JP4589903B2 - Component analyzer - Google Patents

Description

  The present invention relates to a technique for analyzing residual agricultural chemicals attached to the surface of a subject. More specifically, the present invention relates to a technique for analyzing the components of agricultural chemicals in a component analyzer that analyzes the components of a subject such as the concentration of residual agricultural chemicals using an optical analysis technique.

  Conventionally, with regard to residual agricultural chemicals attached to fruits and vegetables, in order to determine whether or not the concentration of residual agricultural chemicals conforms to the residual agricultural chemical standard values stipulated by the Food Sanitation Law, simple analysis using contract analysis or ELISA method It was inspected by destructive inspection by analysis. In addition, as a method for measuring the concentration of residual agricultural chemicals attached to the sample surface without destroying the sample, the sample is irradiated with infrared light, and the diffuse reflected light is used by FT-IR (Fourier transform infrared spectrophotometer). The technique of analyzing is known (for example, refer to Patent Document 1).

In addition, as a method for measuring residual agricultural chemicals in nondestructive inspection, the type of agricultural chemical is determined by comparing the spectral distribution of the reflected light and the peak distribution for each type of agricultural chemical, and the spectrum. A technique for obtaining a residual pesticide value by a maximum peak value in the distribution is known (for example, Patent Document 2). In such a pesticide residue measurement method, infrared reflected light is subjected to spectrum analysis, and the detection data detected by the spectrum analysis is applied to search software provided in the residue pesticide measurement device, and is registered in advance in the search software. The types of pesticides were identified by correlation with several types of data.
JP 2004-325135 A JP 2005-177627 A

However, in the case of destructive inspection, there are problems such as high commission costs and time taken for sample preparation, and the fact that all inspections are impossible due to destructive inspection. . In addition, in the method for measuring residual pesticides disclosed in Patent Document 2, pesticide type identification is based on a single search. If two or more pesticide spectral data are very similar, search for them once. It is difficult to discriminate by, and there is a possibility that it does not meet the new residue standard system that needs to judge many kinds of pesticides. Therefore, an object of the present invention is to provide is to provide a particle analyzer that enables more of many agricultural chemicals, a more accurate analysis.

  The problem to be solved by the present invention is as described above. Next, means for solving the problem will be described.

In Claim 1, the search spectrum (9) which memorize | stores one or several pesticide spectrum regarding the spectrum of a pesticide adhesion subject (6) or a pesticide single-piece | unit, the measurement spectrum converted from the reflected light of infrared rays, and the said search by calculating the correlation between the pesticide spectrum of the library (9), the component analyzer for measuring pesticide residues adhering to the surface of the analyte (6) (1), a plurality of pesticides spectrum similar to the set spectrum, A component analysis apparatus for determining a correlation between a single agrochemical spectrum included in a set spectrum having a set correlation degree or higher and a measurement spectrum after determining a correlation between the set spectrum and the measurement spectrum, wherein the measurement spectrum and search Before the determination of the degree of correlation with the pesticide spectrum in the library, the peak of the measured spectrum is detected and the peak is detected in advance. It is configured to determine the presence or absence of residual pesticides depending on whether or not the noise level setting value (n0) is equal to or higher than the set noise level. Thus, the pesticide having a high correlation between the measurement spectrum and the pesticide spectrum is determined as a residual pesticide, and the residual pesticide discrimination calculation processing is performed .

According to claim 2, in the component analyzer according to claim 1, the arithmetic means (10) constituting the component analyzer performs a Fourier transform on the optical information sent from the AD converter (33). A measurement spectrum expressed from the wave number and intensity is created (S1), an abnormal spectrum is called from the search library (9) (S2), and a correlation between the measurement spectrum and the abnormal spectrum is calculated (S3). It is calculated whether or not it is equal to or more than the predetermined abnormality reference value (S4), and when a plurality of abnormal spectra are stored in the search library (9), those are sequentially retrieved from the search library (9). The abnormal spectrum is called, the correlation between the measured spectrum and each abnormal spectrum is calculated, and if the maximum value of the correlation is equal to or greater than the abnormal reference value, it is determined that the analysis has failed (S4: ES), the display unit (30) displays a message for prompting reanalysis (S11), finishes the component analysis, next determines the presence or absence of residual agricultural chemicals, detects the peak of the measured spectrum (S5), The noise level setting value (n0) stored in the search library (9) is called (S6), and it is determined whether or not the peak or valley difference from the beginning of the peak is greater than or equal to a preset noise level setting value (n0). If the peak is less than or equal to the noise level setting value (n0), it is determined that there is no residual pesticide (S7: NO), the subject is determined to be shippable (6a), and the determination result is It is displayed on the display unit (30) (S12), and the discrimination process is terminated. If the peak is equal to or higher than the noise level set value, it is discriminated that there is a residual pesticide (S7: YES), and the measurement proceeds to the next stage. Between spectra and pesticide spectra High pesticide is determined that pesticide residues, performs arithmetic processing of the pesticide residue determination (S8), and displays the determination result on the display unit (30) (S9), is to end the component analysis.

  As effects of the present invention, the following effects can be obtained.

A measurement library converted from infrared reflected light and the search, comprising a search library (9) for storing one or a plurality of pesticide spectra relating to the spectrum of a pesticide-attached specimen (6) or a single pesticide. by calculating the correlation between the pesticide spectrum of the library (9), the component analyzer for measuring pesticide residues adhering to the surface of the analyte (6) (1), a plurality of pesticides spectrum similar to the set spectrum, A component analysis apparatus for determining a correlation between a single agrochemical spectrum included in a set spectrum having a set correlation degree or higher and a measurement spectrum after determining a correlation between the set spectrum and the measurement spectrum, wherein the measurement spectrum and search Prior to the determination of the degree of correlation with the pesticide spectrum in the library, the peak of the measured spectrum is detected and the peak is preset. It is configured to determine the presence or absence of residual agricultural chemicals based on whether or not the noise level setting value (n0) or higher. If the peak is equal to or higher than the noise level setting value (n0), it is determined that there is residual agricultural chemicals, and in the next stage Since the pesticide with a high correlation between the measured spectrum and the pesticide spectrum is determined as the residual pesticide and the residual pesticide discrimination calculation process is performed , the residual pesticide attached to the surface of the subject without destroying or damaging the subject Can be analyzed immediately.
It is possible to prevent the risk that a specimen exceeding the pesticide residue reference value cannot be shipped, and to add value to the specimen or grain to be shipped.
Moreover, even if a calibration curve is not created for each pesticide as in the past, if a library storing one or a plurality of pesticide spectra and a set spectrum in which the pesticide spectra whose spectrum waveforms are very similar to each other is created, is created. It is possible to quickly analyze the components of various pesticides. And it becomes possible to analyze more agrochemical components more accurately than the conventional one by repeating stepwise specific work.

As in claim 2, in the component analyzer according to claim 1, the arithmetic means (10) constituting the component analyzer performs a Fourier transform on the optical information sent from the AD converter (33). A measurement spectrum expressed from the wave number and intensity is created (S1), an abnormal spectrum is called from the search library (9) (S2), and a correlation between the measurement spectrum and the abnormal spectrum is calculated (S3). It is calculated whether or not it is equal to or more than the predetermined abnormality reference value (S4), and when a plurality of abnormal spectra are stored in the search library (9), those are sequentially retrieved from the search library (9). The abnormal spectrum is called, and the correlation between the measured spectrum and each abnormal spectrum is calculated. If the maximum value of the correlation is equal to or greater than the abnormal reference value, it is determined that the analysis has failed (S4: YE ), A message indicating that reanalysis is urged is displayed on the display unit (30) (S11), the component analysis is terminated, and then the presence or absence of residual pesticides is determined, and the peak of the measured spectrum is detected (S5). Call the noise level setting value (n0) stored in the library (9) (S6), and determine whether the peak or valley difference from the beginning of the peak is greater than or equal to the preset noise level setting value (n0) (S7) If the peak is less than or equal to the noise level setting value (n0), it is determined that there is no residual pesticide (S7: NO), it is determined that the subject can be shipped (6a), and the determination result is displayed. (S12), and the discrimination process is completed. If the peak is equal to or higher than the noise level setting value, it is discriminated that there is a residual pesticide (S7: YES), and the measurement spectrum moves to the next stage. High correlation between chemicals and pesticide spectra Pesticide was determined as residual pesticides, performs arithmetic processing of the pesticide residue determination (S8), and displays the determination result on the display unit (30) (S9), so it ends the component analysis, processing for subject one The time is shortened, and the component analysis can be performed on the residual pesticides of more subjects.

  Next, embodiments of the invention will be described. 1 is a perspective view of the component analyzer 1, FIG. 2 is a side sectional view of the component analyzer 1, FIG. 3 is a diagram showing a waveform of a measurement spectrum, and FIG. 4 is a flowchart relating to component analysis.

  The subject 6 measured by the component analyzer 1 according to the present invention is a general term for strawberries, tangerines, oranges, melons, tomatoes and other fruits and vegetables or materials. For example, when the subject 6 is fruits and vegetables, the fruits and vegetables can be selected according to the type, amount, concentration, etc. of residual agricultural chemicals (hereinafter referred to as residual agricultural chemical concentration). More specifically, as shown in FIG. 1, the subject 6 is taken out from the container 40 or the like, the measurement spectrum on the surface of the subject 6 is measured by the component analyzer 1, and stored in the search library 9 described later. Based on the correlation between the spectrum and the measured spectrum, it is discriminated into a subject 6a that can be shipped (the fruits and vegetables whose correlation is less than the shipping standard value) and a waste subject 6b (the fruits and vegetables whose correlation is greater than or equal to the shipping standard value). Is.

  As shown in FIGS. 1 and 2, the determination as to whether or not the subject 6 can be shipped by the component analyzer 1 is carried out by placing the subject 6 on the stage 3 of the component analyzer 1 and the surface of the subject 6. Is irradiated with infrared rays to convert the optical information of the reflected light into a measurement spectrum, and the rate at which the pesticide spectrum stored in advance in the search library 9 matches the waveform of the measurement spectrum (hereinafter referred to as correlation). Is calculated, and the correlation is compared with a shipping standard value described later.

  As shown in FIG. 2, the component analysis apparatus 1 is emitted from a housing 23 having a stage 3 provided on the upper surface, a light projecting unit 7 disposed inside the housing 23, and the light projecting unit 7. An interferometer 14 that divides and synthesizes (interferes) infrared rays and projects the reflected light to the reflection means 12 for irradiation, reflection means 12 and 13 for irradiation that guides the combined infrared rays to the irradiation mirror 11 below, The infrared ray guided by the irradiation reflecting means 12 and 13 is reflected and irradiated to the side of the subject 6 on the mounting table 5 (arrow B), and reflected by the subject 6 (arrow C). The light receiving mirror 17 that reflects the infrared rays to the light receiving reflecting means 18 and 19 and the light path changing means 15, the light path changing means 15 that projects the infrared light reflected by the light receiving mirror 17 to the detection unit 32, and the optical path The infrared light from the changing means 15 is detected, and the detected infrared optical information Connected to the AD converter 33 as an output signal, the AD converter 33 for transmitting the output signal to the arithmetic means 10 after converting the output signal to a digital signal, the arithmetic means 10, and the arithmetic means 10 And the search library 9 and the like. On the upper surface of the housing 23, a switch 21 for activating the light projecting unit 7, the detecting unit 32, and the like, and a measurement spectrum converted by the calculating means 10 described later and a calculated discrimination result are displayed. A display unit 30 is disposed.

  In the component analyzer 1 of the present invention, when the subject 6 is placed on the stage 3 and the switch 21 is pressed, the light projecting unit 7 transmits infrared rays to the subject 6 via the interferometer 14 and the irradiation mirror 11. The reflected light reflected by the subject 6 is absorbed by the light receiving unit 8 via the light receiving mirror 17 and the optical path changing means 15. The infrared light absorbed by the light receiving unit 8 is detected by the detection unit 32, converted into a digital signal by the AD converter 33, and the converted optical information is transmitted to the calculation means 10. Then, the control means 16 to be described later performs spectrum analysis to determine the presence or absence of residual agricultural chemicals attached to the surface of the subject 6, and the subject to which the subject 6 can be shipped (the subject 6a that can be shipped). It is determined whether the subject is a subject that cannot be shipped (discarded subject 6b), and the determination result is displayed on the display unit 30.

  Since the component analysis apparatus 1 of the present invention measures residual agricultural chemicals adhering to the surface of the subject 6 using infrared rays, the position of the surface of the subject 6 to be measured (the lower surface of the subject 6) is the optimum position. Unless adjusted to (hereinafter referred to as the focal position), an accurate infrared spectrum of the surface of the subject 6 cannot be obtained. When the position of the surface of the subject 6 is not optimal, the intensity of the reflected light received by the light receiving unit 8 becomes weak, so that the difference in intensity according to the wave number (spectrum unevenness) cannot be clearly identified in the spectrum waveform. As a result, the presence / absence of residual agricultural chemicals and the determination of the subject 6 cannot be performed accurately.

  Therefore, in the component analyzer 1, as shown in FIGS. 1 and 2, the stage 3 is provided substantially horizontally on the upper surface of the housing 23 that is the outer wall of the component analyzer 1, and the center of the stage 3 in the plan view is substantially A hole portion 3a is formed in the hole. In the state where the subject 6 is placed so that the operator closes the hole 3a, the switch 21 is pressed to irradiate the lower surface of the subject 6 with infrared rays. With this configuration, the distance from the hole 3a to the mirrors 11 and 17 is always constant. Therefore, if the hole 3a is substantially coincident with the focal position of the infrared rays in advance, the subject 6 is placed in the hole. The surface of the subject 6 substantially coincides with the focal position only by placing the lens 3a so as to close it.

  Specifically, the angle of the mirrors 12, 13,... Is adjusted in advance to a position where the spectral unevenness obtained from the infrared rays reflected from the surface of the subject 6 placed in the hole 3 a is clear. Thus, the position of the surface of the subject 6 is substantially matched with the infrared focal position.

  As shown in FIG. 2, the control unit 16 controls the angle of the light projecting unit 7 and the mirrors 11, 17..., And transmits an operation command (signal) to the display unit 30. The optical information is acquired from the search library 9 in which various types of pesticide-attached specimens or pesticide spectra of a single pesticide are stored, and the AD converter 33 from the detection unit 32, and converted into measurement spectra, The calculation unit 10 is configured to determine the subject 6 by calculating the correlation between the measured spectrum and the agrochemical spectrum stored in the search library 9 described later.

<Search library 9>
The search library 9 is a database such as a magnetic disk connected to the computing means 10 or the like, and stores one or a plurality of types of agrochemical spectra and shipping standard values relating to a pesticide-attached specimen or agrochemical alone created in advance by experiments or the like. ing. Furthermore, it is preferable to classify the agricultural chemicals having similar spectrum waveforms in the search library 9 as an aggregate spectrum. Here, the above-mentioned shipping standard value is a value calculated from a standard value defined by the Food Sanitation Law or the like, and is set as a shipping standard value before measurement. The search library 9 also stores abnormal spectra and abnormal reference values that are obtained by reflection on the fruit and fruit (subject 6) stickers, overripe spots, defective parts exposed inside, and the like. Yes. Further, since there is a high possibility that noise is included in the measurement result, it is necessary to set the noise level when detecting the peak of the measurement spectrum as one of means for taking this noise into consideration. The search library 9 also stores this noise level setting value. The search library 9 may store a program for controlling the operation of the equipment in the component analyzer 1.

  The search library 9 may update the pesticide spectrum by using an external storage medium or the like, store a new pesticide spectrum, or update the noise level setting value, shipping standard value, or the like. Is preferred. Furthermore, it is possible to input the pesticide sprayed inside and outside the field into the search library 9 in advance as a pesticide that may remain. Further, the search library 9 itself is not limited to the one provided in the housing 23 as in the present embodiment, but is a portable storage medium in which the agrochemical spectrum and the abnormal spectrum are stored in an updatable manner. There may be. Then, when a new pesticide-attached specimen or a pesticide spectrum of a single pesticide is created, the storage contents of the storage medium are updated, and the latest pesticide spectrum and abnormalities are measured when measuring and determining the residual pesticide attached to the specimen 6. The spectrum may be stored in a memory or the like of the calculation unit 10.

  Here, the component analysis by the calculating means 10 is demonstrated using the flowchart figure of FIG. In the examples, the presence or absence of residual pesticides is detected step by step. The computing means 10 creates a measurement spectrum represented by wave number and intensity by Fourier transforming the optical information sent from the AD converter 33 (S1), and calls the abnormal spectrum from the search library 9 ( S2), the correlation between the measured spectrum and the abnormal spectrum is calculated (S3), and it is calculated whether the correlation is equal to or greater than the predetermined abnormal reference value (S4). When a plurality of abnormal spectra are stored in the search library 9, those abnormal spectra are sequentially called from the search library 9, the correlation between the measured spectrum and each abnormal spectrum is calculated, and the maximum value of the correlation is obtained. If it is not less than the abnormal reference value, it is determined that the analysis has failed (S4: YES), the display unit 30 is prompted to reanalyze (S11), and the component analysis is terminated.

<Residue presence / absence and pesticide residue determination>
In the next stage, the presence or absence of residual pesticides is determined. A peak (maximum value) of the measured spectrum is detected (S5), a noise level setting value is called from the search library 9 (S6), and a noise level setting in which the difference between the peak and the top (or valley) is preset. It is determined whether or not the value is greater than or equal to the value (FIG. 3n0) (S7). That is, the noise level setting value n0 is stored in the search library 9, and the noise level setting value is called from the search library 9, and the peak level of the measured spectrum is compared with the noise level setting value from the beginning of the peak. If the peak is less than or equal to the noise level set value, it is determined that there is no residual agricultural chemical (S7: NO), the subject 6a is determined to be shippable, and the determination result is displayed on the display unit 30 (S12). The discrimination process is terminated. If the peak is equal to or higher than the noise level setting value, it is determined that there is residual agricultural chemical (S7: YES), and the next stage is determined to determine the agricultural chemical having a high correlation between the measured spectrum and the agricultural chemical spectrum as the residual agricultural chemical. The discrimination calculation process is performed (S8), the discrimination result is displayed on the display unit 30 (S9), and the component analysis is terminated.

<Determination subroutine>
In the calculation process for determining the residual pesticide, in the first stage, the correlation with a registered spectrum having a high degree of correlation with each other, that is, a collective spectrum or a single pesticide spectrum is determined. That is, a single pesticide spectrum having a characteristic waveform for each wave number (to be described later) and a collective spectrum having a similar waveform are called from the search library 9 (S81), and a search parameter for correlation calculation is called. (S82) The correlation between the measured spectrum, the pesticide spectrum and the aggregate spectrum is calculated (S83), and the pesticide spectrum having a high degree of correlation is specified (S84: YES). Further, it is determined whether or not the pesticide spectrum is a collective spectrum (S85). If it is not a collective spectrum (S85: YES), it can be determined that the pesticide is a specific pesticide, and the correlation is equal to or higher than the predetermined shipping standard value. It is calculated whether or not there is (S86), and if the correlation is less than the shipping standard value (S86: YES), it is determined whether or not all the residual pesticides have been specified, that is, whether or not all wave numbers have been determined ( S87) If the identification is completed (S87: YES), the determination process is terminated and the process proceeds to S9. On the other hand, when the degree of correlation is low (S84: NO), the process proceeds to the next wave number, and the same processing is repeated until all the agricultural chemical spectra of the residual agricultural chemical are specified. If it is determined in S85 that the spectrum is an aggregate spectrum, the process proceeds to a step of determining even a minute waveform difference. That is, a search parameter that specializes the difference in the agricultural chemical spectrum included in the aggregate spectrum is called (S88), and the process returns to the correlation calculation (S83) so that it can be individually discriminated and the process is repeated. The search parameters include, for example, selection of a target wave number region, waveform data calculation processing (K / M conversion, differentiation processing, etc.), search algorithms, etc. Parameter conditions that can be accurately searched with each pesticide are experiments, etc. And is stored in the search library 9 in advance.

  When calculating the correlation with the measured spectrum for two or more kinds of pesticide spectra stored in the search library 9 for the aforementioned collective spectrum, the calculated spectrum waveform may be very similar depending on the called search parameter. is there. Under the search parameter, the same pesticide spectrum is discriminated, and there is a possibility that the purpose of discriminating all pesticide species determined as residual pesticides may not be achieved. Therefore, parameter conditions that bring about a spectrum waveform that closely resembles such a pesticide spectrum by experiments or the like are classified as specific pesticide species in advance, and a set of pesticide spectra of the corresponding pesticide species is stored in the search library 9 as a set spectrum. . Furthermore, parameter conditions specialized for the difference in the agricultural chemical spectrum that can individually identify the aggregate spectrum are obtained in advance by experiments or the like, and these are stored in the search library 9. In this way, it is possible to individually discriminate pesticide species stored in advance in the search library 9. In addition, it is possible to provide a component analyzer that enables more accurate analysis of more agrochemical components, which is also a problem to be solved by the present invention.

In other words, component analyzer of the present invention is measured in advance one or more pesticides spectra for pesticide attached analyte or pesticides alone is stored in the search library 9, the measured spectrum which is converted from the infrared reflected light Search parameters (calculation method, target wavenumber region, search algorithm, etc.) are set appropriately between the pesticide spectrum and the aggregate spectrum, and correlation calculation is performed to individually determine the pesticide spectrum and aggregate spectrum included in the measurement spectrum. It is determined and whether or not the subject 6 can be shipped is determined based on whether or not the correlation between the agricultural chemical spectrum and the measured spectrum is equal to or greater than a preset shipping reference value.

  Specifically, the calculation method of the correlation is such that the calculation means 10 has an inter-wavenumber X · Y (or wavelength) in which a characteristic waveform of the pesticide spectrum stored in the search library 9 as shown in FIG. 3 appears. In this case, the correlation coefficient between the pesticide spectrum and the measurement spectrum is calculated, and whether the largest correlation coefficient is greater than or equal to the shipping standard value is calculated. When a plurality of pesticide spectra are stored in the search library 9, the largest correlation coefficient with the measured spectrum is obtained for each pesticide spectrum stored in the search library 9 and obtained for each pesticide spectrum. A maximum value is selected from the largest correlation coefficients, and it is determined whether or not the maximum value is equal to or greater than a shipment reference value. If comprised in this way, the measurement and discrimination | determination of a residual pesticide with respect to the test object 6 to which the multiple types of residual pesticide adheres can be performed more correctly. Specifically, even when a plurality of types of residual pesticides are attached to the subject 6, since the correlation coefficient with each pesticide spectrum is measured high, the shipping standard value is set high so that more accurate discrimination is possible. Can be done.

  In addition, the calculation means 10 is configured to previously synthesize a plurality of types of agrochemical spectra to create a plurality of synthesized spectra, and to calculate the correlation between each synthesized spectrum and the measured spectrum, as described above. Also good. In this case, the correlation is obtained without dividing the measured spectrum or the synthesized spectrum.

  Here, a computer such as a personal computer disposed outside the housing 23 may be used as the control means 16 including the search library 9 and the computing means 10. It is not limited to what is done.

  Finally, a method for measuring and discriminating residual agricultural chemicals by the component analyzer 1 of the present invention will be summarized. When the subject 6 is placed on the stage 3 of the component analysis apparatus 1 and the switch 21 is pushed by the operator, the calculation means 10 projects infrared rays from the light projecting unit 7 to start measurement of residual agricultural chemicals. However, the switch 21 may project infrared light directly from the light projecting unit 7 without using the calculation means 10. The infrared rays are reflected by the surface of the subject 6, and the calculation means 10 converts the output signal relating to the reflected light transmitted from the AD converter 33 into a measurement spectrum. Specifically, the optical information of the reflected light detected by the detection unit 32 is transmitted as a digital signal to the calculation means 10 via the AD converter 33, and the calculation means 10 generates a waveform as a time function of the reflected light. Fourier transform the measured spectrum as a function of frequency.

  The result of the Fourier transform is displayed on the display unit 30 as a spectrum. The calculating means 10 calls the abnormal spectrum sequentially from the search library 9, calculates the correlation between the abnormal spectrum and the measured spectrum, and determines whether or not the correlation is equal to or greater than a predetermined abnormality reference value. If the correlation is greater than or equal to the abnormal reference value, the operator re-places the subject 6 on the stage 3 or automatically travels around the transfer line or the like and places it again on the stage 3 for retrieval. The abnormal spectrum is sequentially called from the library 9, the correlation between the abnormal spectrum and the measured spectrum is calculated, and it is determined whether or not the correlation is equal to or greater than a predetermined abnormality reference value.

  When the correlation is less than the abnormal reference value, the noise level setting value is called from the search library 9 and compared with the peak of the measurement spectrum. If the peak is less than the noise level setting value, the subject 6 can be shipped. On the other hand, if the peak is equal to or higher than the noise level setting value, the pesticide spectrum and the search parameter are sequentially called from the search library 9 to determine the pesticide spectrum or the aggregate spectrum included in the measurement spectrum. Here, when the determined spectrum is determined to be a collective spectrum, a search parameter that can individually determine the collective spectrum is called, and these correlations are calculated to identify individual pesticide spectra. Then, the correlation between the pesticide spectrum and the measurement spectrum is calculated, and it is determined whether or not the correlation is equal to or greater than a predetermined shipping standard value. This discrimination process is performed for all residual agricultural chemicals. When the correlation is less than the shipping reference value, the worker ships the subject 6 as the shipping subject 6a, and when the correlation is equal to or higher than the shipping reference value, the operator removes the subject 6 to be discarded. Treat or wash as specimen 6b.

  A search library 9 for storing one or a plurality of pesticide spectra relating to the spectrum of a pesticide-attached specimen or a single pesticide, and calculating the correlation between the measured spectrum converted from the reflected infrared light and the pesticide spectrum of the search library, A component analyzer 1 for measuring residual pesticides attached to the surface of a specimen, wherein a plurality of similar pesticide spectra are set as a set spectrum, and after determining the correlation between the set spectrum and the measured spectrum, a degree of correlation greater than or equal to a set correlation degree Since the correlation between the measured spectrum and the single pesticide spectrum is determined, the residual pesticide adhering to the subject surface can be immediately analyzed without destroying or damaging the subject. It is possible to prevent the risk that a specimen exceeding the pesticide residue reference value cannot be shipped, and to add value to the specimen or grain to be shipped. Also, without creating a calibration curve for each pesticide as in the past, a library that stores one or more pesticide spectra and a set spectrum of pesticide spectra whose spectrum waveforms are very similar to each other is created. It is possible to quickly analyze the components of various pesticides. And it becomes possible to analyze more agrochemical components more accurately than the conventional one by repeating stepwise specific work.

  Before the determination of the degree of correlation between the measured spectrum and the agrochemical spectrum of the search library 9, the presence or absence of residual agrochemical is determined based on whether the measured spectrum is equal to or higher than a preset noise level setting value. The processing time is shortened, and the component analysis can be performed on the residual pesticides of more specimens.

The perspective view of the component analyzer 1. FIG. FIG. 3 is a side sectional view of the component analyzer 1. The figure which shows the waveform of a measurement spectrum. The flowchart figure which concerns on component analysis.

DESCRIPTION OF SYMBOLS 1 Component analyzer 3 Stage 6 Subject 7 Light projection part 9 Search library 10 Calculation means 30 Display part

Claims (2)

A search library (9) for storing one or a plurality of pesticide spectra relating to the spectrum of a pesticide-attached specimen (6) or a single pesticide, a measurement spectrum converted from infrared reflected light, and the pesticides in the search library (9) by calculating the correlation between the spectrum and the component analyzer for measuring pesticide residues adhering to the surface of the analyte (6) (1), a plurality of pesticides spectrum similar to the set spectrum, the measurement with the set spectrum A component analysis apparatus for determining a correlation between a single pesticide spectrum included in a set spectrum having a set correlation degree or more and a measured spectrum after determining a correlation with the spectrum, and comprising the measurement spectrum and the pesticide spectrum in the search library The peak of the measured spectrum is detected before the correlation level is determined, and the noise level is set in advance. It is configured to determine the presence or absence of residual pesticide depending on whether it is equal to or greater than a fixed value (n0). If the peak is equal to or higher than the noise level setting value (n0), it is determined that there is residual pesticide, and in the next stage, A component analysis apparatus characterized in that a pesticide having a high correlation in pesticide spectrum is determined as a residual pesticide, and the residual pesticide discrimination processing is performed. The component analysis apparatus according to claim 1, wherein the arithmetic means (10) constituting the component analysis apparatus is represented by wave number and intensity by performing Fourier transform on the optical information sent from the AD converter (33). (S1), the abnormal spectrum is called from the search library (9) (S2), the correlation between the measured spectrum and the abnormal spectrum is calculated (S3), and the correlation is determined in advance. It is calculated whether or not it is a reference value or more (S4), and when a plurality of abnormal spectra are stored in the search library (9), the abnormal spectra are sequentially called from the search library (9), The correlation between the measured spectrum and each abnormal spectrum is calculated, and if the maximum value of the correlation is greater than or equal to the abnormal reference value, it is determined that the analysis has failed (S4: YES), and the display unit (3 ) Prompts re-analysis (S11), ends the component analysis, then determines whether there is residual pesticide, detects the peak of the measured spectrum (S5), and stores it in the search library (9) The set noise level setting value (n0) is called (S6), and it is determined whether or not the peak or valley difference from the beginning of the peak is greater than or equal to a preset noise level setting value (n0) (S7). If the peak is less than or equal to the noise level setting value (n0), it is determined that there is no pesticide residue (S7: NO), it is determined that the subject can be shipped (6a), and the determination result is displayed on the display unit (30). (S12), the discrimination process is terminated, and if the peak is equal to or higher than the noise level set value, it is discriminated that there is a residual pesticide (S7: YES), and the next step is taken to correlate the measured spectrum with the pesticide spectrum. High pesticide residue Judgment, performs arithmetic processing of the pesticide residue determination (S8), the determination result displayed on the display unit (30) (S9), component analyzer, characterized by terminating the component analysis.

Images