Portable near-infrared rice adulteration detection device and detection method

Abstract

This invention discloses a portable near-infrared rice adulteration detection device and method, comprising: a housing, a light source, a grinding and transmission device, a spectral acquisition device, and a calculation module; the light source provides the necessary light environment for detection; the grinding and transmission device grinds sample rice into rice flour and drops it into a quartz dish, which is then placed on a sample detection stage using a telescopic motor; the spectral acquisition device collects spectral data from different objects; the calculation module adjusts parameters, receives signals emitted by the spectral acquisition device, processes the data, and draws conclusions through data processing and comparison, which are displayed on a touchscreen 1. This invention enables rapid detection of the proportion of adulterated rice, using the fatty acid content of the rice as an indicator of its aging degree.

Description

Technical Field

[0001] This invention relates to the field of rice quality testing technology, specifically to a portable near-infrared rice adulteration detection device and method. Background Technology

[0002] During prolonged storage, the aging process of rice itself leads to quality deterioration, reduced edible value, and even the formation of carcinogens. Selling reprocessed aged rice not only disrupts the market for normal rice, but the toxins produced by various chemical reactions during the aging process also pose a health risk to consumers. Therefore, efficient and rapid analysis to identify adulterated rice is crucial for quality control and ensuring rice safety.

[0003] Studies have shown that physicochemical indicators such as fatty acid value, moisture content and water molecule state, and sulfhydryl content are fundamental factors in evaluating grain quality, and these indicators are of significant reference value in exploring its aging mechanism. Related research indicates that the fatty acid value of rice is a sensitive factor reflecting rice freshness and a sensitive indicator for detecting rice quality and determining its age. It has a good linear relationship with storage time; the fatty acid value of rice increases with prolonged storage time. Therefore, a detection device and method are needed to detect the fatty acid value of rice, thereby providing a new and feasible detection method for rice quality control. Summary of the Invention

[0004] Purpose of the invention: To address the above-mentioned technical problems, this invention provides a portable near-infrared rice adulteration detection device and method, which can quickly detect the fatty acid value of rice, thereby providing a new and feasible detection method for rice quality control.

[0005] Technical solution: To solve the above problems, the present invention discloses a portable near-infrared rice adulteration detection device, comprising: a housing (1), a light source (4), a grinding and transmission device (2), a spectral acquisition device (5), and a calculation module;

[0006] The emitted light from the light source is transmitted through an optical fiber to the sample detection stage to irradiate the sample for detection. The spectral information carrying the fatty acid content of the sample enters the optical fiber and is then received by the spectral acquisition device (5).

[0007] The grinding and conveying device (2) includes a funnel (2-1), a connecting pipe (2-2), a grinding wall (2-3), a spiral rod (2-4), a rotary motor (2-5), a second touch screen, a linear motor, a sample testing stage (2-9), a quartz vessel, an inlet placement stage (2-8), and a telescopic motor. It is used to grind the sample rice into rice flour and drop it into the quartz vessel. The telescopic motor is used to place the quartz vessel on the sample testing stage (2-9) for testing.

[0008] The spectral acquisition device includes a spectrometer, a reflective optical fiber, a detection platform, and a baffle, and is used to acquire spectral information of different objects.

[0009] The computing module is fixed on the top of the housing and includes a first touch screen, which is used for parameter adjustment and receiving signals emitted by the spectral acquisition device and performing data processing. At the same time, it draws conclusions through data processing and comparison, and the conclusions are displayed on the first touch screen.

[0010] Furthermore, the telescopic motor includes a first telescopic motor, a second telescopic motor, a third telescopic motor, and a fourth telescopic motor. The first telescopic motor uses a telescopic rod to send the quartz vessel placed on the inlet placement platform (2-8) to the position where the rice flour falls in the grinding and conveying device (2).

[0011] Furthermore, after the sample is placed into the funnel (2-1), the rotary motor (2-5) drives the screw rod (2-4) to rotate. The sample is crushed into powder by the crushing action of the grinding wall (2-3) and the screw rod (2-4) and falls into the quartz vessel. The second touch screen indicates that the grinding is complete.

[0012] Furthermore, the connecting tube at the end of the crushed wall (2-3) is designed with an angled opening, so that the sample falls into the quartz vessel by gravity.

[0013] Furthermore, the second touchscreen indicates that the grinding is complete, and the linear motor starts, causing the quartz vessel to vibrate to spread the powder evenly.

[0014] Furthermore, after the powder is spread evenly, the second touchscreen indicates that the powder spreading is complete. The second telescopic motor then uses a telescopic rod to deliver the quartz vessel containing the sample to the sample testing stage (2-9) for testing. Simultaneously, the fourth telescopic motor moves a baffle via the telescopic rod to block external light and prevent interference from the external environment on data acquisition. The near-infrared detection device is activated, and the light source illuminates the bottom center of the quartz vessel containing the sample. The spectrometer collects the spectral data of the sample, and the second touchscreen displays that the testing is complete. The fourth telescopic motor retracts the baffle, and the third telescopic motor uses the telescopic rod to deliver the tested quartz vessel containing the sample to the outlet of the grinding and conveying device. The calculation module processes and compares the data to draw a conclusion, which is displayed on the first touchscreen.

[0015] Furthermore, when a sample needs to be retested, clicking the second touchscreen will trigger a retest, and the fourth telescopic motor will send the quartz vessel containing the sample back to the sample testing station (2-9) via a telescopic rod for testing.

[0016] The detection method based on the portable near-infrared rice adulteration detection device includes the following steps:

[0017] (1) Power on the equipment and place the quartz vessel on the inlet placement platform (2-8);

[0018] (2) The first telescopic motor sends the quartz vessel placed on the inlet placement platform (2-8) to the position where the rice flour falls in the grinding and conveying device (2) through the telescopic rod;

[0019] (3) After the sample is placed in the funnel (2-1), the rotary motor (2-5) drives the screw rod (2-4) to rotate. The sample is crushed into powder by the crushing action of the crushing wall (2-3) and the screw rod (2-4) and falls into the quartz vessel. The second touch screen indicates that the grinding is complete.

[0020] (4) The second touch screen indicates that the grinding is complete, the linear motor starts, and drives the quartz vessel to vibrate to spread the powder evenly. The second touch screen indicates that the powder spreading is complete.

[0021] (5) The second telescopic motor delivers the quartz vessel containing the sample to the sample testing stage (2-9) via the telescopic rod for testing; at the same time, the fourth telescopic motor moves the baffle via the telescopic rod to block external light and prevent the external environment from interfering with data acquisition. The near-infrared detection device is started, the light source shines on the sample, and the spectral information of the detected sample is transmitted to the first touch screen. The second touch screen displays that the detection is complete.

[0022] (6) The fourth telescopic motor retracts the baffle, and the third telescopic motor delivers the quartz vessel containing the sample to the outlet of the grinding and conveying device via the telescopic rod. The calculation module compares the data and draws a conclusion, which is displayed on the first touch screen.

[0023] (7) When a sample needs to be retested, click the second touch screen for retesting. The fourth telescopic motor will send the quartz vessel containing the sample back to the sample testing station (2-9) via the telescopic rod to wait for testing.

[0024] Furthermore, the spectrometer collects spectral data from the quartz vessel containing the sample, the second touchscreen displays that the detection is complete, and then the touchscreen inputs the spectral information of the sample into the prediction model to calculate whether the fatty acid content of the sample exceeds the standard and displays it on the touchscreen.

[0025] Furthermore, the prediction model is a Support Vector Machine (SVC) classification model; or simply SVC model. The steps for building an SVC model are as follows:

[0026] (1) Select multiple rice samples as the calibration set and multiple rice samples as the prediction set;

[0027] (2) Let all rice flour samples stand at room temperature for a period of time, and collect the near-infrared spectral data of the calibration set and the prediction set respectively using a portable near-infrared rice adulteration detection device. That is, the spectral information of the calibration set and the prediction set at each wavelength, and save them as xlsx table format.

[0028] (3) According to the national standard method, the content of fatty acid values ​​of the calibration set and the prediction set is obtained through experiments;

[0029] (4) Remove noise from both sides of the near-infrared spectral data of the calibration set and the prediction set, and use the competitive adaptive reweighted sampling (CARS) algorithm to extract the characteristic wavelengths in the band of 900-1700nm. Select the spectral information of the characteristic wavelengths and use the obtained spectral information and doping ratio to establish the SVC model.

[0030] Furthermore, the sample (rice) to be tested also includes paddy rice that has undergone aging treatment. The aging treatment involves splitting the paddy rice at a 5:2 ratio. If the total weight is 7 kg, 5 kg is used for artificial aging (artificial aging conditions: temperature 37℃, air humidity 85%), and the remaining 2 kg is vacuum-refrigerated and used as new rice in subsequent adulteration experiments. The fatty acid value of the artificially aged paddy rice is measured every 7 days. When the fatty acid value reaches 25 mg KOH / 100g, artificial aging is completed. The aged rice is then mixed with new rice at 10 different ratios: 0%, 5%, 10%, 15%, 20%, 30%, 40%, 60%, 80%, and 100%. The mixed rice serves as the test sample. The spectral data is divided into a calibration set and a prediction set. The calibration set samples are used to build the model, and the prediction set samples are used to test the predictive performance of the model.

[0031] Beneficial effects: This invention is based on near-infrared detection technology and belongs to the category of non-destructive testing instruments. It is environmentally friendly, easy to operate, and portable. This testing equipment can perform sample grinding, sample transfer, spectral acquisition, adulteration prediction, and data storage. Spectral data acquisition performed at the bottom center of a quartz dish inside the testing platform improves detection efficiency and accuracy. The touchscreen can read the spectral information acquired by the spectrometer, and then data analysis can be used to predict multiple qualities of rice. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the structure of the device of the present invention;

[0033] Figure 2 This is a partial cross-sectional view of the housing of the device of the present invention;

[0034] Figure 3 This is an exploded view of the grinding and conveying equipment of the present invention;

[0035] Figure 4 This is a schematic diagram of the grinding equipment of the present invention.

[0036] Figure 5 This is a schematic diagram of the detection platform of the device of the present invention.

[0037] Figure 6 The diagram shown is a flowchart of the method described in this invention.

[0038] In the diagram: 1-Box body, 1-1-Box body wiring fixing interlayer, 2-Grinding transmission equipment, 2-1-Function funnel, 2-2-Connecting pipe, 2-3-Grinding wall, 2-4-Screw rod, 2-5-Rotary motor, 2-6-Second touch screen, 2-7-Grinding equipment housing, 2-8-Sample inlet placement platform, 2-9-Sample testing platform, 2-10-Second telescopic motor, 2-11-First telescopic motor, 2-12-Reflective optical fiber, 2-13-Fourth telescopic motor, 2-14-Testing platform housing, 2-15-Third telescopic motor.

[0039] 2-16-Linear motor, 2-17-Station, 2-18-Quartz vessel, 2-19-Sample exit stage, 3-Power interface

[0040] 4-Light source, 5-Spectrum acquisition device, 6-First touch screen Detailed Implementation

[0041] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0042] like Figure 1 , 2 As shown in Figures 3, 4, and 5, the present invention provides a portable near-infrared rice adulteration detection device and method, comprising: a housing 1, a light source 4, a grinding and transmission device, a spectral acquisition device, and a calculation module. The grinding and transmission device includes a funnel 2-1, a connecting pipe 2-2, a grinding wall 2-3, a spiral rod 2-4, a rotary motor 2-5, a second touchscreen, a linear motor, a sample detection stage 2-9, a quartz vessel, an inlet placement stage 2-8, and a telescopic motor; the spectral acquisition device comprises a spectrometer, a reflective optical fiber 2-12, a sample detection stage 2-9, and a detection platform housing 2-14; the calculation module consists of a first touchscreen.

[0043] Light source 4 is fixed above and to the side of the wiring fixing interlayer 1-1 in the housing. It transmits data to the detection platform 9 via reflective optical fiber 2-12. The spectral information carrying fatty acid content is transmitted through the sample into the transmitting optical fiber and then received by the spectrometer. A spectral acquisition device is used to collect spectral information from different objects. A grinding and transmission device is used to grind rice into rice flour. The transmission device uses a telescopic motor to control the state of the push rod for preprocessing, sample transport, and blocking the light from the detection platform exit. A calculation module is used for parameter adjustment, receiving signals emitted by the spectral acquisition device, and processing data. Simultaneously, through data processing and comparison, a conclusion is drawn, which is displayed on the first touchscreen. In this embodiment, fresh rice provided by the National Grain Reserve is used, and the rice undergoes dehulling, polishing, and aging treatment. Old rice was mixed with new rice at 10 different proportions: 0%, 5%, 10%, 15%, 20%, 30%, 40%, 60%, 80%, and 100%. The mixed rice was used as the test sample. The spectral data was divided into a calibration set and a prediction set. The calibration set samples were used to build the model, and the prediction set samples were used to test the predictive performance of the model.

[0044] The grinding and conveying device 2 includes a funnel 2-1, a connecting pipe 2-2, a grinding wall 2-3, a spiral rod 2-4, a rotary motor 2-5, a second touch screen, a linear motor, a sample testing platform 2-9, a quartz vessel, an inlet placement platform 2-8, and a telescopic motor. It is used to grind sample rice into rice flour and drop it into the quartz vessel. The telescopic motor is used to place the quartz vessel on the sample testing platform 2-9 for testing.

[0045] The spectral acquisition device includes a spectrometer, a reflective optical fiber, a detection platform, and a baffle, and is used to acquire spectral information of different objects.

[0046] The computing module is fixed on the top of the housing and includes a first touch screen, which is used for parameter adjustment and receiving signals emitted by the spectral acquisition device and performing data processing. At the same time, it draws conclusions through data processing and comparison, and the conclusions are displayed on the first touch screen.

[0047] The spectral acquisition device is located on one side of the grinding device and is used for spectral acquisition. It consists of a spectrometer, a reflective optical fiber 2-12, a sample detection stage 2-9, and a detection platform housing 2-14, and is used to acquire spectral information of different objects. The detection platform is located on the side of the grinding device housing and is snapped into a slot.

[0048] The calculation module consists of a first touch screen, which is fixed on the top of the housing 1. It is used to receive signals emitted by the spectral acquisition device and perform data processing. At the same time, it draws conclusions through data processing and comparison, and the conclusions are displayed on the first touch screen.

[0049] like Figure 6As shown, this invention discloses a portable near-infrared rice adulteration detection device and method, comprising the following steps:

[0050] (1) Power on the equipment and place the quartz vessel at the inlet of the grinding and conveying equipment;

[0051] (2) Click the second touch screen to start the equipment. The grinding and conveying equipment will be introduced. The first telescopic motor will extend the push rod. The quartz vessel will reach the designated position. The first telescopic motor will retract the push rod. The second touch screen will prompt you to put in the sample (meter).

[0052] (3) Put the sample (meters) into the funnel, click the second touch screen, the rotary motor drives the screw to rotate, the sample (meters) is turned into powder by the screw and the grinding wall and falls into the quartz vessel, the second touch screen indicates that the grinding is complete;

[0053] (4) The linear motor starts and drives the platform to vibrate and spread the powder evenly. The second touch screen indicates that the powder spreading is complete.

[0054] (5) The second telescopic motor extends the push rod. After the quartz vessel containing the sample (meter) arrives at the detection platform, the second telescopic motor retracts the push rod. At the same time, the fourth telescopic motor located at the waste outlet extends the push rod to block the light at the exit of the detection platform. The near-infrared detection equipment starts and transmits the detected data to the first touch screen. The second touch screen displays that the detection is complete.

[0055] (6) The fourth telescopic motor retracts the push rod, the third telescopic motor extends the push rod, the sample (meter) reaches the outlet of the grinding and conveying equipment, the third telescopic motor retracts the push rod, the touch screen prompts that the test is completed and gives the test result;

[0056] (7) When a sample (meter) needs to be retested, click the second touch screen to retest. The fourth telescopic motor push rod will extend and push the sample (meter) back into the testing platform for testing. The first touch screen will indicate that the test is complete and give the test result.

[0057] (8) Select the desired model from the local file, click the first touch screen, the first touch screen sends a signal to control the operation of the spectrometer, so that the spectrometer collects the spectral information of the quartz vessel 2-18 containing the sample (meter), and then the first touch screen inputs the spectral information of the sample into the prediction model, thereby calculating whether the fatty acid value content of the sample exceeds the standard, and displays whether the fatty acid value content exceeds the standard at the bottom of the first touch screen interface. If you click repeat measurement, the fatty acid value content will be displayed in real time whether it exceeds the standard. At this time, you can choose to save the data or image to your local file. Figure 6 The diagram shows a flowchart of the method described in this invention.

[0058] The steps for establishing the SVC model are as follows:

[0059] (1) Select multiple rice samples as the calibration set and multiple rice samples as the prediction set;

[0060] (2) Let all rice flour samples stand at room temperature for a period of time, and collect near-infrared spectral data of the calibration set and the prediction set respectively, that is, the spectral information of the calibration set and the prediction set at each wavelength, and save them in xlsx table format.

[0061] (3) According to the national standard method, the content of fatty acid values ​​of the calibration set and the prediction set is obtained through experiments;

[0062] (4) Remove noise from both sides of the near-infrared spectral data of the calibration set and the prediction set, extract the characteristic wavelengths in the band of 900-1700nm using the CARS algorithm, select the spectral information of the characteristic wavelengths, and use the obtained spectral information to establish the SVC model.

[0063] The specific method is as follows: Fresh rice was extracted from the national grain reserve and dehulled and polished. A portion of the rice was artificially aged, and the aged rice was mixed with new rice at 10 different proportions: 0%, 5%, 10%, 15%, 20%, 30%, 40%, 60%, 80%, and 100%, resulting in 200 experimental samples of the aged rice mixture. Near-infrared spectral data of the calibration and prediction sets were collected separately, i.e., spectral information of the calibration and prediction sets at various wavelengths. The calibration set samples were used to build the model, and the prediction set samples were used to test the predictive performance of the model. The data were saved in XLSX table format. According to national standards, fatty acid values ​​of the calibration and prediction sets were obtained experimentally. Noise data on both sides of the spectral data were removed. The CARS algorithm was used to extract characteristic wavelengths in the 900-1700nm band. Cross-validation (CV) was used to determine the optimal parameter combination (C, g). The spectral information of the characteristic wavelengths was collected, and the SVC model was built using the obtained spectral information. The accuracy of the calibration set was 95.3%, and that of the prediction set was 96%, indicating high detection precision. In conclusion, the established SVC model meets the detection precision requirements.

Claims

1. A portable near infrared rice adulteration detection device, characterized in that, It comprises a box (1), a light source (4), a grinding transmission device (2), a spectrum acquisition device (5), and a calculation module. The emitted light of the light source (4) is transmitted to the sample detection platform through an optical fiber to irradiate the sample for detection, and the spectrum information carrying the fatty acid value content of the sample enters the optical fiber and is then received by the spectrum acquisition device (5); The grinding transmission device (2) comprises a hopper (2-1), a connecting pipe (2-2), a rolling wall (2-3), a screw rod (2-4), a rotary motor (2-5), a second touch screen, a linear motor, a sample detection platform (2-9), a quartz vessel, an import placement platform (2-8), and a telescopic motor, which are used to grind the sample rice into rice powder and drop it into the quartz vessel, and the telescopic motor is used to place the quartz vessel on the sample detection platform (2-9) for detection. The spectrum acquisition device (5) comprises a spectrometer, a reflection optical fiber, a detection platform, and a baffle, which are used to acquire the spectrum information of different objects. The calculation module is fixed on the top of the box (1) and comprises a first touch screen, which is used for parameter adjustment and receiving the signals emitted by the spectrum acquisition device (5) and performing data processing, and the conclusion is obtained through comparison of the data processing and displayed on the first touch screen. The telescopic motor comprises a first telescopic motor, a second telescopic motor, a third telescopic motor, and a fourth telescopic motor, and the first telescopic motor sends the quartz vessel placed on the import placement platform (2-8) to the position where the rice powder falls in the grinding transmission device (2) through a telescopic rod. After the sample is placed in the hopper (2-1), the rotary motor (2-5) drives the screw rod (2-4) to rotate, and the sample is ground into powder under the rolling action of the rolling wall (2-3) and the screw rod (2-4) and dropped into the quartz vessel, and the second touch screen prompts that the grinding is completed. After the second touch screen prompts that the grinding is completed, the linear motor is started to drive the quartz vessel to vibrate to flatten the powder; after the powder is flattened, the second touch screen prompts that the powder flattening is completed, the second telescopic motor sends the quartz vessel containing the sample to the sample detection platform (2-9) for detection through a telescopic rod; at the same time, the fourth telescopic motor moves the baffle through a telescopic rod to block external light to prevent external environment from interfering with data acquisition, the near-infrared detection equipment is started, the light source (4) irradiates the center end bottom of the quartz vessel containing the sample, the spectrometer acquires the spectrum data of the sample, the second touch screen displays that the detection is completed, the fourth telescopic motor retracts the baffle, the third telescopic motor sends the quartz vessel containing the sample after detection to the outlet of the grinding transmission device (2) through a telescopic rod, and the calculation module obtains the conclusion through comparison of the data processing, which is displayed on the first touch screen. The connecting pipe at the end of the rolling wall (2-3) is designed as a beveled opening, and the sample falls into the quartz vessel through gravity.

2. The portable near infrared rice adulteration detection device according to claim 1, characterized in that, When the sample needs to be re-detected, click the second touch screen to recheck, and the fourth telescopic motor sends the quartz vessel containing the sample back to the sample detection platform (2-9) for detection through a telescopic rod.

3. The portable near infrared rice adulteration detection device according to claim 1, characterized in that, The following steps are included:

4. The method for detecting the portable near-infrared rice adulteration detection device according to claim 1, characterized in that, (1) Power on the equipment and place the quartz vessel on the import placement platform (2-8). ​ (2) The first telescopic motor sends the quartz container placed on the inlet placing table (2-8) to the grinding transmission device (2) through the telescopic rod to the position where the powder falls off; (3) After the sample is placed in the funnel (2-1), the rotary motor (2-5) drives the screw rod (2-4) to rotate, and the sample is crushed into powder under the crushing action of the crushing wall (2-3) and the screw rod (2-4), and then falls into the quartz container, and the second touch screen prompts that the grinding is completed; (4) After the second touch screen prompts that the grinding is completed, the linear motor starts to drive the quartz container to vibrate to flatten the powder, and the second touch screen prompts that the powder flattening is completed; (5) The second telescopic motor sends the quartz container containing the sample to the sample detection table (2-9) for detection through the telescopic rod; meanwhile, the fourth telescopic motor moves the baffle through the telescopic rod to shield external light and prevent external environment from interfering with data collection, and the near-infrared detection equipment starts to work, the light source (4) irradiates the sample, and the spectral data of the detected sample are transmitted to the first touch screen, and the second touch screen displays that the detection is completed; (6) The fourth telescopic motor retracts the baffle, the third telescopic motor sends the quartz container containing the sample after detection to the grinding transmission device (2) outlet through the telescopic rod, and the calculation module compares the data to draw a conclusion, which is displayed on the first touch screen; (7) When the sample needs to be re-detected, click the second touch screen to recheck, and the fourth telescopic motor sends the quartz container containing the sample back to the sample detection table (2-9) for detection.

5. The method of claim 4, wherein the portable near infrared rice adulteration detection device is characterized by, The spectrometer collects the spectral information of the sample and transmits it to the first touch screen, and then the first touch screen inputs the spectral information of the sample into the prediction model to calculate whether the fatty acid value content of the sample exceeds the standard and displays it on the touch screen.

6. The method of claim 5, wherein the portable near infrared rice adulteration detection device is characterized by: The prediction model is a support vector machine classification model, and the establishment steps are: (1) Select multiple rice samples as the calibration set and multiple rice samples as the prediction set; (2) Let all the rice powder samples stand at room temperature for a period of time, collect the near-infrared spectral data of the calibration set and the prediction set by the portable near-infrared rice adulteration detection equipment, i.e. the spectral information of the calibration set and the prediction set at each wavelength, and save them in xlsx table format; (3) According to the national standard method, the content of the fatty acid value of the calibration set and the prediction set is obtained through experiments; (4) Remove the noise on both sides of the near-infrared spectral data of the calibration set and the prediction set, extract the characteristic wavelengths in the wavelength range of 900-1700 nm by using the competitive self-adaptive reweighted sampling method, select the spectral information of the characteristic wavelengths, and establish an SVC model using the obtained spectral information and the adulteration ratio.

Images