A transmission spectroscopy non-destructive testing device and method

Abstract

The application discloses a kind of transmission spectrum nondestructive testing device and method, belong to fruit quality nondestructive testing field.The present application includes device shell, halogen tungsten lamp, collimating lens, pinhole diaphragm, multispectral sensor, raspberry controller, display screen and battery.Working, the light emitted by halogen tungsten lamp light source is collimated after collimating lens and passes through cherry tomato, and the light carrying cherry tomato internal quality information is filtered by pinhole diaphragm after passing through, and is received by multispectral sensor, and multispectral sensor converts optical signal into electrical signal and transmits to raspberry, and raspberry is analyzed in real time to obtain soluble solids content, hardness index and display.The present application can complete detection using separate button, uses multispectral sensor as spectral collection element, and greatly reduces the development cost of device.The present application realizes portable cherry tomato internal quality detection device, and can be used for cherry tomato key quality index on-site rapid nondestructive testing.

Description

Technical Field

[0001] This invention relates to a transmission spectroscopy nondestructive testing device and method, belonging to the field of nondestructive testing technology for fruit quality. Background Technology

[0002] Cherry tomatoes, a cultivated variety of the genus *Solanum* in the Solanaceae family, belong to the same fruit family as common tomatoes, but possess unique biological characteristics and economic value. With the continuous expansion of cherry tomato cultivation and the increasing abundance of market supply, consumers are gradually shifting their focus from simply on the appearance of the product to placing higher demands on its intrinsic quality. Many factors influence the taste of cherry tomatoes, among which soluble solids content (SSC) primarily reflects the soluble sugar content and is closely related to its sweetness, while firmness directly affects the crispness and storage and transportation resistance of the fruit. Therefore, the rapid and accurate detection of SSC and firmness in cherry tomatoes is crucial for their post-harvest grading, storage and transportation optimization, and enhancement of market competitiveness.

[0003] Traditional methods for testing the quality of cherry tomatoes rely on manual judgment based on external characteristics such as color, size, and texture. This method suffers from significant identification errors and inconsistent classification standards. While physicochemical testing is more accurate for determining cherry tomato quality, it is cumbersome, time-consuming, consumes large amounts of reagents, is costly, and only allows for random sampling. In recent years, near-infrared spectroscopy has made significant progress in fruit quality testing, offering unique advantages of being green and non-destructive. By combining various advanced algorithms and optimization strategies, it has achieved high-precision prediction of key indicators such as firmness, SSC (Self-Containing Score), and pH value. However, current research mainly focuses on theoretical verification in laboratory environments, and the testing systems are mostly large, benchtop devices that are expensive and difficult to meet the needs of real-time on-site testing.

[0004] Therefore, there is an urgent need to propose a non-destructive testing device and method based on transmission spectroscopy to solve the above-mentioned technical problems. Summary of the Invention

[0005] To address the aforementioned problems, a transmission spectroscopy nondestructive testing device and method are provided. A brief overview of the invention is given below to provide a basic understanding of certain aspects of the invention. It should be understood that this overview is not an exhaustive summary of the invention. It is not intended to identify key or essential parts of the invention, nor is it intended to limit the scope of the invention.

[0006] The technical solution of the present invention:

[0007] A transmission spectroscopy non-destructive testing device includes: a halogen tungsten lamp, a collimating lens, a stage, an aperture, and a multispectral sensor arranged in sequence, all installed inside the device housing. A cooling fan is provided on the side of the halogen tungsten lamp.

[0008] Preferably, the first mounting bracket includes: a power supply module, a control switch, and a controller that are electrically connected in sequence; the controller is electrically connected to the display screen; and the controller is electrically connected to the multispectral sensor via control buttons.

[0009] The charging and discharging module, battery pack, switch, and boost module are electrically connected in sequence. The boost module is electrically connected to the halogen lamp and cooling fan.

[0010] Preferably, the second mounting bracket includes: a first outer shell and a second outer shell connected by bolts in the middle, the first outer shell and the second outer shell are arranged crosswise, the left side structure is the acquisition chamber, and the right side structure is the handle;

[0011] The left side structure of the first outer shell is fixedly installed with a halogen tungsten lamp, a cooling fan, and a collimating lens. The left side structure of the second outer shell is fixedly installed with a stage with a circular hole in the center, an aperture, and a multispectral sensor. The left side structures of the first and second outer shells are arranged correspondingly. During closed detection, the halogen tungsten lamp, collimating lens, circular hole of the stage, aperture, and multispectral sensor are coaxially arranged. A sponge is placed at the circular hole of the stage.

[0012] The right side structure of the first outer casing houses a boost module, a charge / discharge module, and a battery pack. A switch is located on the outer wall of the right side structure of the first outer casing. A control switch is located on the outer wall of the right side structure of the second outer casing. A display screen and control buttons are located on the top of the outer wall of the right side structure of the second outer casing. A power supply module and a controller are located inside the right side structure of the second outer casing.

[0013] Preferably, it also includes: a halogen lamp holder and a halogen lamp holder bracket. The halogen lamp holder and the halogen lamp holder bracket are fixedly installed on the upper part of the left side structure of the first housing. The halogen lamp holder bracket is bolted to the halogen lamp housing, and the halogen lamp holder is electrically connected to the halogen lamp.

[0014] Preferred configurations include: a 5V turbine cooling fan, an OLED display screen, a UPS power supply module, a 5V boost module, a 5V integrated charging and discharging module, and an 18650 battery pack.

[0015] Preferred method: The transmission spectroscopy nondestructive testing device includes the following steps:

[0016] The visible-near-infrared light emitted by the halogen tungsten lamp passes through a collimating lens. The collimated light then passes through the sample, and the light carrying information about the sample's interior passes through an aperture and is received by a spectral sensor. The spectral sensor converts the light signal into an electrical signal and transmits it to the controller for processing. The calculation results are then displayed on the screen.

[0017] Preferably, the method includes the following steps:

[0018] Step a. Turn on the controller: Turn on the control switch to power the controller, OLED display, and multispectral sensor;

[0019] Step b. Acquire reference black spectrum T d Press the control button to collect the black spectrum T of the current environment. d ;

[0020] Step c. Turn on the halogen lamp: Turn on the switch to turn on the halogen lamp light source and the cooling fans on both sides of the halogen lamp, and wait for it to preheat;

[0021] Step d. Acquire reference white spectrum T w Press the control button to collect the white spectrum T of the current environment. w ;

[0022] Step e. Collect the transmission spectrum T of the sample. s Using the bolt as an axis, open the collection chamber, place the sample horizontally on the stage of the collection chamber, ensuring its curved surface aligns with the elliptical hole on the stage to create a sealed, light-blocking environment. Simultaneously, align its equatorial portion with the small aperture below. Close the collection chamber and press the control button. At this point, the spectral information of the cherry tomato is transmitted to the multispectral sensor, completing the sample's transmission spectrum T. s collection;

[0023] Step f. Calculate the sample transmission correction spectrum T c Based on the collected black reference spectrum T d White reference spectrum T w and the original spectrum T s Using formula Calculate the transmission corrected spectrum T of the sample. c Assume the wavelength range of the calibration spectrum Tc is 400-950 nm.

[0024] Step g. Establishing an internal quality prediction model for samples: The transmission correction spectrum collected by the detection device is used as the input, and the physicochemical values ​​of key quality indicators of the samples obtained by traditional destructive testing methods such as sugar refractometer and hardness tester are used as the output; after the first derivative FD preprocessing method, partial least squares regression (PLSR) and principal component regression (PCR) quantitative prediction models are established. By comparing the correlation coefficient of the calibration set, the root mean square error of the calibration set (RMSEC), the correlation coefficient of the prediction set, the root mean square error of the prediction set (RMSEP), and the residual prediction deviation (RPD) evaluation index, the optimal internal quality prediction model for cherry tomatoes is the partial least squares regression (PLSR) model.

[0025] Step f. Model Deployment: Deploy the in-sample quality partial least squares regression prediction model to the controller;

[0026] Step h. Predict the internal quality index of the sample: Turn on the halogen lamp switch to turn on the halogen lamp light source, turn on the control switch to turn on the controller, collect the spectrum of the sample to be tested, correct it with the formula, and then input it into the partial least squares regression (PLSR) prediction model to obtain the internal quality index of the sample.

[0027] Preferred: The sample is a cherry tomato. In step h, the internal quality indicators of the sample include the specific values ​​of soluble solids content (SSC) and firmness, and the results are displayed on the screen.

[0028] The present invention has the following beneficial effects:

[0029] 1. All components of the portable detection device of the present invention are integrated inside the device, making it convenient to carry and easy to operate;

[0030] 2. This invention selects a transmission mode for detection and device design, which, compared with the traditional reflection mode, can better acquire the spectral information of the internal pulp and reduce the interference of external stray light and epidermal texture information on the detection results;

[0031] 3. This invention uses a high-current boost module to ensure the stability of the halogen tungsten lamp light source, and the module itself generates low heat, reducing the risk of softening of the device casing;

[0032] 4. This invention uses a multispectral sensor as the spectral acquisition element, which greatly reduces the development cost of the device;

[0033] 5. This invention proposes an optimal detection model for the internal quality of cherry tomatoes, which can achieve accurate detection of key quality parameters of samples. Attached Figure Description

[0034] Figure 1 This is an overall structural diagram of a transmission spectroscopy non-destructive testing device.

[0035] Figure 2 This is a top view of a transmission spectroscopy non-destructive testing device.

[0036] Figure 3 This is a schematic diagram of the acquisition system control of a transmission spectroscopy nondestructive testing device.

[0037] Figure 4 This is a schematic diagram of the control system for a transmission spectroscopy nondestructive testing device.

[0038] Figure 5 These are the original visible-near-infrared transmission spectra of the sample in Embodiment 1 of the present invention and the spectra after processing with the first derivative FD.

[0039] Figure 6 This is a scatter plot of the predicted and measured values ​​of soluble solids content and hardness in Example 1 of the present invention.

[0040] Figure 7 This is a physical image of a transmission spectroscopy non-destructive testing device. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention is described below with reference to specific embodiments shown in the accompanying drawings. However, it should be understood that these descriptions are merely exemplary and not intended to limit the scope of the invention. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concept of the invention.

[0042] Specific implementation method one: Combining Figure 1-4 This embodiment describes a transmission spectroscopy non-destructive testing device, comprising: a halogen tungsten lamp 3, a cooling fan 4, a collimating lens 5, a stage 6, an aperture 7, a multispectral sensor 8, and a housing 10; the halogen tungsten lamp 3, the collimating lens 5, the stage 6, the aperture 7, and the multispectral sensor 8 are arranged coaxially from top to bottom and installed inside the housing 10 of the device; a cooling fan 4 is provided on the side of the halogen tungsten lamp 3; this provides a real-time, portable non-destructive testing method for the internal quality of cherry tomatoes.

[0043] Specific Implementation Method Two: Combining Figure 1-4 This embodiment describes a non-destructive testing device for transmission spectroscopy, comprising: a halogen tungsten lamp 3, a cooling fan 4, a collimating lens 5, a stage 6, an aperture 7, a multispectral sensor 8, and a housing 10; the halogen tungsten lamp 3, the collimating lens 5, the stage 6, the aperture 7, and the multispectral sensor 8 are arranged coaxially from top to bottom and installed inside the housing 10; the cooling fan 4 is provided on the side of the halogen tungsten lamp 3; the stage 6 is used to place samples (e.g., cherry tomatoes);

[0044] It also includes: control switch 11, display screen 12, control button 13, controller 14, power supply module 15, boost module 16, charging and discharging module 17, switch 18 and battery pack 19; power supply module 15, control switch 11 and controller 14 are electrically connected in sequence, controller 14 is electrically connected to display screen 12, and controller 14 is electrically connected to multispectral sensor 8 through control button 13;

[0045] The charging / discharging module 17, battery pack 19, switch 18, and boost module 16 are sequentially and electrically connected. Boost module 16 is electrically connected to halogen lamp 3 and cooling fan 4. Control switch 11 is a Raspberry Pi control switch, controller 14 is a Raspberry Pi controller, and switch 18 is a halogen lamp switch. Control button 13 can be set to turn on when pressed and automatically turn off when not pressed. During operation, the light emitted from the halogen lamp light source is collimated by a collimating lens and passes through the cherry tomato. The light carrying internal quality information of the cherry tomato is filtered through a pinhole aperture and received by a multispectral sensor. The multispectral sensor converts the light signal into an electrical signal and transmits it to the Raspberry Pi. The Raspberry Pi analyzes and displays the soluble solids content and hardness index in real time. This invention uses a single button to complete the detection and uses a multispectral sensor as the spectral acquisition element, greatly reducing the development cost of the device. This invention realizes a portable cherry tomato internal quality detection device, which can be used for rapid, non-destructive on-site detection of key quality indicators of cherry tomatoes.

[0046] The outer shell 10 includes a first outer shell and a second outer shell, which are arranged symmetrically on top of each other. It also includes bolts 9. The first and second outer shells of the outer shell 10 are rotatably connected by bolts 9, so that the first and second outer shells are arranged in a cross manner. The left side structure is the collection chamber and the right side structure is the handle. Based on the ease of operation, the existing structure is used to carry out a reasonable layout design.

[0047] The left side structure of the first outer shell is fixedly installed with a halogen tungsten lamp 3, a cooling fan 4, and a collimating lens 5. The left side structure of the second outer shell is fixedly installed with a stage 6 with a circular hole in the center, an aperture 7, and a multispectral sensor 8. The left side structures of the first and second outer shells are arranged in a corresponding manner to form a detection chamber inside the installation chamber. When the detection is closed, the halogen tungsten lamp 3, the collimating lens 5, the circular hole of the stage 6, the aperture 7, and the multispectral sensor 8 are arranged coaxially. A sponge is placed at the circular hole of the stage 6. The sponge is a black, high-temperature resistant, annular light-shielding sponge. The aperture 7 is a pinhole aperture.

[0048] The right side of the first outer shell houses a boost module 16, a charge / discharge module 17, and a battery pack 19. A switch 18 is located on the outer wall of the right side of the first outer shell. A control switch 11 is located on the outer wall of the right side of the second outer shell. A display screen 12 and control buttons 13 are located on the top of the outer wall of the right side of the second outer shell. A power supply module 15 and a controller 14 are located inside the right side of the second outer shell. This design facilitates operation of the buttons / switches and observation of the display screen, allowing for one-handed operation and convenient portability. The sample is placed at the circular hole of the stage to ensure it is stably centered. Holding the handle closes the collection chamber.

[0049] It also includes: halogen lamp holder 1 and halogen lamp holder bracket 2. The halogen lamp holder 1 and halogen lamp holder bracket 2 are fixedly installed on the upper part of the left side structure of the first housing. The halogen lamp holder bracket 2 is bolted to the housing of the halogen lamp 3. The halogen lamp holder 1 is electrically connected to the halogen lamp 3.

[0050] Cooling fan 4 uses a 5V turbine cooling fan, display screen 12 uses an OLED display, power supply module 15 uses a Raspberry Pi UPS power supply module, boost module 16 uses a 5V boost module, charging and discharging integrated module 17 uses a 5V charging and discharging integrated module, and battery pack 19 uses an 18650 battery pack.

[0051] Specific implementation method three: Combining Figure 1-4 This embodiment describes a transmission spectroscopy nondestructive testing device and method. The device comprises a halogen tungsten lamp 3, a cooling fan 4, a collimating lens 5, a stage 6, an aperture 7, a multispectral sensor 8, and a housing 10. The halogen tungsten lamp 3, collimating lens 5, stage 6, aperture 7, and multispectral sensor 8 are arranged coaxially from top to bottom and installed inside the housing 10. A cooling fan 4 is provided on the side of the halogen tungsten lamp 3. The stage 6 is used to place samples (e.g., cherry tomatoes).

[0052] The method includes the following steps:

[0053] The visible-near-infrared light emitted by the halogen tungsten lamp 3 as a light source passes through the collimating lens 5. The collimated light passes through the sample, and the light carrying the internal information of the sample passes through the aperture 7 and is received by the spectral sensor 8. The spectral sensor 8 converts the light signal into an electrical signal and transmits it to the controller 14 for processing. The calculation result is displayed on the display screen 12.

[0054] Specific implementation method four: Combination Figure 1-4 This embodiment describes a transmission spectroscopy nondestructive testing device and method. The device comprises a halogen tungsten lamp 3, a cooling fan 4, a collimating lens 5, a stage 6, an aperture 7, a multispectral sensor 8, and a device housing 10. The halogen tungsten lamp 3, collimating lens 5, stage 6, aperture 7, and multispectral sensor 8 are arranged coaxially from top to bottom and installed inside the device housing 10. A cooling fan 4 is provided on the side of the halogen tungsten lamp 3. The stage 6 is used to place samples (e.g., cherry tomatoes).

[0055] The method includes the following steps:

[0056] Step a. Turn on the controller: Flip the control switch 11 to power on the controller 14, OLED display 12, and multispectral sensor 8;

[0057] Step b. Acquire reference black spectrum T dAccording to the prompts on display screen 12, first press control button 13 to collect the black spectrum T of the current environment. d ;

[0058] Step c. Turn on the halogen lamp: Turn on the halogen lamp light source 3 and the cooling fans 4 on both sides of the halogen lamp by switching switch 18. Wait 15 seconds for the light source to preheat.

[0059] Step d. Acquire reference white spectrum T w According to the prompts on display screen 12, press control button 13 to collect the white spectrum T of the current environment. w ;

[0060] Step e. Collect the transmission spectrum T of the sample. s Using bolt 9 as the axis, open the collection chamber and place the sample horizontally on the stage 6 of the collection chamber, ensuring its curved surface aligns with the elliptical hole on the stage 6 to create a closed, light-shielding environment. Simultaneously, align its equatorial portion with the aperture 7 below. Close the collection chamber 9 and, according to the prompts on the display screen 12, press the control button 13. At this point, the spectral information of the cherry tomato is transmitted to the multispectral sensor 8, completing the transmission spectrum T of the sample. s collection;

[0061] Step f. Calculate the sample transmission correction spectrum T c Based on the collected black reference spectrum T d White reference spectrum T w and the original spectrum T s Using formula Calculate the transmission corrected spectrum T of the sample. c Assume the wavelength range of the calibration spectrum Tc is 400-950 nm.

[0062] Step g. Establishing an internal quality prediction model for samples: The transmission correction spectrum collected by the detection device is used as the input, and the physicochemical values ​​of key quality indicators of the samples obtained by traditional destructive testing methods such as sugar refractometer and hardness tester are used as the output; after the first derivative FD preprocessing method, partial least squares regression (PLSR) and principal component regression (PCR) quantitative prediction models are established. By comparing the correlation coefficient of the calibration set, the root mean square error of the calibration set (RMSEC), the correlation coefficient of the prediction set, the root mean square error of the prediction set (RMSEP), and the residual prediction deviation (RPD), the optimal internal quality prediction model for cherry tomatoes is the partial least squares regression (PLSR) model.

[0063] Step f. Model Deployment: Deploy the in-sample quality partial least squares regression prediction model onto controller 14;

[0064] Step h. Predict the internal quality index of the sample: Turn on the halogen lamp light source 3 by switching the halogen lamp switch 18, turn on the controller 14 by switching the control switch 11, collect the spectrum of the sample to be tested, and after correction by the formula, input it into the partial least squares regression (PLSR) prediction model to obtain the internal quality index of the sample.

[0065] In step h, the specific values ​​of the internal quality indicators of the sample include the soluble solids content (SSC) and hardness, and the specific values ​​of the indicator results are displayed on the display screen 12.

[0066] This invention is based on visible / near-infrared transmission spectroscopy technology. The detection optical path and hardware circuit are designed according to the size and shape characteristics of cherry tomatoes. Real-time detection software is developed based on spectral sensors. A portable device that can quickly and non-destructively detect the SSC and hardness of cherry tomatoes is developed. The device uses self-made instruments to collect spectral data, establishes optimal prediction models for SSC and hardness respectively, and embeds them into the portable detection device to achieve rapid and non-destructive on-site detection of cherry tomato quality.

[0067] Example 1:

[0068] Combination Figure 1-7 The aforementioned non-destructive testing device for cherry tomatoes using transmission spectroscopy, such as... Figure 1 , Figure 2As shown, the portable cherry tomato internal quality transmission spectroscopy detection device comprises a halogen tungsten lamp holder 1, a halogen tungsten lamp holder bracket 2, a halogen tungsten lamp 3, a cooling fan 4, a collimating lens 5, a stage 6, an aperture 7, a multispectral sensor 8, bolts 9, a housing 10, a control switch 11, a display screen 12, control buttons 13, a controller 14, a UPS power supply module 15, a boost module 16, a charge / discharge integrated module 17, a halogen tungsten lamp switch 18, and a battery pack 19; the halogen tungsten lamp has a voltage of 6V, a power of 10W, and a wavelength range of 350-2500 nm. nm; The halogen tungsten lamp holder bracket 2 is made of high-temperature resistant printing material and is fixed on the outer shell 10 of the detection device. A central opening is used to fix the halogen tungsten lamp holder 1, and the overall structure is hollow to facilitate heat dissipation of the halogen tungsten lamp 3; The halogen tungsten lamp holder 1 is a G4 lamp holder and is fixed on the halogen tungsten lamp holder bracket 2; The pins of the halogen tungsten lamp 3 are inserted into the halogen tungsten lamp holder 1; The cooling fans 4 are symmetrically installed on both sides of the halogen tungsten lamp 3; The collimating lens 5 is fixed to the outer shell 10 of the device via a 3D printed bracket and is coaxial with the halogen tungsten lamp 3; The stage 6 has a circular hole in the center with a diameter of 10mm, coaxial with the collimating lens 5, and the area around the circular hole is black and high-temperature resistant. A ring-shaped light-blocking sponge is used; the pinhole aperture 7, with a diameter of 6mm, is fixed to the device housing 10 and coaxial with the collimating lens 5; the multispectral sensor 8 is positioned opposite the lower part of the pinhole aperture 7 and is fixed to the device housing 10 via a 3D-printed bracket, connected to the I2C pin of the controller 14; the bolt 9 is used to fix the two parts of the device housing and also serves as a rotating shaft to achieve sample collection; the device housing 10 is divided into upper and lower parts connected by the bolt 9, forming a clamping structure, consisting of a handle and a collection chamber. The collection chamber consists of two opposing container-like structures used to hold the sample, with the upper part of the collection chamber serving as the top. The device has an opening to facilitate heat dissipation for the internal halogen tungsten lamp 3; the control switch 11 is connected to the UPS power supply module 15 to power the Raspberry Pi controller 14; the OLED display screen 12 is located on the upper handle of the device and is connected to the I2C pin of the controller 14; the control button 13 is located at the lower part of the OLED display screen, with one end of its pin connected to the multispectral sensor 8 and the other end connected to the Raspberry Pi controller 14; the Raspberry Pi controller 14 is connected to the Raspberry Pi UPS power supply module 15 via its own pins and is fixed directly above the Raspberry Pi UPS power supply module 15; the UPS power supply module 15... The boost module 16 is fixed inside the upper part of the handle of the device; the input end is connected to the battery pack 19, and the output end is connected to the halogen tungsten lamp 3 and the 5V turbine cooling fan; the output end of the charge-discharge integrated module 17 is connected to the battery pack 19 and fixed to the lower part of the handle of the device housing 10; the switch 18 is connected between the boost module 16 and the battery pack 19 and is used to control the power supply of the halogen tungsten lamp 3 and the 5V turbine cooling fan 4; the 18650 battery pack 19 consists of two 3.7V batteries connected in parallel and fixed to the lower part of the handle of the device; the UPS power supply module 15 contains an independent 3.7V battery.A 7V lithium battery converts 3.7V to 5V via its own boost module, powering the controller 14. The 5V boost module 16, with a maximum current capacity of 21A, raises the voltage of the battery pack 19 from 3.7V to 5V, powering the halogen lamp 3 and cooling fan 4. The multispectral sensor 8, display screen 12, and control buttons 13 are all controlled by the controller 14. The multispectral sensor 8 collects sample spectral signals through the aperture 7, converts them into electrical signals, transmits them to the controller 14 for processing and calculation, and displays the results on the display screen 12. The outer shell 10 is made of 3D printing material PETG-HF, and the halogen lamp holder is made of high-temperature resistant 3D printing material PET-CF. All printing materials are black to prevent reflection caused by the halogen lamp 3 irradiating the inner wall. The portable cherry tomato internal quality detection device based on transmission spectroscopy has a length, width, and height of 240 mm, 60 mm, and 90 mm, and weighs 256 g. Figure 7 The black button below the OLED display is used to control the multispectral sensor to collect spectra. After the Raspberry Pi receives the spectral signal, it is fed into the detection model and the result is output to the display screen.

[0069] A non-destructive testing method for cherry tomatoes using transmission spectroscopy includes the following steps:

[0070] Sample preparation: In this embodiment, 165 cherry tomatoes with uniform shape and size and no surface damage were selected, numbered, and placed at room temperature for more than 24 hours.

[0071] Visible-near-infrared spectral acquisition includes the following steps:

[0072] a. Turn on the Raspberry Pi controller: Flip the Raspberry Pi control switch to power the Raspberry Pi controller, OLED display, and multispectral sensor 8;

[0073] b. Acquisition of reference black spectrum T d According to the prompts on display screen 12, first press control button 13 to collect the black spectrum T of the current environment. d ;

[0074] c. Turn on the halogen lamp: Turn on the halogen lamp light source 3 and the cooling fans 4 on both sides of the halogen lamp by toggling the halogen lamp switch 18. Wait 15 for the light source to preheat.

[0075] d. Acquire reference white spectrum T w According to the prompts on display screen 12, press control button 13 to collect the white spectrum T of the current environment. w ;

[0076] e. Transmission spectrum T of the collected sample sUsing bolt 9 as the axis, open the collection chamber and place the sample horizontally on the stage 6 of the collection chamber, ensuring its curved surface aligns with the elliptical hole on the stage 6 to create a closed, light-shielding environment. Simultaneously, align its equatorial portion with the aperture 7 below. Close the collection chamber 9 and, according to the prompts on the display screen 12, press the control button 13. At this point, the spectral information of the cherry tomato is transmitted to the multispectral sensor 8, completing the transmission spectrum T of the sample. s collection;

[0077] f. Calculate the sample transmission correction spectrum T c Based on the collected black reference spectrum T d White reference spectrum T w and the original spectrum T s Using formula Calculate the transmission corrected spectrum T of the sample. c ;

[0078] The wavelength range of the corrected spectrum Tc for the steamed bun is 400-950 nm.

[0079] Physicochemical index determination: Physicochemical values ​​of soluble solids and hardness of key quality indicators of the sample obtained by traditional destructive testing methods such as sugar content refractometer and hardness tester.

[0080] Construction of internal quality prediction model for cherry tomatoes: The transmission calibrated spectrum collected by the detection device is used as the input, and the physicochemical values ​​of key quality indicators of the sample obtained by traditional destructive testing methods such as sugar refractometer and hardness tester are used as the output; after the first derivative FD preprocessing method, partial least squares regression (PLSR) and principal component regression PCR quantitative prediction models are established. By comparing the correlation coefficient of the calibration set, the root mean square error of the calibration set (RMSEC), the correlation coefficient of the prediction set, the root mean square error of the prediction set (RMSEP), and the residual prediction deviation (RPD), the optimal internal quality prediction model for cherry tomatoes is the partial least squares regression (PLSR) model.

[0081] Model Deployment: Deploy the partial least squares regression prediction model for the internal quality of cherry tomatoes onto the Raspberry Pi controller;

[0082] Predicting internal quality indicators of cherry tomatoes: Turn on the halogen lamp light source 3 by switching the halogen lamp switch 18, turn on the Raspberry Pi controller by switching the Raspberry Pi control switch 11, collect the spectrum of the sample to be tested, and after correction by the formula, input it into the partial least squares regression (PLSR) prediction model to obtain the specific values ​​of the internal quality indicators of cherry tomatoes, namely the soluble solids content (SSC) and hardness, and display the results on the display screen 12.

[0083] The analysis results are shown below: Figure 5The images show the original transmission spectrum and the transmission spectrum curves after first-order derivative (FD) preprocessing, collected by the portable cherry tomato internal quality transmission spectrum detection device described in Example 1. Figure 6 The image shows a scatter plot of the predicted and actual values ​​obtained from the optimal prediction model for soluble solids and hardness in Example 1. The correlation coefficients between the calibration set and the prediction set for soluble solids content are 0.926 and 0.913, respectively; the root mean square errors (RMS) of the calibration set and the prediction set are 0.322 and 0.365, respectively; and the residual prediction deviation (RPD) is 2.203. The correlation coefficients between the calibration set and the prediction set for hardness are 0.873 and 0.871, respectively; the RMS of the calibration set and the prediction set are 0.949 and 1.048, respectively; and the residual prediction deviation (RPD) is 2.046.

[0084] It should be noted that in the above embodiments, as long as the technical solutions are not contradictory, they can be permuted and combined. Those skilled in the art can exhaust all possibilities based on the mathematical knowledge of permutation and combination. Therefore, the present invention will not describe the technical solutions after permutation and combination one by one, but it should be understood that the technical solutions after permutation and combination have been disclosed by the present invention.

[0085] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A transmission spectroscopy nondestructive testing device, characterized in that: include: The halogen tungsten lamp (3), collimating lens (5), stage (6), aperture (7), and multispectral sensor (8) arranged in sequence are installed inside the housing (10), and a cooling fan (4) is provided on the side of the halogen tungsten lamp (3). It also includes: a power supply module (15), a control switch (11), and a controller (14) connected in sequence, the controller (14) being connected to the display screen (12) and the controller (14) being connected to the multispectral sensor (8) via a control button (13); The charging and discharging module (17), battery pack (19), switch (18), and boost module (16) are electrically connected in sequence. The boost module (16) is electrically connected to the halogen lamp (3) and cooling fan (4). The first and second outer shells of the outer shell (10) are rotatably connected by bolts (9). The first and second outer shells are arranged in a cross pattern. The structure on the left is the collection chamber, and the structure on the right is the handle. The left side structure of the first outer shell is fixedly installed with a halogen tungsten lamp (3), a cooling fan (4), and a collimating lens (5). The left side structure of the second outer shell is fixedly installed with a stage (6) with a circular hole in the center, an aperture (7), and a multispectral sensor (8). The left side structures of the first and second outer shells are set up correspondingly. During closed detection, the halogen tungsten lamp (3), the collimating lens (5), the circular hole of the stage (6), the aperture (7), and the multispectral sensor (8) are set coaxially. A sponge is provided at the circular hole of the stage (6). The first outer shell has a boost module (16), a charge-discharge module (17), and a battery pack (19) inside the right side structure. A switch (18) is provided on the outer wall of the right side structure of the first outer shell. A control switch (11) is provided on the outer wall of the right side structure of the second outer shell. A display screen (12) and a control button (13) are provided on the top of the outer wall of the right side structure of the second outer shell. A power supply module (15) and a controller (14) are provided inside the right side structure of the second outer shell. Place the sample at the circular hole of the stage to ensure it is stably centered. Hold the handle to close the collection chamber. The sample is a cherry tomato.

2. The transmission spectroscopy nondestructive testing device according to claim 1, characterized in that: Also includes: Halogen lamp holder (1) and halogen lamp holder bracket (2) are fixedly installed on the upper part of the left side structure of the first housing. The halogen lamp holder bracket (2) is bolted to the housing of the halogen lamp (3). The halogen lamp holder (1) is electrically connected to the halogen lamp (3).

3. The transmission spectroscopy nondestructive testing device according to claim 2, characterized in that: The cooling fan (4) adopts a 5V turbine cooling fan, the display screen (12) adopts an OLED display screen, the power supply module (15) adopts a UPS power supply module, the boost module (16) adopts a 5V boost module, the charging and discharging integrated module (17) adopts a 5V charging and discharging integrated module, and the battery pack (19) adopts an 18650 battery pack.

4. A non-destructive testing method using transmission spectroscopy, characterized in that: The transmission spectroscopy nondestructive testing device according to any one of claims 1-3 includes the following steps: The visible-near infrared light emitted by the halogen tungsten lamp (3) passes through the collimating lens (5), and the collimated light passes through the sample. The light carrying the internal information of the sample passes through the aperture (7) and is received by the multispectral sensor (8). The multispectral sensor (8) converts the light signal into an electrical signal and transmits it to the controller (14) for processing. The calculation result is displayed on the display screen (12).

5. The transmission spectroscopy nondestructive testing method according to claim 4, characterized in that: The method includes the following steps: Step a. Turn on the controller: Turn on the control switch (11) to complete the power supply operation of the controller (14), display screen (12), and multispectral sensor (8); Step b. Acquire reference black spectrum T d Press the control button (13) to collect the black spectrum of the current environment. T d ; Step c. Turn on the halogen lamp: Turn on the switch (18) to turn on the halogen lamp (3) and the cooling fans (4) on both sides of the halogen lamp, and wait for preheating; Step d. Acquire reference white spectrum T w Press the control button (13) to collect the white spectrum of the current environment. T w ; Step e. Collect the transmission spectrum of the sample. T s Using bolt (9) as the axis, open the collection chamber, place the sample horizontally on the stage (6) of the collection chamber, and make its arc-shaped surface fit against the elliptical hole on the stage (6) to form a closed light-blocking environment. At the same time, align its equatorial part with the aperture (7) below, close the collection chamber, and press the control button (13). At this time, the spectral information of the cherry tomato is transmitted to the multispectral sensor (8) to complete the transmission spectrum of the sample. T s collection; Step f. Calculate the transmission correction spectrum of the sample. T c Based on the collected black reference spectrum T d White reference spectrum T w and original spectrum T s Using formula Calculate the transmission corrected spectrum of the sample. T c ; Set the corrected spectrum T c The wavelength range is 400-950 nm; Step g. Establishing an internal quality prediction model for samples: The transmission correction spectrum collected by the detection device is used as the input, and the physicochemical values ​​of key quality indicators of the samples obtained by traditional destructive testing methods such as sugar refractometer and hardness tester are used as the output; after the first derivative FD preprocessing method, partial least squares regression (PLSR) and principal component regression (PCR) quantitative prediction models are established. By comparing the correlation coefficient of the calibration set, the root mean square error of the calibration set (RMSEC), the correlation coefficient of the prediction set, the root mean square error of the prediction set (RMSEP), and the residual prediction deviation (RPD) evaluation index, the optimal internal quality prediction model for cherry tomatoes is the partial least squares regression (PLSR) model. Step h. Model deployment: Deploy the in-sample quality partial least squares regression prediction model onto the controller (14); Step i. Predict the internal quality index of the sample: Turn on the switch (18) to turn on the halogen lamp (3), turn on the control switch (11) to turn on the controller (14), collect the spectrum of the sample to be tested, and after correction by the formula, input it into the partial least squares regression (PLSR) prediction model to obtain the internal quality index of the sample.

6. The transmission spectroscopy nondestructive testing method according to claim 5, characterized in that: The sample is a cherry tomato. In step i, the internal quality indicators of the sample include the specific values ​​of soluble solids content (SSC) and hardness, and the results are displayed on the display screen (12).

Images