JP2015014527A - Abnormality detection device and abnormality detection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To allow abnormality detection even in the case of an inspection object being a material with a high water content.SOLUTION: An abnormality detection device is configured to include an irradiation area A1 irradiated with infrared rays to acquire an image of an inspection object 1, in a first flow channel 11 inclined relative to a horizontal plane and to acquire a spectrum image in a state where the inspection object 1 is moving on the surface of the inclined flow channel, whereby a thickness of the inspection object 1 can be adjusted, spectrum data relating to the inspection object can be obtained with a reduced influence of absorption due to water, and abnormality can be accurately detected. In addition, the thickness of the inspection object 1 can be easily adjusted because the angle of inclination of the first flow channel 11 can be changed in accordance with viscosity of the inspection object. Further, the abnormality detection device is configured to include a diffuse reflection plate 50 provided on the rear side of the first flow channel 11, whereby the same reflectance as the case that the surface of the first flow channel 11 is made into a diffuse reflection plate can be obtained and contamination of the diffuse reflection plate can be prevented.

Description

  The present invention relates to an abnormality detection device that detects an abnormality mixed in an inspection object and an abnormality detection method using the abnormality detection device.

  2. Description of the Related Art Conventionally, there has been known an apparatus for imaging an inspection object and analyzing the image data to determine whether there is an abnormality after detecting the foreign object or defective product mixed in the inspection object flowing on the processing line. Yes. For example, in Patent Literature 1, UV light, visible light, or infrared light is applied to an inspection object that moves on a conveyor, and an image is captured by a camera disposed above the conveyor, and a foreign object is detected by image processing. The method is shown. Here, in the case of performing a spectrum analysis of a foreign object or a defective product obtained by imaging an inspection object by the method of Patent Document 1, light in the near-infrared wavelength region in which a specific absorption spectrum for each substance is likely to appear. The measurement is preferably performed using various methods, and various methods have been studied.

JP 2006-133052 A

  However, when the object to be inspected is a substance containing a large amount of water, absorption by water increases in the near-infrared wavelength region, particularly in the wavelength region of 1350 nm or more, and it is difficult to confirm peaks derived from other materials. . Therefore, it is difficult to detect foreign matters or defective products in the measurement of the wavelength region of near infrared light.

  The present invention has been made in view of the above, and an abnormality detection device capable of detecting an abnormality even when a substance containing a large amount of water is a test object, and an abnormality using the abnormality detection device. An object is to provide a detection method.

  In order to achieve the above object, the abnormality detection device according to the present invention is inclined with respect to a horizontal plane, moves a test object containing moisture and fluidity on the surface, and diffuses on the back surface opposite to the surface. A flow path to which a reflector is attached, a light source unit that emits near-infrared light to an inspection object that moves on an irradiation area of the flow path, and an inspection target that is irradiated with near-infrared light from the light source unit An imaging unit that spectrally captures light from an object and outputs spectral data; and an analysis unit that detects an abnormality based on the spectral data output by the imaging unit. It is made of a material having optical properties, and is characterized in that the reflection surface of the diffuse reflection plate is attached to the irradiation region so as to face the back surface of the flow path.

  Further, the abnormality detection method according to the present invention is such that a test object that is inclined with respect to a horizontal plane, contains moisture and has fluidity is moved on the surface, and a diffuse reflector is attached to the back surface opposite to the surface. An abnormality detection method using an abnormality detection device attached to the irradiation region so that the irradiation region of the flow channel is made of a light-transmitting material and the reflection surface of the diffuse reflector is opposed to the back surface of the flow channel, A step of emitting near-infrared light to the inspection object moving on the irradiation area of the flow path from the section, and spectrally imaging the light from the inspection object irradiated with the near-infrared light from the light source section And outputting the spectrum data, and detecting an abnormality based on the spectrum data output in the step of outputting the spectrum data.

  According to the above-described abnormality detection device and abnormality detection method, the irradiation region for irradiating near infrared light from the light source unit is provided in the channel inclined with respect to the horizontal plane, and the inspection object moves on the surface of the inclined channel. In this state, the spectrum data is acquired. As a result, the thickness of the inspection object can be adjusted so that the near-infrared light reaches the bottom side (the surface side of the flow path) of the inspection object, and the influence of water absorption is reduced. Since spectrum data concerning the inspection object can be obtained, even if the inspection object contains water and has fluidity, it is possible to accurately detect abnormalities contained therein. Furthermore, with respect to the irradiation region of the flow path, the amount of light incident on the imaging unit can be increased by providing a diffuse reflection plate on the back side as a configuration made of a light-transmitting material. Further, by providing the diffuse reflector on the back side, the diffuse reflector and the inspection object do not come into contact with each other, and the diffuse reflector can be prevented from being stained.

  Here, as a configuration that effectively exhibits the above-described operation, for example, an aspect in which an ND filter is further provided between the inspection target and the imaging unit can be cited.

  With the configuration further including the ND filter as described above, for example, the background data can be acquired using the ND filter and can be corrected using the ND filter, and the amount of diffuse reflected light is greatly absorbed by water. Even when a small number of inspection objects are imaged, small changes included in the spectrum data can be detected, and abnormality detection accuracy is improved.

  Moreover, the aspect which is at least one part of the wavelength range of 1300 nm-2400 nm is mentioned for near infrared light.

  As described above, even when light included in a wavelength range of 1300 nm to 2400 nm that is particularly affected by water absorption is used as near-infrared light, abnormal detection can be suitably performed.

  ADVANTAGE OF THE INVENTION According to this invention, even if it is a case where the substance containing much moisture is a test subject, the abnormality detection apparatus which can detect abnormality, and the detection method using this abnormality detection apparatus are provided.

It is a figure which shows the structure of the abnormality detection apparatus which concerns on this embodiment. It is a figure which shows the measurement result by the abnormality detection apparatus which concerns on this embodiment. It is a figure which shows the measurement result by the abnormality detection apparatus which concerns on this embodiment. It is a figure which shows the measurement result by the abnormality detection apparatus which concerns on this embodiment.

  DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted.

  An abnormality detection apparatus 100 according to an embodiment of the present invention will be described with reference to FIG. The abnormality detection apparatus 100 is an apparatus that inspects for the presence or absence of abnormalities such as foreign matter mixed in the inspection object 1 having fluidity moving on the flow path 10 and alteration of the inspection object 1. Examples of the inspection object 1 of the abnormality detection apparatus 100 according to the present embodiment include raw materials such as foods, products, and the like that contain a lot of moisture and have fluidity. Examples of such an inspection object include fruit compote, honey, jam, and the like, and a configuration in which a liquid and a solid are mixed may be used. In addition, examples of detection targets by the abnormality detection apparatus 100 include foreign substances such as hair and other living body-derived substances, metals and foreign substances derived from manufacturing apparatuses, and the like. Anomalies such as alteration of the inspection object 1 are also detected by the anomaly detection device 100. The abnormality detection apparatus 100 measures the spectrum of diffusely reflected light obtained by irradiating the inspection object 1 with measurement light, and based on the spectrum, foreign matter and denatured material attached to the inspection object 1 (hereinafter, referred to as “diffuse reflected light”). Foreign matter and alteration are collectively referred to as “abnormality”). For this reason, the abnormality detection device 100 includes a light source unit 20, an imaging unit 30, and an analysis unit 40. In addition, a diffuse reflector 50 for making light from the light source unit 20 incident on the imaging unit 30 is attached to the back side of the flow path 10 of the inspection object 1.

  The flow path 10 includes a first flow path 11 that is inclined with respect to a horizontal plane, and a second flow path 12 that has a gentler inclination angle than the first flow path 11 that is continuously provided downstream of the first flow path 11. It is comprised including. By moving the inspection object 1 on the inclined flow path 10, the thickness of the inspection object 1 moving on the flow path 10 is adjusted. For example, when the second flow path 12 is a horizontal plane and the moving speed of the inspection object 1 in the second flow path 12 is 1 m per minute, the inclination of the first flow path 11 is changed and the first flow is changed. It is assumed that the moving speed of the inspection object 1 on the road 11 is 5 m per minute. In this case, while moving the first flow path 11, the thickness of the inspection object 1 can be reduced to 1/5 compared to when the second flow path 12 is moved. Since the relationship between the inclination angle and the moving speed greatly affects the viscosity of the inspection object 1, it is preferable to obtain the corresponding relationship in advance according to the type of the inspection object 1.

  The light source unit 20 irradiates measurement light having a certain wavelength band toward a predetermined irradiation region A1 on the flow path 10. The wavelength range of the measurement light emitted by the light source unit 20 is appropriately selected according to the inspection object 1 and foreign matters attached to the inspection object 1 to be detected and abnormalities such as alteration. When using near infrared light as measurement light, specifically, light having a wavelength range of 800 nm to 2500 nm is preferably used, but visible light can be used as measurement light instead of near infrared light. . In addition, a light source (SC light source) that generates super continuum (SC) light can be used as the light source unit 20.

  The irradiation area A1 is a partial area on the surface of the first flow path 11 in the flow path 10 in which the inspection object 1 moves. This irradiation area A1 extends in the width direction (x-axis direction) perpendicular to the moving direction of the inspection object 1 (y-axis direction in FIG. 1), from one end of the first flow path 11 to the other end. It is the area | region extended in the line shape which covers. The width of the irradiation area A1 in the direction perpendicular to the extending direction of the irradiation area A1 (y-axis direction) is 10 mm or less in this embodiment. (The width of the irradiation region may be 10 mm or more, but is easily affected by heat.)

  Here, the area | region used as irradiation area | region A1 among the 1st flow paths 11 is made into the material which has a light transmittance at least. An example of such a material is quartz glass. Furthermore, a diffuse reflector 50 is provided on the back surface of the first flow path opposite to the surface of the first flow path 11 (the surface on which the inspection object 1 is placed), and the reflection surface of the first flow path is the first flow path. It is attached to face the back side.

  Moreover, the 1st flow path 11 to which the diffuse reflection board 50 was attached can also be set as the structure which downstream area | region 11A to which the diffuse reflection board 50 was attached can rotate below centering on the axis | shaft 11B. In this case, the inspection object 1 on the downstream region 11A can be removed from the flow path 10 by dropping downward. The rotation of the downstream region 11A is preferably performed based on an analysis result by an analysis unit 40 described later. By setting it as such a structure, it becomes possible to remove easily the test target object 1 from which abnormality was detected from the flow path 10. When the downstream region 11A that discharges the inspection object 1 in which an abnormality is detected is provided on the flow path 10, the discharge may be performed in the immediate vicinity of a point where the abnormality is detected (an area in which imaging is performed by the imaging unit 30). desirable. Even if the movement speed of the inspection object 1 on the flow path is constant, the movement speed may be different from that of the inspection object 1 depending on the size or shape of the portion (or substance) that shows the abnormality. is there. When the moving speed is different, if the distance from the region where the imaging unit 30 captures an image is increased, it becomes difficult to specify the position of the portion showing the abnormality in proportion to the distance, and it is difficult to reliably discharge the region. Therefore, it is preferable to remove a portion showing an abnormality in the vicinity of the irradiation region A1, and a configuration including the irradiation region A1 as in the downstream region 11A of the abnormality detection device 100 according to the present embodiment may be employed.

  In FIG. 1, for the sake of explanation, the lower side of the shaft 11B on the downstream side of the first flow path 11 is configured to be rotatable. However, for example, the entire first flow path 11 is configured to be rotatable. It is good. In that case, it is preferable to perform measurement related to abnormality detection on the lower end side of the first flow path 11, but it is also possible to perform measurement related to abnormality detection on the upstream side and to allow the upstream side to be rotatable. In addition, it is considered that a time lag of about 0.01 seconds occurs until the downstream region 11A moves based on an instruction from the analysis unit 40 after the analysis unit 40 detects an abnormality as a result of the spectrum analysis. Therefore, it is necessary to set the position of the downstream area 11A (rotating area) in consideration of the moving speed of the inspection object.

  The light source unit 20 includes a light source that emits light and an irradiation unit that emits light from the light source to the one-dimensional line-shaped irradiation region A1. When the SC light is emitted from the light source unit 20, the light source includes a seed light source and a nonlinear medium. The light emitted from the seed light source is input to the nonlinear medium, and the spectrum is broadened to a wide band by the nonlinear optical effect in the nonlinear medium. Output light. In addition, for example, a cylindrical lens is preferably used as the irradiation unit.

  The near-infrared light L1 output from the light source unit 20 is diffusely reflected by the inspection object 1 placed on the irradiation area A1. A part of the light enters the imaging unit 30 as diffusely reflected light L2. Further, the light that has passed through the inspection object 1 passes through the first flow path 11 from the irradiation area A1, is diffusely reflected by the diffuse reflector 50, and part of the light enters the imaging unit 30 as diffusely reflected light L2.

  The imaging unit 30 has a function as a hyperspectral sensor that acquires a hyperspectral image. A hyperspectral image is an image in which one pixel is composed of N pieces of wavelength data, and includes spectral information including reflection intensity data corresponding to a plurality of wavelengths for each pixel. That is, a hyperspectral image has a three-dimensional configuration that combines two-dimensional elements as an image and elements as spectral data because of the feature that each pixel constituting the image has intensity data of multiple wavelengths. It is data. In the present embodiment, the hyperspectral image refers to an image composed of pixels having intensity data in at least five wavelength bands per pixel.

  The imaging unit 30 includes a camera lens 31, a slit 32, a spectroscope 33, and a light receiving unit 34. The imaging unit 30 has a visual field region 30 </ b> S extending in a direction perpendicular to the moving direction of the inspection object 1 (x-axis direction). The visual field region 30S of the imaging unit 30 is a linear region included in the irradiation region A1 on the surface of the first flow path 11, and the diffuse reflected light L2 that has passed through the slit 32 forms an image on the light receiving unit 23. It is.

  The slit 32 is provided with an opening in a direction parallel to the extending direction (x-axis direction) of the irradiation region A1. The diffuse reflected light L <b> 2 that has entered the slit 32 of the imaging unit 30 enters the spectroscope 33. The spectroscope 33 splits the diffuse reflected light L2 in the direction perpendicular to the extending direction of the slit 32, that is, in the direction perpendicular to the extending direction of the irradiation region A1 (y-axis direction).

  The light receiving unit 34 includes a light receiving surface 23 in which a plurality of light receiving elements are two-dimensionally arranged, and each light receiving element receives light dispersed by the spectroscope 33. Thereby, the light of each wavelength of the diffusely reflected light L2 reflected at each position along the width direction (x-axis direction) on the first flow path 11 by the plurality of light receiving elements is received. Each light receiving element outputs a signal corresponding to the intensity of received light as information on a two-dimensional planar point composed of a position and a wavelength. A signal output from the light receiving element of the light receiving unit 34 is sent from the imaging unit 30 to the analyzing unit 40 as image data related to the hyperspectral image.

  In addition, it is good also as a structure further equipped with the ND filter 38 between the imaging part 30 and the test object 1. FIG. The ND (Neutral Density) filter 38 is a filter that reduces the amount of light entering the lens at a predetermined rate regardless of the wavelength of the light. By using the ND filter, even if the amount of light from the inspection object 1 is insufficient, it can be detected with higher accuracy. This point will be described later.

  The analysis unit 40 obtains the spectrum of the diffuse reflected light L2 from the input signal, and performs an inspection based on the obtained spectrum. When detecting foreign matter contained in the inspection object 1, the inspection is performed according to the following principle. The foreign matter has an absorption band in the wavelength range of the measurement light (near infrared light in the present embodiment) output from the light source unit 20. Therefore, a specific absorption peak derived from a foreign object is detected by referring to spectral information that is included for each pixel and is composed of a plurality of intensity data in a hyperspectral image obtained by diffusely reflected light L2 diffusely reflected by the inspection object 1. Thus, a foreign object is detected. Moreover, when measuring the quantity of the water | moisture content and sugar contained in the test object 1, it test | inspects on the following principles. For example, since sugar has absorption peaks in the vicinity of a wavelength of 1500 nm and a wavelength of 2100 nm, when the inspection object 1 contains a sugar, it is irradiated with near infrared light in the range of at least 100 nm before and after these wavelengths and diffused. By analyzing the spectrum of the reflected light L2, it is possible to detect a peak derived from sugar in the food. The type and content of the sugar can be obtained from the position and intensity of the peak derived from the sugar, and the abnormality of the test object 1 can be detected or the quality can be evaluated. The result of analysis by the analysis unit 40 is notified to the operator of the abnormality detection device 100 by outputting the result to, for example, a monitor connected to the analysis unit 40 or a printer.

  The analysis unit 40 includes a CPU (Central Processing Unit), a RAM (Random Access Memory) and a ROM (Read Only Memory) that are main storage devices, a communication module that performs communication with other devices such as a detection unit, and a hard disk It is comprised as a computer provided with hardware, such as auxiliary storage devices. And the function as the analysis part 40 is exhibited when these components operate | move. An outline of analysis processing by the analysis unit 40 and a specific method thereof will be described later.

  As an abnormality detection method using the above-described abnormality detection device 100, the inspection object 1 is disposed in the first flow path 11 by introducing the inspection object 1 having fluidity to the upstream side of the first flow path 11. It moves from the upstream side to the downstream side along the inclination, and further moves to the second flow path 12. At this time, by irradiating near-infrared light L1 to the irradiation area A1 on the first flow path 11 from which the inspection object 1 moves from the light source unit 20, diffuse reflected light from the inspection object 1 or inspection object Of the transmitted light of the object 1, the light diffusely reflected by the diffuse reflector 50 is incident on the camera lens 31 of the imaging unit 30, and after being split by the slit 32 and the spectroscope 33, is received by a plurality of light receiving elements. As a result, a signal output from the light receiving element of the light receiving unit 34 is sent as image data related to the hyperspectral image from the imaging unit 30 to the analysis unit 40, and the analysis unit 40 determines whether there is an abnormality in the inspection object 1. To detect.

  Here, since the inspection object 1 by the abnormality detection apparatus 100 contains water and has fluidity, when it is irradiated with near infrared light, the influence of water absorption increases, and the inspection object 1 It is conceivable that spectrum data cannot be obtained. Therefore, in order to acquire an image of the inspection object 1, an irradiation area A1 for irradiating near-infrared light is provided in the first flow path 11 inclined with respect to the horizontal plane, and the inspection object 1 is on the surface of the inclined path. The spectrum image is acquired in a moving state. Accordingly, the thickness of the inspection object 1 can be adjusted so that the near infrared light reaches the bottom side of the inspection object 1 (the surface side of the first flow path 11), and the influence of absorption by water. Since the spectrum data relating to the inspection object can be obtained in a state where the size of the inspection object is small, even if the inspection object contains water and has fluidity, the abnormality contained therein can be detected with high accuracy. Moreover, in the abnormality detection apparatus 100, since adjustment of the thickness of the test object 1 can respond by changing the inclination-angle of the 1st flow path 11 according to the viscosity of the test object 1, it adjusts easily. be able to.

  Further, in order to detect an abnormality in the inspection object 1, it is preferable to reduce the influence of water absorption, but since the inspection object 1 contains water, removing the influence is not possible. Since it is difficult, a configuration for increasing the amount of light incident on the imaging unit 30 is desired. Therefore, the abnormality detection device 100 is configured to provide the diffuse reflection plate 50 having a high reflectance in the entire measurement wavelength range on the back surface side of the first flow path 11. Further, in the abnormality detection apparatus 100, an area that connects the diffusive reflector 50 and the imaging unit 30, and is an area that images the inspection object 1, that is, an irradiation area A1 that irradiates near-infrared light from the light source unit 20. Is made of a material having translucency with respect to near-infrared light as measurement light. In order to increase the amount of light incident on the imaging unit 30, for example, it is desirable that the surface of the first flow path 11 serving as a background in the imaging of diffuse reflected light in the imaging unit 30 be a diffuse reflection plate. With this configuration, the reflective surface of the diffusive reflecting plate comes into contact with the object to be inspected, so that it is easy to get dirty and the reflectance is lowered. Accordingly, by providing the diffuse reflection plate 50 on the back surface side as described above, the same reflectance as that obtained when the surface of the first flow path 11 is a diffusion reflection plate can be obtained, and dirt on the diffusion reflection plate can be obtained. Can be prevented. Furthermore, cleaning is easy when the surface of the first flow path 11 is dirty, and measurement accuracy can be easily maintained.

  And said structure is used especially suitably when using the light containing at least one part of a wavelength range of 1300 nm-2400 nm as near-infrared light. That is, when the light included in the wavelength range having a large influence of water absorption is used as measurement light, the abnormality detection apparatus 100 according to the present embodiment can more effectively exhibit the above-described action.

  Moreover, as a structure which detects the abnormality of the test object 1, it measures with the transmission system which radiate | emits near-infrared light from one side of the test object 1, and receives the light which permeate | transmitted the test object 1 on the other side. A method is also conceivable. However, when the transmission system is employed, when the near infrared light does not pass through the inspection object 1, a black object is imaged on the imaging unit side. Here, when the inspection object 1 contains a substance that hardly transmits near infrared light, such as a lump of pulp, for example, the inspection object 1 includes an object that does not transmit near infrared light as a result of measurement. Even if it is not possible to identify whether it is included in a normal product or an abnormality, the accuracy of abnormality detection is not sufficient. On the other hand, in the abnormality detection device 100 according to the present embodiment, the near-infrared light is configured such that the imaging unit 30 receives diffusely reflected light from the inspection object 1 by irradiating near-infrared light. Even if a substance that is difficult to transmit is included, diffuse reflected light can be obtained by the imaging unit 30, and therefore it is possible to identify whether or not it is abnormal by analyzing the spectrum.

Next, a setup for the purpose of improving measurement accuracy in measurement using the abnormality detection apparatus 100 will be described. In order to stably acquire the spectrum data related to the inspection object in the abnormality detection apparatus 100 according to the present embodiment, it is necessary to correct the spectrum data captured for the inspection object 1. As the method, there is a method in which correction is performed by the following equation so that the spectral data photographed in a light-shielded state is converted to 0% reflectance and the spectral data photographed from the diffuse reflector is converted to 100% reflectance.
Data = [Data (real) −Data (dark)] / [Data (white) −Data (dark)]
(If the reflectance is displayed in%, multiply 100 by the above formula)
Data (dark): Received light intensity when shaded
Data (white): Received light intensity when photographing a diffuse reflector
Data (real): Received light intensity when photographing the inspection object

  By performing the correction described above, the reflectance output from each light receiving element can be adjusted so as not to be blown out even when Data (white) is photographed. On the other hand, when the above-described correction is performed, when the inspection object 1 is measured, the reflectance of the inspection object 1 obtained in the imaging unit 30 becomes low due to the influence of absorption of water in the liquid portion. In some cases, the dynamic range of the imaging unit 30 cannot be used effectively. Specifically, for example, there may be a case where the maximum value of the reflectance of the inspection object is obtained only about 10%. As described above, when the maximum value of the reflectance when the inspection object is measured is low, it may be difficult to detect the abnormality.

  Therefore, when acquiring Data (white) for performing the above correction, instead of directly photographing the diffuse reflector, a 10% transmission ND filter is disposed between the diffuse reflector and the diffuse reflector. It is effective to perform a correction so that the output from the light receiving element obtained at that time is equivalent to a reflectance of 100%. More specifically, the amount of light from the light source 20 is such that the output from the light receiving element is 80 to 90% with respect to the full scale of the output by the imaging unit 30 when photographing through an ND filter of 10% transmission. By adjusting the value, 80 to 90% of output can be obtained even when a dark inspection object is imaged, and the S / N ratio of the measurement data can be increased by fully utilizing the function of the imaging unit 30. It is possible to improve the detection rate. In addition, the brightness when the inspection object is imaged (the intensity of light received by each light receiving element in the light receiving unit) depends on the influence of the thickness due to the viscosity of the inspection object and the absorption characteristics of the inspection object. receive. Therefore, the optimal transmittance of the ND filter changes for each inspection object. In the case of using the ND filter, it is preferable to measure the inspection object once without the ND filter and then select an optimal ND filter transmittance from the maximum reflectance.

  Here, an example of the result of evaluation using the abnormality detection apparatus 100 of FIG. 1 is shown. In FIG. 1, the object to be inspected is marmalade, and the part (foreign matter) showing an abnormality is plastic. In FIG. 1, a diffusive reflector having a reflectivity of about 97% over the entire wavelength region is arranged so that the reflecting surface is the upper surface, and quartz corresponding to an irradiation region made of a light-transmitting material is formed thereon. Glass was placed thereon, an inspection object or foreign matter was placed thereon, and near-infrared light having a wavelength range of 1100 nm to 2200 nm was irradiated from the light source unit 20, and the spectrum was acquired in the imaging unit 30. In addition, for the correction of data, the above correction method is used, the output from the light receiving element when the diffuse reflector is measured through the quartz glass without placing the inspection object is set to reflectivity 100%, and the lens is covered. Thus, the output from the light receiving element in a state where the light was shielded was corrected so that the reflectance was 0%.

  FIG. 2 shows the reflectance of a plastic that becomes a foreign substance, the reflectance of a 1.5 mm thick marmalade, and the reflectance when a 1.5 mm thick marmalade is applied on the plastic. According to the result of FIG. 2, an absorption peak peculiar to plastics can be confirmed in the vicinity of a wavelength of 1750 nm, but this absorption peak can be confirmed even when an image of a plastic coated with a 1.5 mm thick marmalade is taken. That is, if the thickness is about 1.5 mm, the absorption spectrum can be detected when plastic is contained in the marmalade, that is, the plastic in the marmalade can be detected as a foreign substance. It is.

  Next, FIG. 3 shows the result of studying the shape change of the reflection spectrum when the thickness of the marmalade is changed. FIG. 3 shows the reflection spectrum when the thickness of the marmalade is 6 mm. FIG. 3 also shows three types: plastic only, marmalade only, and plastic coated marmalade. According to the result of FIG. 3, in the region of wavelength 1400 nm or more, the absorption due to plastic near 1750 nm cannot be detected because the absorption derived from marmalade is too large. This is mainly due to absorption by water in the marmalade, and the same phenomenon appears even when a liquid containing water other than marmalade is measured.

  As described above, it was confirmed that in order to detect a foreign substance from a liquid containing water that has a large absorption with respect to light, it is necessary to measure with a reduced liquid layer thickness. In contrast, the abnormality detection apparatus 100 according to the present embodiment has a configuration in which the thickness of the inspection object is controlled by moving on the inclined first flow path 11, thereby increasing the thickness of the inspection object. Can be reduced to such a thickness that foreign matter can be detected, and as a result, even a liquid containing a large amount of moisture can be detected.

  FIG. 4 shows the result when correction is performed using an ND filter. FIG. 4 shows the reflectance when a 1.5 mm-thick marmalade is used as the inspection object, similarly to the measurement whose result is shown in FIG. Furthermore, in FIG. 4, a reflection spectrum (thick line) after correction using Data (White) using a 50% transmission ND filter is added to a sample in which 1.5 mm thick marmalade is applied to plastic. ing. As a result of adjusting the amount of light with the 50% ND filter used, the amount of light corresponding to 100% is approximately doubled, and the apparent reflectance is approximately twice that of the case where the ND filter is not used. It has become. Therefore, it became easier to see the absorption peak of the plastic measured through the marmalade (indentation near the wavelength of 1750 nm). When the ND filter is used in this way, it is possible to detect foreign matters in the marmalade even when the thickness of the marmalade increases, for example, when the thickness is about 4 to 5 mm.

  As mentioned above, although embodiment of this invention was described, this invention is not limited to the said embodiment, A various change can be made. For example, in the above embodiment, the configuration using the SC light source has been described, but the type of the light source is not particularly limited. Moreover, although the said embodiment demonstrated the structure which acquires a hyperspectral image, it is not limited to a hyperspectral image, If it is an apparatus which can acquire spectral data, it will not specifically limit.

DESCRIPTION OF SYMBOLS 3 ... Test object, 10 ... Flow path, 11 ... 1st flow path, 20 ... Light source part, 30 ... Imaging part, 38 ... ND filter, 40 ... Analysis part, 50 ... Diffuse reflector, 100 ... Abnormality detection apparatus.

Claims (4)

A flow path that is inclined with respect to a horizontal plane, contains moisture and has a fluidity to be moved on the surface thereof, and a diffusion reflector is attached to the back surface opposite to the surface.
A light source unit that emits near-infrared light to the inspection object moving on the irradiation region of the flow path;
An imaging unit that spectrally captures and images light from the inspection object irradiated with the near-infrared light from the light source unit, and outputs spectral data;
An analysis unit for detecting an abnormality based on the spectrum data output by the imaging unit;
With
The flow path is made of a material in which the irradiation region has translucency,
The abnormality detection apparatus attached to the said irradiation area so that the reflective surface of the said diffused reflection plate may oppose the said back surface of the said flow path.   The abnormality detection device according to claim 1, further comprising an ND filter between the inspection object and the imaging unit.   The abnormality detection apparatus according to claim 1, wherein the near-infrared light is light including at least a part of a wavelength range of 1300 nm to 2400 nm. An inspection object that is inclined with respect to a horizontal plane, contains moisture, and has fluidity is moved on the surface, and a diffuse reflector is attached to the back surface opposite to the surface, and the irradiation area of the flow path is translucent. An abnormality detection method by an abnormality detection device attached to the irradiation region so that the reflection surface of the diffuse reflector is opposed to the back surface of the flow path,
Emitting near-infrared light to the inspection object moving on the irradiation area of the flow path from a light source unit;
Spectrally imaging the light from the inspection object irradiated with the near-infrared light from the light source unit, and outputting spectral data;
Detecting abnormalities based on the spectral data output in the step of outputting the spectral data;
An abnormality detection method comprising:

Images