positioned along a transport path through which multiple test objects are transported for purposes such as inspecting products and sorting waste.

[0003] Spectroscopic analysis does not always require the atmosphere management processes associated with vacuum / reduced pressure treatments, atmosphere-controlled treatments, immersion treatments in solutions, and drying treatments used to limit throughput. Because spectroscopic analysis can identify the properties of analytes in ambient atmospheres, its application to waste resin classification has been explored in recent years.

[0004] As a form of spectroscopic analysis, infrared absorption spectroscopy and Raman scattering spectroscopy are known. In infrared absorption spectroscopy, the absorption spectrum of incident light from the test object, encompassing the infrared wavelength range, is obtained. In Raman scattering spectroscopy, the scattered spectrum of incident light from the test object, encompassing the ultraviolet wavelength range, is obtained. Raman scattering spectroscopy is less susceptible to light attenuation due to sample thickness and is therefore used to identify waste materials of varying sizes. Raman scattering spectroscopy, which disperses the scattered light, utilizes the Raman shifts characteristic of interatomic bonds in hydrocarbons. Therefore, Raman scattering spectroscopy is suitable for resin identification.

[0005] The intensity of Raman scattered light is several orders of magnitude lower than that of the elastic scattering component (Rayleigh scattering) contained in secondary light, allowing for methods that utilize converged primary light to illuminate the test object to improve detection sensitivity per unit area. A known classification device classifies test objects into target test objects and non-target test objects based on the detected spectrum and whether predetermined target conditions are met.

[0006] BUNSEKI KAGAKU Vol. 61, No. 12, pp. 1027-1032 (2012) discusses a waste resin sorting device including an identification mechanism. This identification mechanism projects the spectrum from a spectroscopic element onto a two-dimensionally arranged charge-coupled device (CCD) image sensor to simultaneously acquire the spectrum of the test substance across the entire observed wavenumber range without performing wavelength scanning. The identification device discussed in BUNSEKI KAGAKU Vol. 61, No. 12, pp. 1027-1032 (2012) projects the spectrum from the spectroscopic element along the longitudinal (row direction) of the CCD image sensor to acquire spectral images at high speed by reading the spectrum projected by the spectroscopic element in the column direction.

[0007] Meanwhile, due to the development of semiconductor manufacturing technology in recent years, 2K×4K image sensors, such as complementary metal-oxide-semiconductor (CMOS) image sensors, capable of high-definition and high-speed readout, are available. Utilizing such image sensors and beam-splitting elements capable of high-speed, high-definition imaging, compact recognition devices with high-speed and high-volume recognition capabilities can be provided. Japanese Patent Application Publication No. 2019-105628 discusses a recognition device including a beam-splitting element and a CMOS image sensor. The beam-splitting element disperses light from multiple light-collecting units in parallel, and the CMOS image sensor captures multiple spectra from the beam-splitting element in parallel. The recognition device discussed in Japanese Patent Application Publication No. 2019-105628 reduces size by integrating the device after multiple light-collecting units into a single beam-splitting element and a single two-dimensional imaging unit. Furthermore, the recognition device discussed in Japanese Patent Application Publication No. 2019-105628 uses a rolling shutter, rolling reset method to drive the CMOS image sensor to reduce the influence between adjacent spectra projected parallel in the column direction.

[0008] To increase the amount of recognition processed per unit time, the recognition device discussed in Japanese Patent Application Publication No. 2019-105628 is expected to employ a method that projects multiple spectra in multiple columns along the row and column directions of the image sensor. In this case, if the viewing angle and the number of mounted elements of the image sensor are increased, and the number of spectra per unit area of ​​the image sensor (projection density) is maintained, it is expected that the mutual interference between spectra projected onto adjacent regions will be reduced. However, it has been found that even when increasing the number of projections while maintaining the projection density of the spectra, the accuracy of the recognition processing may decrease. Therefore, it has been found that the recognition processing volume per unit time may not increase as expected while maintaining recognition accuracy. Summary of the Invention

[0009] According to one aspect of an embodiment, an identification device includes: a plurality of collecting units configured to collect scattered light from a plurality of test objects; a first beam splitting unit configured to disperse light from a portion of the plurality of collecting units; a second beam splitting unit configured to disperse light from the remaining portion of the plurality of collecting units; an imaging unit configured to acquire a first spectrum projected from the first beam splitting unit and a second spectrum projected from the second beam splitting unit, the imaging unit including a plurality of light receiving elements arranged at least in a first direction; and an acquisition unit configured to acquire information about the test objects based on an output signal from the imaging unit, wherein one of the wavenumbers of the first spectrum and the second spectrum increases in the first direction, while the other of the wavenumbers of the first spectrum and the second spectrum decreases in the first direction.

[0010] According to another aspect of the embodiments, an identification device includes: a transport unit comprising a plurality of transport routes and configured to transport a plurality of test objects in parallel; a first collection unit corresponding to a route among the plurality of routes and a second collection unit corresponding to the other routes among the plurality of routes; a first beam splitting unit configured to disperse light from the first collection unit; a second beam splitting unit configured to disperse light from the second collection unit; an imaging unit configured to acquire a first spectrum projected from the first beam splitting unit and a second spectrum projected from the second beam splitting unit, the imaging unit comprising a plurality of light receiving elements arranged at least in a first direction; and an acquisition unit configured to acquire information about the test objects based on an output signal from the imaging unit, wherein the wavenumber of the first spectrum increases in the direction of increase of the address of the light receiving element in the first direction relative to the address of the light receiving element in the first direction opposite to the direction of increase of the wavenumber of the second spectrum relative to the address of the light receiving element in the first direction.

[0011] Other features of this disclosure will become clear from the following description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0012] Figure 1A and Figure 1B This is a diagram illustrating a schematic construction of an identification device according to a first exemplary embodiment.

[0013] Figure 2A and Figure 2B This is a diagram showing the projection area of ​​the spectrum on the imaging unit according to the first exemplary embodiment, and the correspondence between the projection wavenumber and the row direction address. Figure 2C This is a graph showing the projected wavenumbers.

[0014] Figure 3A and Figure 3B This is a diagram showing the projection area of ​​the spectrum on the imaging unit according to the second exemplary embodiment, and the correspondence between the projection wavenumber and the row direction address.

[0015] Figure 4A and Figure 4B This is a diagram showing the projection area of ​​the spectrum on the imaging unit according to the third exemplary embodiment, and the correspondence between the projection wavenumber and the row direction address.

[0016] Figure 5A , Figure 5B and Figure 5C This is a schematic diagram showing the Raman scattering spectra of acrylonitrile butadiene styrene (ABS) resin, polystyrene, and polypropylene, the projection regions of the spectra, and the correspondence between the projection wavenumber and the row direction address of the projection spectrum. Detailed Implementation

[0017] Various exemplary embodiments of this disclosure will now be described with reference to the accompanying drawings.

[0018] The following will refer to Figure 1A , Figure 1B , Figure 2A and Figure 2B A description of an identification device according to a first exemplary embodiment. Figure 1A and Figure 1B This is a diagram schematically illustrating the construction of the identification device 1000 according to this exemplary embodiment. Figure 2A It is shown Figure 1A A partial detailed view of the spectral information acquisition unit 100 of the identification device 1000 shown. Figure 2B This is a diagram showing the projection of the spectra 280sa (odd) and 280sb (even) onto the imaging unit 170. Furthermore, Figure 2C This is a projection wavenumber-light receiver address mapping diagram showing the relationship between the number (row direction number) of the light receiver element arranged in the row direction 172r of the imaging unit 170 and the wavenumber of the spectrum projected in the row direction 172r. The term "light receiver element number" is also called the row direction address of the light receiver element or the row direction number of the light receiver element.

[0019] exist Figure 1A In this diagram, the z-direction corresponds to the vertical direction, i.e., the direction of gravity, and the x-direction corresponds to the conveying direction dc. Furthermore, the y-direction corresponds to the conveying width direction dw, and the xy-plane corresponds to the horizontal plane.

[0020] The conveying width direction dw is consistent with the direction that is parallel to the conveying surface 200S and perpendicular to the conveying direction dc.

[0021] Identification device

[0022] like Figure 1A As shown, the identification device 1000 includes an illumination unit 22. The illumination unit 22 illuminates a test object 900i transported in the transport direction dc with illumination light 220 to focus the illumination light 220 onto the test object 900i. The test object 900i is fed to the transport unit 200 by the feeder 500 and transported by the transport unit 200 along the transport direction dc. The illumination light 220 is also referred to as "converging light 220" or "primary light 220". The identification device 1000 also includes a light collecting unit 20 and an acquisition unit 30. The light collecting unit 20 corresponds to the illumination unit 22 and collects the scattered light from the test object 900i. The acquisition unit 30 acquires identification information for identifying the attributes of the test object 900i based on the light collected by the light collecting unit 20.

[0023] In addition, such as Figure 1A As shown, the identification device 1000 includes a conveying unit 200 and a discrimination device 300. The conveying unit 200 includes a conveyor belt that conveys the test object 900i in the x-direction at a conveying speed vc. The discrimination device 300 is disposed downstream of the conveying unit 200 in the conveying direction dc. Figure 1B This is a diagram showing the conveyor unit 200 and multiple conveyor tracks TR-p (p = 1 to 4). Figure 1B Corresponding to Figure 1A The projection diagram of the light collection unit 20 of the identification device 1000 and the discrimination device 300 is shown with plane A-A' as the projection plane. Figure 1B Section B-B' in the middle corresponds to Figure 1A The schematic construction diagram shown is shown below.

[0024] Next, we will refer to Figure 2A , Figure 2B and Figure 2C The spectral information acquisition unit 100 of the identification device 1000 is described in detail below. The spectral information acquisition unit 100 relates to the features of this exemplary embodiment and includes a spectroscopic element and an imaging unit.

[0025] Spectral information acquisition unit

[0026] The identification device 1000 includes a spectral information acquisition unit 100, which is configured to acquire spectral information about the light collected from the test object 900i. The spectral information acquisition unit 100 is a unit that acquires the Raman shift based on the wavenumber difference between the Raman scattered light contained in the secondary light from the test object 900i and the excitation light contained in the primary light.

[0027] like Figure 1A and Figure 2AAs shown, the spectral information acquisition unit 100 includes illumination units 22a(p) and 22b(p) and light collection units 20a(p) and 20b(p). Illumination units 22a(p) and 22b(p) illuminate the test object 900i with illumination light 220a(p) and illumination light 220b(p), respectively. Light collection units 20a(p) and 20b(p) collect secondary light from the test object 900i. Figure 1A This shows the view from the y-direction and... Figure 1B A schematic diagram of the xz plane corresponding to the transport route TRp where the route number p is odd. Figure 1A The diagram shows a feeder 500, a camera 76p, a light collecting unit 20a(p), an illumination unit 22a(p), and an air nozzle 330p arranged on a transport path TRp in odd-numbered columns p. Figure 1A Although not shown in the diagram, the feeder 500, camera 76p, light collection unit 22b(p), irradiation unit 20b(p), and air nozzle 330p are also arranged on the transport path p corresponding to the even-numbered column p.

[0028] Figure 1B The plan view shown is from the z-direction. Figure 1A A schematic diagram of the construction of plane A-A' in the middle.

[0029] The light collection unit 20a(p) and the illumination unit 22a(p) are arranged coaxially, and the illumination unit 22a(p) is optically coupled to the light source 25a, which includes a laser light source, via optical fiber 130p. Similarly, the illumination unit 22b(p) and the light collection unit 20b(p) are arranged coaxially. For simplicity, the suffix (p) of the route number p indicating the transmission route TRp will be omitted below.

[0030] Light collection units 20a and 20b are optically coupled to the spectral image acquisition unit 10, enabling the spectral information acquisition unit 100 to acquire optical information about the material contained in the test object 900i. Collection unit 27a includes an illumination unit 22a and a light collection unit 20a, and collection unit 27b includes an illumination unit 22b and a light collection unit 20b.

[0031] According to this exemplary embodiment, the suffixes a and b at the end of each number are associated with different collection units 27a and 27b, respectively. Specifically, the suffixes (a, b, c, d...) at the end of each number are used to associate with multiple components equipped to process light collected in parallel from different light collection units. The suffixes a and b are associated with transport routes TRp with odd route numbers p and transport routes TRp with even route numbers p.

[0032] Collection Unit

[0033] Figure 2AThis is a schematic diagram illustrating an example of the construction of the spectral information acquisition unit 100. The spectral information acquisition unit 100 includes a collection unit 27a, which comprises an illumination unit 22a for illuminating one test object 900i with light, and a light collection unit 20a for collecting Raman scattered light from the test object 900i. Similarly, the spectral information acquisition unit 100 includes a collection unit 27b, which comprises an illumination unit 22b for illuminating another test object 900i with light, and a light collection unit 20b for collecting Raman scattered light from the other test object 900i. When viewed from the dichroic mirror 250a, the illumination unit 22a and the light collection unit 20a are coaxially arranged on the test object side (object side), and even if the test object 900i has a height difference or is tilted, positional deviation is unlikely to occur between the center of the illumination point and the center of the collected scattered light beam. Similarly, when viewed from the dichroic mirror 250b, the illumination unit 22b and the light collection unit 20b are coaxially arranged on the test object side (object side).

[0034] Collection units 27a and 27b guide light collected from different test objects to beam-splitting elements 150a and 150b via optical fibers 190a and 190b, respectively, to identify different test objects in parallel. Beam-splitting elements 150a and 150b will be described below.

[0035] According to this exemplary embodiment, collection units 27a and 27b are respectively arranged on transport routes TRp and TRp+q. Using this arrangement, the shared imaging unit 170 performs parallel identification processing on two test objects transported in parallel to different transport routes TRp and TRp+q. As a variation of this exemplary embodiment, the following form is also included within the scope of this disclosure: collection units 27a and 27b are respectively located upstream and downstream of the same transport route TRp and collect light from two different test objects.

[0036] Irradiation unit

[0037] like Figure 1A and Figure 2A As shown, irradiation units 22a and 22b each have a focal length DF to form a focal plane 65 at a predetermined distance WD from the transport surface 200S, and are both arranged above the transport unit 200. Irradiation units 22a and 22b are arranged to focus the irradiation light 220 toward the upper surface of the test object 900i, thereby increasing the scattering intensity of Raman scattering light, which is several orders of magnitude lower than Rayleigh scattering light. The unit including irradiation unit 22a and light source 25a can be referred to as an "irradiation optical system". Light sources 25a and 25b can be replaced by using a common light source (not shown).

[0038] like Figure 2AAs shown, illumination units 22a and 22b respectively include objective lenses 260a and 260b, dichroic mirrors 250a and 250b, collimating lenses 230a and 230b, cylindrical lenses 240a and 240b, and reflecting mirrors 210a and 210b. Objective lenses 260a and 260b may include convex lenses, collimating lenses, concave lenses, and / or zoom lenses.

[0039] Synthetic quartz can be used as the material for glass lenses such as collimating lenses 230a and 230b, cylindrical lenses 240a and 240b, and objective lenses 260a and 260b. For the aforementioned lenses illuminated with high-output light from light sources 25a and 25b, lenses containing synthetic quartz as the lens material reduce the background components of fluorescence and Raman scattered light.

[0040] Objective lenses 260a and 260b serve as converging lenses in illumination units 22a and 22b, respectively, to converge light from light sources 25a and 25b onto test objects 900i (i=0, 1, 2…). Objective lenses 260a and 260b form a focal plane 65 at a position corresponding to the numerical aperture NA at a focal length DF from objective lens 260, and have a focal diameter… The focal spot (not shown) and focal depth ΔDF. The height of the focal plane 65 relative to the transport surface 200S is set with reference to the height hi distribution of the test object 900i (i = 1, 2, 3...).

[0041] Collimating lenses 230a and 230b, as well as cylindrical lenses 240a and 240b, reduce scattering of light emitted from light sources 25a and 25b to shape the light into parallel beams. Cylindrical lenses 240a and 240b can be other collimating optical elements, such as a pair of deformable prisms. Furthermore, wavelength filters, such as laser line filters, can be provided at various locations on the pupil surfaces of illumination units 22a and 22b. This adjusts the wavelength characteristics of the light emitted by illumination units 22a and 22b to illuminate the test object 900i.

[0042] like Figure 2A As shown, at least a portion of the illumination units 22a and 22b can be shared with the light collection units 20a and 20b. Since the light collection units 20a and 20b and the illumination units 22a and 22b are arranged coaxially according to this exemplary embodiment, the objective lenses 260a and 260b and the dichroic mirrors 250a and 250b are shared by the light collection unit 20 and the illumination unit 22.

[0043] light source

[0044] Light sources 25a and 25b are light sources that emit excitation light to irradiation units 22a and 22b via optical fibers 130a and 130b, respectively. The irradiation optical system for dispersing Raman scattered light uses laser light sources with excitation wavelengths from 400 nm to 1100 nm as light sources 25a and 25b. In Raman scattering, the excitation efficiency increases at shorter wavelengths, while the background fluorescence component decreases at longer wavelengths.

[0045] In other embodiments, the excitation wavelengths of the laser sources applied to light sources 25a and 25b are selected such that the difference in Raman shift between the target material and the non-target material is clear.

[0046] At least one of the wavelengths 532 nm, 633 nm, 780 nm, and 1064 nm can be used as the excitation wavelength of the laser source. Although the use of light sources 25a and 25b as the light source for irradiation units 22a and 22b is described herein, this exemplary embodiment is not limited to what is described herein, and other laser sources, such as semiconductor-excited solid-state lasers or gas lasers, can be used.

[0047] Light collection unit

[0048] Light collecting units 20a and 20b are arranged above the conveying surface 200S to collect secondary light from the upper surface of the test object 900i conveyed by the conveying unit 200. In other words, in order to collect secondary light from the upper surface of the test object 900i passing through the irradiation area, light collecting units 20a and 20b are arranged above the conveying unit 200 to face the irradiation area irradiated by irradiation light 220a and 220b from irradiation units 22a and 22b.

[0049] Light collecting units 20a and 20b respectively include objective lenses 260a and 260b, dichroic mirrors 250a and 250b, imaging lenses 270a and 270b, and optical fibers 190a and 190b. Similar to illumination unit 22, the objective lenses 260a and 260b of light collecting units 20a and 20b include convex lenses, collimating lenses, concave lenses, and / or zoom lenses. To reduce unwanted light in spectral measurements, light collecting unit 20 may include wavelength filters (such as bandpass filters or longpass filters) for reducing the excitation light component of the primary light.

[0050] Light collecting units 20a and 20b each employ objectives with large numerical apertures to improve light-gathering efficiency, or objectives with small numerical apertures to maintain working distance and depth of focus. More specifically, the SCHOTT B-270 with an effective lens diameter of 25 mm, a focal length of 20 mm, and a numerical aperture of 0.5 can be used as the objective lens.

[0051] Spectral image acquisition unit

[0052] like Figure 2A As shown, light from the two collection units 27a and 27b is guided in parallel through optical fibers 190a and 190b to the spectral image acquisition unit 10 according to this exemplary embodiment. The spectral image acquisition unit 10, starting from the collection units 27a and 27b, sequentially includes: imaging lenses 110a and 110b, long-pass filters 120a and 120b, beam splitters 150a and 150b, imaging lenses 160a and 160b, and an imaging unit 170. The suffixes a and b added to the components according to this exemplary embodiment correspond to the collection units 27a and 27b. The suffixes a and b can correspond to two test objects undergoing recognition processing in parallel.

[0053] Spectroradiometers 150a and 150b are arranged to disperse the light collected by collecting units 27a and 27b and project the spectrum onto imaging unit 170 along the row direction 172r of light receiving element 350.

[0054] like Figure 2A and Figure 2B As shown, the spectra 280sa (odd number) of collecting unit 27a (odd number) and the spectra 280sb (even number) of collecting unit 27b (even number) are projected onto different positions on the row direction 172r of the light receiving element 350. The symbols or numbers in parentheses are the numbers assigned to each collecting unit of the imaging unit 170. According to this exemplary embodiment, the collecting unit corresponds to... Figure 1B The route number p of the transport route TRp in the text.

[0055] Collection units 27a (odd numbers) are arranged corresponding to transport routes TRp (p=1, 3, 5, 7) with odd route numbers p. Specifically, collection units 27a(1), 27a(3), 27a(5), and 27a(7) are arranged corresponding to transport routes TRp (p=1, 3, 5, 7). Similarly, collection units 27b (even numbers) are arranged corresponding to transport routes TRp (p=2, 4, 6, 8) with even route numbers p. Specifically, collection units 27b(2), 27b(4), 27b(6), and 27b(8) are arranged corresponding to transport routes TRp (p=2, 4, 6, 8).

[0056] According to this exemplary embodiment, the spectra 28sa(1), 28sa(2)...27a(7) and 27b(8) of the eight collection units 27a(1), 27b(2)...27a(7) and 27b(8) are projected onto the imaging unit 170 in a matrix of four rows and two columns.

[0057] Furthermore, the spectrum 280sa (odd number) of collection unit 27a and the spectrum 280sb (even number) of collection unit 27b are arranged in the row direction 172r, and there is a boundary between element numbers 1341 and 1361, such as Figure 2B As shown. In this case, the wavenumbers of spectra 280sa (odd) and 280sb (even) increase in opposite directions relative to the element address of the optical receiving element 350, as... Figure 2C As shown. In other words, in the row direction 172r, according to this exemplary embodiment, one of the wavenumbers of the spectrum 280sa (odd number) and 280sb (even number) increases while the other decreases, as... Figure 2C As shown.

[0058] Similarly, spectra 280sa (odd) and 280sb (even) are projected onto different positions on the imaging unit 170 in the row direction 172r. Spectra 280sa (1), 280sa (3), 280sa (5), and 280sa (7) are projected onto different positions on the column direction 172c of the imaging unit 170. Similarly, spectra 280sb (2), 280sb (4), 280sb (6), and 280sb (8) are projected onto different positions on the column direction 172c of the imaging unit 170. The row and column directions of the arrangement of the light receiving elements 350 are perpendicular to each other and are also referred to as the first direction and the second direction intersecting the first direction.

[0059] The low wavenumber ends of the spectra 280sa (odd) and 280sb (even) are close to each other at adjacent element numbers 1341 and 1361 in the row direction 172r. Specifically, relative to the address of the light receiving element 350 in the row direction 172r, the projected spectra 280sa (odd) and 280sb (even) increase in opposite directions. This is achieved by... Figure 2A The mirror-image arrangement of the spectroscopic elements 150a and 150b is shown to achieve an increase in the spectral spectral 280sa (odd number) and 280sb (even number) in opposite directions, as shown. Figure 2C The projection wavenumber-element address correspondence is shown in the diagram. Specifically, the beam splitter 150a projects the spectral component corresponding to the larger diffraction angle to the smaller row direction address of the light receiving element 350. On the other hand, the beam splitter 150b is arranged to project the spectral component corresponding to the larger diffraction angle to the larger row direction address of the light receiving element 350.

[0060] The identification device 1000 according to this embodiment includes collection units 27a(p) and 27b(p), i.e., light collection units 22a(p) and 22b(p), corresponding to the odd and even numbers of the route number p of the transport route TRp. The identification device 1000 according to this exemplary embodiment includes a plurality of optical fibers 190a serving as a first light guiding unit. The plurality of optical fibers 190a guides multiple light beams from the plurality of light collection units 22a(p), which are part of the plurality of light collection units, to a beam splitter 150a(p). Similarly, the identification device 1000 according to this exemplary embodiment includes a plurality of optical fibers 190b(p) serving as a second light guiding unit. The plurality of optical fibers 190b(p) guides multiple light beams from the plurality of light collection units 22b(p), which are the remaining part of the plurality of light collection units, to a beam splitter 150b(p). Beam-splitters 150a(p) and 150b(p), arranged on different transport paths TRp and having opposite projection wavenumber directions relative to the element address of imaging unit 170, may also be referred to as "first beam-splitter" and "second beam-splitter". Similarly, beam-splitters 150a(p) and 150b(p) may also be referred to as "first beam-splitter unit" and "second beam-splitter unit". In other words, the first beam-splitter unit and the second beam-splitter unit respectively include beam-splitters 150a and 150b having different cone angles relative to imaging unit 170.

[0061] The technical implications of back-projection in the wavenumber direction will be described below.

[0062] The wavenumber dependence of Raman scattered light will be described below. For example... Figure 5A As shown, the wavenumber dependence of Raman scattered light intensity may be negatively tilted to the right on the wavenumber axis, exhibiting a near-monotonically decreasing trend. To increase the number of objects to be identified that can be processed in parallel, multiple spectra can be arranged parallel in the positive direction along the row direction of imaging unit 170. The positive direction is also referred to as the "same direction" or "common direction." In this case, as... Figure 5B As shown, the high wavenumber and low wavenumber ends of two adjacent spectra are close to each other, and a large intensity difference is found in the detection signals of the Raman scattered light from adjacent light receiving elements.

[0063] like Figure 5C As shown in the wavenumber-component address mapping diagram, parallel projections of two spectra in the forward (same) direction can cause interference between adjacent spectra. This interference between adjacent spectra is also known as crosstalk. Crosstalk between adjacent spectra includes at least one of electrical and optical effects.

[0064] Imaging unit 170 includes low-impedance circuitry, but the electrical constant distribution on the circuitry is not zero, and there is more than just a slight electrical constant distribution. Furthermore, high-speed readout is required. Additionally, imaging unit 170 is not a perfect light absorber for Raman scattered light components, or excitation light components and fluorescence components that were not successfully removed by the filter and are therefore incident. Therefore, electrical crosstalk may affect the vicinity of the intended projection position through the transmission of some charge photoelectrically converted by light receiving element 350 along the circuitry equipped with parasitic capacitors. On the other hand, optical crosstalk may affect the vicinity of the intended projection position as halo pattern noise through the concentric return and re-incidence of backscattered light from light receiving unit 171. In either the electrical or optical effects, the low-intensity detection signal acquired by light receiving elements with low Raman scattered signal intensity is often affected by crosstalk.

[0065] According to this exemplary embodiment, the high wavenumber ends of the two spectra 280sa (odd) and 280sb (even) are close to each other in the row direction 172r. In other words, the fingerprint regions of the spectra 280sa (odd) and 280sb (even) with relatively high average signal intensity of Raman scattered light are close to each other. The two spectra 280sa (odd) and 280sb (even) have very small differences in Raman signal intensity and are adjacent to each other under the condition that the spectra 280sa (odd) and 280sb (even) are unlikely to affect each other.

[0066] Therefore, compared with the parallel method where two spectra are adjacent to each other in the row direction under the condition of large intensity difference between the detection signals of Raman scattered light, the mutual influence of the spectra of the imaging unit 170 in the wavenumber direction is reduced by using the identification device 1000 according to this exemplary embodiment.

[0067] Imaging unit

[0068] For imaging unit 170, an imaging device including two-dimensionally arranged light-receiving elements is employed, such as a charge-coupled device (CCD) image sensor or a complementary metal-oxide-semiconductor (CMOS) image sensor. Although the plurality of light-receiving elements 350 of imaging unit 170 are arranged in a matrix, in the case of a triangular arrangement, the row direction and column direction are associated with two of the three axes, or with one of the three axes and a composite direction obtained by combining the remaining two axes.

[0069] While the transport unit 200 transports the test object 900i, the identification device 1000 identifies the attributes of the test object 900i, and the discrimination device 300 described below identifies the test object 900i based on the identification result. Therefore, in order to increase the throughput of the discrimination process, the transport speed vc of the transport unit 200 is increased. The spectra 280sa (odd) and 280sb (even) projected onto the imaging unit 170 originate from the Raman scattered light from the test object 900i moving on the transport surface 200S. Therefore, when the test object 900i being transported is in an irradiated area irradiated by the irradiation light 220 from the irradiation unit 22, spectra 280sa (odd) and 280sb (even) are formed on the imaging unit 170. For example, when the transport speed vc of the transport unit 200 is 2 m / sec and the length of the test object 900i in the transport direction dc is 10 mm, the imaging unit 170 can detect the spectrum formed by the Raman scattered light from the test object 900i in 5 milliseconds or less. Therefore, in one embodiment, the imaging unit 170 has a high frame rate. An imaging unit with a high frame rate is a CMOS image sensor, and a CMOS image sensor is suitable for use as the imaging unit 170.

[0070] Furthermore, since the intensity of the Raman scattered light from the test object 900i is very low as described above, the intensity of the light incident on the light receiving element 350 of the imaging unit 170 is also very low. Therefore, in one embodiment, an imaging unit 170 with high sensitivity to the wavenumber region is used at the acquisition of the spectrum 280sa (odd) and 280sb (even). Generally, a rolling shutter image sensor has a simpler pixel structure and a higher aperture ratio than a global shutter image sensor, and the size of the photoelectric conversion element can be increased. Therefore, using a rolling shutter image sensor increases sensitivity and dynamic range. In addition, the rolling shutter image sensor has a simple pixel structure, which has the advantage of lower cost than a global shutter image sensor. For the above reasons, a rolling shutter CMOS image sensor is used as the imaging unit 170 according to this exemplary embodiment.

[0071] For the imaging unit 170, a rolling reset image sensor can be used to sequentially reset the light receiving elements 350 row by row. This maximizes the exposure time of each row of light receiving elements 350 and increases sensitivity.

[0072] The imaging unit 170 according to this exemplary embodiment has a cropping reading function for reading specific rows in the light receiving unit 171. In the light receiving unit 171, light receiving elements 350 are arranged in two dimensions along the row direction 172r and the column direction 172c. Therefore, when the morphological information acquisition unit 70 detects that the test object 900i has reached the light-collectible area where the light collecting unit 20 can collect light, the imaging unit 170 reads specific rows in the light receiving unit 171 corresponding to the light collecting unit 20.

[0073] Imaging unit 170 includes readout circuit 173, horizontal scanning circuit 174, vertical scanning circuit 175, and output circuit 176. Imaging unit 170 sequentially reads signals from a plurality of pixels arranged in a matrix, row by row. Vertical scanning circuit 175 selects a row in light receiving unit 171 and drives the selected row. Readout circuit 173 reads the signal output from the pixels of the row selected by vertical scanning circuit 175 and transmits the readout signal to output circuit 176 based on the control of horizontal scanning circuit 174. This is how reading is performed in the main scanning direction (row direction). Furthermore, the row selected by vertical scanning circuit 175 is shifted, and readout circuit 173 performs reading in the main scanning direction based on the control of horizontal scanning circuit 174. This process is repeated, and the selected row is shifted in the sub-scanning direction (column direction), thereby reading signals from the entire light receiving unit 171. The readout signal is output as an output signal through output terminal 177 of output circuit 176 to material information reference unit 180 provided outside imaging unit 170. In this scenario, scanning is performed at high speed in the main scanning direction, while the scanning speed in the sub-scanning direction is lower than that in the main scanning direction.

[0074] Imaging lenses 110a and 110b convert the light collected by light collection units 20a and 20b and guided by optical fibers 190a and 190b into parallel light. Long-pass filters 120a and 120b reduce the excitation light component contained in the collected light and transmit the Raman scattered light component. Beam-splitters 150a and 150b disperse the spectral components obtained by fan-shaped dispersion of the collected light. In this case, beam-splitters 150a and 150b are arranged relative to imaging unit 170 such that their increasing directions are reversed on the projection wavenumber-element address map. In other words, beam-splitters 150a and 150b are arranged relative to imaging unit 170 such that the wavenumber directions of the projected spectra of beam-splitters 150a and 150b projected onto imaging unit 170 are reversed. In other words, beam-splitters 150a and 150b are arranged relative to imaging unit 170 such that the wavenumber changing directions of the projected spectra of beam-splitters 150a and 150b projected onto imaging unit 170 are reversed.

[0075] Imaging lenses 160a and 160b project the light dispersed by beam-splitter elements 150a and 150b onto imaging unit 170. Beam-splitter elements 150a and 150b are transmission gratings. Alternatively, reflection gratings can be used. In this case, a Rowland arrangement or a Czerny-Turner method is employed as the beam-splitter configuration. Beam-splitter elements 150a and 150b can also be referred to as "gratings 150a and 150b".

[0076] Taking into account the captured spectrum, the photoelectric conversion characteristics of the light receiving element 350 of the imaging unit 170, and the transmission characteristics of the optical system, the imaging unit 170 acquires spectral information Si about the test object 900i. Furthermore, the beam-splitting elements 150a and 150b can acquire polarization information, including circular dichroism and optical rotation dispersion, along with the spectrum.

[0077] According to this exemplary embodiment, beam splitter 150a is considered to be an optical element included in the first beam splitting unit. Furthermore, beam splitter 150b is considered to be another optical element included in the first beam splitting unit.

[0078] According to a first exemplary embodiment, the imaging unit 170 is equipped with a plurality of beam-splitting elements 150a and 150b to reverse the direction of wavenumber variation in the projected spectrum. According to a variant of the first exemplary embodiment, for one imaging unit, a single beam-splitting element is used to reverse the direction of wavenumber variation in the projected spectrum. This variant of the first exemplary embodiment is implemented by the following construction. The differences between the variant and the first exemplary embodiment will be described below. Two collection units, including light collection channels A and B, are provided such that a collimated beam obtained by collimating two collected beams from the two collection units into a small-diameter beam using an optical fiber collimator is incident on a single beam-splitting element (grating) at a predetermined distance d between them. From the output side of the single beam-splitting element (the side opposite to the incident side), two diffracted beams are emitted from positions d apart from each other, corresponding to the two small-diameter beams. One of the two diffracted beams forms an image on the imaging unit 170 through a single cylindrical lens A1. The other diffracted beam is relayed by two cylindrical lenses B1 and B2, and then forms an image on the imaging unit 170 through a third cylindrical lens B3. As described above, this variant is achieved by designing the focal point of the projection-side optical system.

[0079] For example, one diffracted beam forms an image on the imaging unit 170 through a cylindrical lens A1 with in-plane focal power, while the other diffracted beam forms an image on the imaging unit 170 through cylindrical lenses B1, B2, and B3. With the focal length of cylindrical lens A1 being 100 mm, the focal lengths of cylindrical lenses B1 and B2 can be set to 25 mm and 50 mm, respectively, so that two projection spectra are formed on the imaging unit 170 by reversing the wavenumber changes of the two projection spectra.

[0080] According to this variant, the incident area of ​​a single beam-splitting element (grating) and channel A on cylindrical lens A1 can be considered as the optical element of the first beam-splitting unit. Furthermore, the incident area of ​​a single beam-splitting element (grating) and channels B on cylindrical lenses B1 to B3 can be considered as the optical element of the second beam-splitting unit. In other words, according to this variant, the first and second beam-splitting units share a beam-splitting element (grating) with different illumination areas. In other words, the first and second beam-splitting units share beam-splitting elements, allowing beams from multiple collecting units to illuminate different positions.

[0081] Material Information Reference Unit

[0082] The spectral information acquisition unit 100 includes a material information reference unit 180, which is configured to acquire material information about the test object 900i based on the spectral information Si acquired by the spectral image acquisition unit 10. The material information reference unit 180 references a material database (not shown) storing Raman scattering reference data and acquires material information Mi identifying materials contained in the test object 900i based on the similarity between the spectral information Si and the reference data. The spectral information acquisition unit 100 stores at least one of the spectral information Si and the material information Mi in a first storage unit 60 via the instruction unit 40 described below.

[0083] Furthermore, the material database referenced by the material information reference unit 180 can be stored on a local server of the identification device 1000 or on a remote server accessible via the Internet or intranet.

[0084] As described above, the spectral information acquisition unit 100 acquires material information Mi about the mixture of materials, additives and impurities contained in the test sample 900i.

[0085] Morphological information acquisition unit

[0086] like Figure 1A and Figure 1BAs shown, the morphology information acquisition unit 70 includes a camera 76p, which is arranged such that the imaging field of view 700p overlaps with the transport path TRp (p=1 to 8) of the transport unit 200. The morphology information acquisition unit 70 also includes an image processing unit 78, which is configured to process the test object image captured by the camera 76p. The morphology information acquisition unit 70 acquires morphology information Fi about the test object 900i. Similar to the material information Mi, the morphology information Fi is information about the properties of the test object 900i.

[0087] Image processing unit 78 performs image processing including contrast and contour extraction, and acquires the length of each test object 900i in the transport direction dc, as well as the reflected color, shape, and material mixing degree of each test object 900i. In other words, image processing unit 78 is an element that performs processing to obtain information about the size of each test object 900i. Shape information acquisition unit 70 may include an optical interruptor and / or a laser interferometer (not shown) instead of camera 76p. Shape information acquisition unit 70 may be referred to as imaging unit 170. Furthermore, shape information acquisition unit 70 is an element selectively employed in recognition device 1000.

[0088] Acquisition Unit

[0089] like Figure 1A As shown, the acquisition unit 30 acquires identification information Di, which indicates whether the test object 900i is a target test object or a non-target test object, based on the material information Mi or spectral information Si acquired by the spectral information acquisition unit 100 and the morphological information Fi acquired by the morphological information acquisition unit 70. The acquisition unit 30 outputs the acquired identification information Di to the instruction unit 40.

[0090] In other words, the acquisition unit 30 identifies the properties of the test object 900i based on the Raman spectrum contained in the secondary light collected by the light collection unit 20. In other words, according to this exemplary embodiment, the acquisition unit 30 identifies the properties of each test object 900i based on the test object image acquired by the camera 76p and the Raman spectrum contained in the secondary light collected by the light collection unit 20.

[0091] As a variation, the spectral information acquisition unit 100 and the morphological information acquisition unit 70 according to this exemplary embodiment can be replaced by a hyperspectral camera or a multi-band camera capable of acquiring morphological information Fi and spectral information Si from captured images. Specifically, the identification device (not shown) according to the variation includes a detection system configured to acquire multidimensional data from which material information and morphological information can be read.

[0092] Control Unit

[0093] The identification device 1000 includes a control unit 400, which includes an instruction unit 40, a second storage unit 80, and a first storage unit 60. The instruction unit 40 controls the identification operation of the identification device 300 based on the attributes of each test item 900i. The second storage unit 80 stores the identification operation control conditions. The first storage unit 60 stores the attributes of each test item 900i. The control unit 400 includes a display unit 140, which provides a graphical user interface (GUI) through which the user can specify control conditions. The display unit 140 can display information acquired by the acquisition unit 30.

[0094] storage unit

[0095] The first storage unit 60 is configured to store identification information Di, material information Mi, spectral information Si, morphological information Fi, and test object 900i in a timed manner via the irradiation area for each test object 900i.

[0096] The second storage unit 80 is configured to store, for each test object 900i, control conditions for controlling the strength Is of the discrimination operation of the discrimination device 300 corresponding to the identification information Di. The control conditions may take the form of a referenced table, a general formula expressed algebraically, or statistical information learned through machine learning.

[0097] Instruction Unit

[0098] The instruction unit 40 estimates the transit time of the test object 900i through the processing area based on the identification information Di from the acquisition unit 30, and generates instructions to control the discrimination operation of the discrimination device 300, in which the discrimination device 300 identifies the test object 900i based on its material and size. The transit time of the test object 900i through the processing area can be estimated based on at least one of the signals from the morphology information acquisition unit 70, the spectral information acquisition unit 100, and the test object sensor (not shown) from the transport unit 200.

[0099] Discrimination device

[0100] like Figure 1AAs shown, the identification device 300 includes air nozzles 330p (p=1 to 8) and an identification control unit 340. The air nozzles 330p (p=1 to 8) discharge compressed air at a discharge rate and discharge flow rate for a predetermined discharge time. The identification control unit 340 controls a solenoid valve (not shown) of the air nozzles 330p. The identification control unit 340 receives control signals from the instruction unit 40 of the identification device 1000. The identification operation of the identification device 300 according to this exemplary embodiment includes the operation of discharging fluid. The fluid to be discharged includes air, dry nitrogen, inert gases (such as rare gases), liquids, and gas-liquid mixtures (aerosols). Based on the control signals from the instruction unit 40, the identification device 300 collects the test item 900i in the target collection basket 620 or the non-target collection basket 600 or 640 according to the properties of the test item 900i.

[0101] The fluid discharge device of the identification device 300 can be replaced by a gate that opens and closes at a predetermined angular velocity or a shutter that opens and closes at a predetermined speed. Furthermore, the components of the morphology information acquisition unit 70, the spectral information acquisition unit 100, and the identification device 300 of the identification device 1000 can be arranged at different positions along the conveying width direction dw of the conveying unit 200 to integrate the system and improve processing speed. The identification device 300 can be considered a component of the identification device 1000 and can be referred to as "identification unit 300".

[0102] Conveying unit

[0103] The conveying unit 200 is a conveying unit that conveys multiple test objects 900i (i=1, 2...) sequentially fed from the feeder 500 in the conveying direction dc (x direction in Figure 1) at a predetermined conveying speed vc. The conveying unit 200 and the feeder 500 together form a conveying unit for conveying the test objects 900i. The feeder 500 is arranged to form a supply section 550p upstream of the light collection units 22a(p) and 22b(p) for each conveying route TRp (p=1 to 8).

[0104] The conveying unit 200 according to this exemplary embodiment includes a conveyor belt that linearly conveys the test object 900i fed from the feeder 500 along the conveying direction dc at a conveying speed vc on the conveying surface 200S. As a variation, the conveying unit 200 may be replaced by a turntable feeder that helically conveys the test object to the outside, a vibratory feeder equipped with a vibrator for moving the test object in a predetermined direction, or a conveying roller composed of multiple rollers.

[0105] The transport unit 200 moves the test object 900i so that it passes through the imaging field of view 700p of the camera 76p. Therefore, the transport unit 200 can also be referred to as the placement section 200 of the shape information acquisition unit 70. Similarly, the transport unit 200 moves the test object 900i so that it passes through the effective light collection area 220p of the light collection units 20 (20a, 20b). Therefore, the transport unit 200 can also be referred to as the placement section 200 of the light collection units 20 (20a, 20b).

[0106] According to this exemplary embodiment, in the case of a conveyor belt, the conveying speed vc of the conveying unit 200 can be set to 0.1 m / s to 5 m / s.

[0107] Furthermore, the classification process of filtering the shape and size of the test object 900i as a feeding pretreatment of the feeder 500 is also an identification method according to a variant of the identification device 1000 according to this exemplary embodiment. The pretreatment is performed using a vibrating conveyor, a vibrating screen, or a particle crushing controller.

[0108] Next, refer to the following Figure 3A and Figure 3B A description of an identification device according to a second exemplary embodiment. Figure 3A This is a diagram showing the projection of the spectrum onto the imaging unit 170 according to the second exemplary embodiment. Figure 3B This is a projection wavenumber-light receiving element address mapping diagram according to this exemplary embodiment, which shows the relationship between the element number of the light receiving element 350 arranged in the row direction 172r of the imaging unit 170 and the wavenumber of the spectrum projected in the row direction 172r.

[0109] The identification device according to the second exemplary embodiment projects three columns of spectra parallel in the row direction, and can be considered a variant of the identification device 1000 according to the first exemplary embodiment. The identification device according to this exemplary embodiment differs from the identification device 1000 according to the first exemplary embodiment in that the identification device according to this exemplary embodiment projects three columns of spectra 280sa(3n-2), 280sb(3n-1), and 280sa(3n) parallel in the row direction 172r onto the imaging unit 170.

[0110] According to this exemplary embodiment, three beam-splitting elements (not shown) are arranged relative to the imaging unit 170 to reverse the tendency of the diffraction angles to be fan-shaped in the row direction 172r, thereby achieving the following: Figure 3B The relationship between the projected wavenumbers of adjacent spectra and the component addresses is shown.

[0111] Spectrum 280sa(3n-2) is projected onto optical receiving element 350 with element addresses 101 to 1301 in the row direction 172r, and spectrum 280sb(3n-1) is projected onto optical receiving element 350 with element addresses 1401 to 2601. For example... Figure 3B As shown, the high wavenumber ends of spectra 280sa(3n-2) and 280sb(3n-1) are close to each other, while the low wavenumber ends of spectra 280sa(3n-2) and 280sb(3n-1) are far apart from each other. Specifically, spectra 280sa(3n-2) and 280sb(3n-1) are projected onto imaging unit 170 such that the high wavenumber ends are closer to each other than the low wavenumber ends.

[0112] Similarly, spectrum 280sa (3n) is projected onto optical receiving element 350 with element addresses 2701 to 3901 in the row direction 172r. Figure 3B As shown, the low wavenumber ends of spectra 280sb(3n-1) and 280sa(3n) are close to each other, while the high wavenumber ends of spectra 280sa(3n-2) and 280sb(3n-1) are far apart from each other. Specifically, spectra 280sb(3n-1) and 280sa(3n) are projected onto imaging unit 170 such that the low wavenumber ends are closer to each other than the high wavenumber ends.

[0113] Specifically, spectra 280sa(3n-2) and 280sb(3n-1) are close to each other in the CH stretching region where the average signal intensity of the Raman scattered light is relatively low. In other words, under the condition that the difference in Raman signal intensity is small and spectra 280sa(3n-2) and 280sb(3n-1) are unlikely to affect each other, spectra 280sa(3n-2) and 280sb(3n-1) are parallel in the row direction 172r. Similarly, spectra 280sb(3n-1) and 280sa(3n) are close to each other in the fingerprint region where the average signal intensity of the Raman scattered light is relatively high. Under the condition that the difference in Raman signal intensity is small and spectra 280sb(3n-1) and 280sa(3n) are unlikely to affect each other, spectra 280sb(3n-1) and 280sa(3n) are parallel in the row direction 172r.

[0114] Therefore, compared with the method of arranging two adjacent spectra in parallel along the row direction under the condition that the intensity difference between the Raman scattered light detection signals increases, the identification device 1000 of this exemplary embodiment achieves a reduction in the mutual influence between the spectra of the imaging unit 170 in the wavenumber direction.

[0115] In the identification device according to the second exemplary embodiment, the crosstalk between adjacent spectra 280sa(3n-2), 280sb(3n-1) and 280sa(3n) is reduced compared to the forward projection method in the forward parallel projection spectrum.

[0116] Next, refer to the following Figure 4A and Figure 4B A description of an identification device according to a third exemplary embodiment. Figure 4A This is a diagram showing the projection of the spectrum onto the imaging unit 170 according to a third exemplary embodiment. Figure 4B This is a projection wavenumber-light receiving element address mapping diagram according to this exemplary embodiment, which shows the relationship between the element number of the light receiving element 350 arranged in the row direction 172r of the imaging unit 170 and the wavenumber of the spectrum projected in the row direction 172r.

[0117] This exemplary embodiment is similar to the first exemplary embodiment in that multiple spectra of light collected by multiple collection units 27a (odd number) corresponding to the transport path TRp of the odd-numbered column p are projected onto the imaging unit 170 by the beam splitter 150a in parallel and at a certain distance from each other in the column direction 172c. Furthermore, this exemplary embodiment is similar to the first exemplary embodiment in that light from the test object 900i on the transport path TRp of the even-numbered column p is collected in parallel, and the beam splitter 150b projects multiple spectra onto different positions in the column direction 172c of the imaging unit 170.

[0118] Furthermore, another similarity to the first exemplary embodiment is that adjacent spectra 208sa (odd) and 208sb (even) are oriented opposite to each other in the projection wavenumber-element address correspondence and have frequency bands such that spectra 208sa (odd) and 208sb (even) are projected to different positions in the row direction 172r.

[0119] On the other hand, this exemplary embodiment differs from the first exemplary embodiment in that the plurality of spectra 208sa (odd numbers) and the plurality of spectra 208sb (even numbers) are shifted from one another not only in the row direction 172r but also in the column direction 172c, and are alternately projected onto the imaging unit 170. Furthermore, this exemplary embodiment differs from the first exemplary embodiment in that the plurality of spectra 208sa (odd numbers) and the plurality of spectra 208sb (even numbers) have frequency bands such that the plurality of spectra 208sa (odd numbers) and the plurality of spectra 208sb (even numbers) are projected onto the overlapping positions of the addresses of the light receiving element 350 in the row direction 172r.

[0120] According to this exemplary embodiment, spectra 208sa (odd) and 208sb (even) are projected onto imaging unit 170 such that, in the row direction 172r, the high wavenumber ends of spectra 208sa (odd) and 208sb (even) are closer to each other than the low wavenumber ends of spectra 208sa (odd) and 208sb (even). Therefore, the intensity difference between the detected signals of the high wavenumber Raman scattered light from the closely spaced spectra 208sa (odd) and 208sb (even) is smaller, and crosstalk between adjacent spectral images is reduced.

[0121] Using the identification device according to the third exemplary embodiment, the spectra 280sa (odd number) and 280sb (even number) projected adjacently in the row direction 172r are in opposite directions in the projection wavenumber-optical receiver address correspondence. Therefore, compared with the forward projection method that projects the spectra in parallel in the forward direction, the identification device according to the third exemplary embodiment reduces crosstalk between spectra.

[0122] While this disclosure has been described with reference to exemplary embodiments, it should be understood that this disclosure is not limited to the disclosed exemplary embodiments. The scope of the appended claims should be interpreted in the broadest possible sense to encompass all such variations and equivalent structures and functions.

Claims

1. An identification device comprising: Multiple collection units are configured to collect scattered light from multiple test objects; The first beam-splitting unit is configured to disperse light from a portion of the collection units among the plurality of collection units; The second beam-splitting unit is configured to disperse light from the remaining portion of the plurality of collecting units; An imaging unit is configured to acquire a first spectrum projected from the first beam splitter and a second spectrum projected from the second beam splitter, the imaging unit comprising a plurality of light receiving elements arranged at least in a first direction; as well as The acquisition unit is configured to acquire information about the test object based on the output signal from the imaging unit. Wherein, one of the wavenumbers of the first spectrum and the second spectrum increases in the first direction, while the other of the wavenumbers of the first spectrum and the second spectrum decreases in the first direction, and Wherein, the first spectrum and the second spectrum are projected such that, in the first direction, the low wavenumber ends of the first spectrum and the low wavenumber ends of the second spectrum are closer to each other than the high wavenumber ends of the first spectrum and the high wavenumber ends of the second spectrum; or, the first spectrum and the second spectrum are projected such that, in the first direction, the high wavenumber ends of the first spectrum and the high wavenumber ends of the second spectrum are closer to each other than the low wavenumber ends of the first spectrum and the low wavenumber ends of the second spectrum.

2. The identification device according to claim 1, wherein, The first spectrum and the second spectrum are projected onto different locations on the imaging unit.

3. The identification device according to claim 1, wherein, The first spectrum and the second spectrum are projected onto different positions in the first direction.

4. The identification device according to claim 1, wherein, The plurality of light-receiving elements are arranged on both the first direction and the second direction intersecting the first direction, and the first spectrum and the second spectrum are projected onto different positions on the second direction.

5. The identification device according to claim 4, wherein, When the first direction is a row direction and the second direction is a column direction, the first spectrum and the second spectrum are projected onto the position where the row direction addresses of the light receiving element overlap.

6. The identification device according to claim 4, wherein, The first spectrum and the second spectrum are projected alternately in the first direction and the second direction.

7. The identification device according to claim 1, further comprising: A plurality of first light guiding units are configured to guide light from a portion of the plurality of collecting units to the first light splitting unit; as well as A plurality of second light guiding units are configured to guide light from the remaining portion of the plurality of collecting units to the second light splitting unit.

8. The identification device according to claim 1, further comprising: The placement section, on which the plurality of test objects are to be placed in the light collection area of ​​the plurality of collection units.

9. The identification device according to claim 8, wherein, The placement part is configured to move the test object in a predetermined direction.

10. The identification device according to claim 9, further comprising: A discrimination unit is configured to classify the plurality of test items and is arranged downstream of the collection unit in the predetermined direction.

11. The identification device according to claim 10, wherein, The acquisition unit controls the classification operation of the discrimination unit based on the information.

12. The identification device according to claim 9, wherein, The placement section includes a conveying unit having a predetermined direction and a width direction intersecting the predetermined direction, and the conveying unit includes multiple routes configured to convey the multiple test objects placed at different positions in the width direction in parallel.

13. The identification device according to claim 1, wherein, The first beam splitting unit and the second beam splitting unit respectively include a first beam splitting element and a second beam splitting element, and the cone angle of the first beam splitting element relative to the imaging unit and the cone angle of the second beam splitting element relative to the imaging unit are different from each other.

14. The identification device according to claim 1, wherein, The first beam splitting unit and the second beam splitting unit share a beam splitting element, such that light from the plurality of collecting units illuminates different positions of each other.

15. The identification device according to claim 1, further comprising: Multiple irradiation units are configured to irradiate the test object using light associated with the multiple collection units.

16. The identification device according to claim 15, wherein, The multiple illumination units are optically coupled to multiple light sources.

17. The identification device according to claim 1, wherein, The plurality of collection units collect Raman scattered light from the test object.

18. An identification device comprising: The transport unit includes multiple transport routes and is configured to transport multiple test objects in parallel; A first collection unit corresponding to one of the plurality of transport routes and a second collection unit corresponding to another of the plurality of transport routes; The first beam-splitting unit is configured to disperse light from the first collecting unit; The second beam-splitting unit is configured to disperse the light from the second collecting unit; An imaging unit is configured to acquire a first spectrum projected from the first beam splitter and a second spectrum projected from the second beam splitter, the imaging unit comprising a plurality of light receiving elements arranged at least in a first direction; as well as The acquisition unit is configured to acquire information about the test object based on the output signal from the imaging unit. Wherein, the direction of increase of the wavenumber of the first spectrum relative to the address of the optical receiving element in the first direction is opposite to the direction of increase of the wavenumber of the second spectrum relative to the address of the optical receiving element in the first direction, and Wherein, the first spectrum and the second spectrum are projected such that, in the first direction, the low wavenumber ends of the first spectrum and the low wavenumber ends of the second spectrum are closer to each other than the high wavenumber ends of the first spectrum and the high wavenumber ends of the second spectrum; or, the first spectrum and the second spectrum are projected such that, in the first direction, the high wavenumber ends of the first spectrum and the high wavenumber ends of the second spectrum are closer to each other than the low wavenumber ends of the first spectrum and the low wavenumber ends of the second spectrum.

19. The identification device according to claim 18, further comprising: Multiple irradiation units are configured to irradiate the test object in association with the first collection unit and the second collection unit.

20. The identification device according to claim 19, wherein, The multiple illumination units are optically coupled to multiple light sources.

21. The identification device according to claim 18, wherein, The first collection unit and the second collection unit collect Raman scattered light from the test object.

Images