Nondestructive testing method and device and setting method and system of light source and detector

Abstract

The invention relates to the technical field of non-destructive testing, and discloses a method for setting a light source and a detector of a non-destructive testing method, which comprises the following steps of: 1, acquiring a plurality of citrus fruit samples and measuring to obtain optical characteristic parameters of three layers of tissues; 2, processing the optical characteristic parameters of the three-layer tissue to obtain optical transmission simulation parameters of the three-layer tissue; 3, constructing a three-layer tissue optical transmission model according to the optical characteristic parameters of the three-layer tissue and the optical transmission simulation parameters of the three-layer tissue; 4, obtaining a plurality of light source incident angles and a plurality of detector receiving angles to obtain a photon number image and a plurality of spectral images; and 5, analyzing the photon number image and the plurality of spectral images, and determining an optimal light source incident angle and an optimal detector receiving angle. According to the method, the signal intensity and quality of the near infrared spectrum can be improved. Meanwhile, the invention further discloses a system based on the method and a nondestructive testing method and device.

Description

Technical Field

[0001] The present invention relates to the technical field of non-destructive testing, and in particular to a non-destructive testing method, a device, and a method and system for setting a light source and a detector. Background Art

[0002] Pomelos, belonging to the Rutaceae family, are the largest citrus fruit. Rich in nutrients and delicious in flavor, they are a beloved traditional fruit in China. Their tissue can be divided into three layers: the oil cell layer, the spongy layer, and the pulpy segments. While visible / near-infrared spectroscopy is widely used for nondestructive testing of internal fruit quality, the thick-skinned nature of pomelos, coupled with their large size and significant spectral transmission attenuation, makes this testing inaccurate.

[0003] CN202210746484 discloses a nondestructive detection method for the soluble solids content of honey pomelo based on near-infrared spectroscopy detection information, comprising the following steps: S1. establishing a correspondence between the soluble solids content of honey pomelo fruit and the near-infrared spectrum of the fruit surface; S2. performing infrared spectroscopy detection on the surface of the honey pomelo fruit to be tested; S3. establishing a correspondence model between the soluble solids content of honey pomelo fruit and the near-infrared spectrum; and S4. using the model to calculate the soluble solids content of the honey pomelo fruit to be tested. This detection method determines the soluble solids content by establishing a relationship between the soluble solids content and the near-infrared spectrum of the fruit surface and then performing near-infrared spectroscopy analysis to obtain the soluble solids content. However, due to the significant transmission attenuation of the near-infrared spectrum in honey pomelo fruit, the detection accuracy may be reduced.

[0004] The technical problem to be solved by the present invention is: how to improve the near-infrared spectral signal intensity and the internal quality non-destructive detection effect. Summary of the Invention

[0005] The main purpose of the present invention is to provide a method for setting the light source and detector of a non-destructive testing method, adjust the incident angle of the light source and the receiving angle of the detector, determine the optimal light source incident angle and the optimal detector receiving angle, thereby improving the signal intensity of the near-infrared spectrum and further improving the non-destructive testing effect.

[0006] At the same time, a system based on the above method and a non-destructive testing method and device are also provided.

[0007] To achieve the above objectives, the technical solutions adopted in this application are:

[0008] A method for setting a light source and a detector for a non-destructive testing method comprises the following steps:

[0009] Step 1: Obtain multiple citrus fruit samples and measure them to obtain the optical property parameters of the three-layer tissue;

[0010] Step 2: Process the optical characteristic parameters of the three-layer tissue using the inverse amplification (IAD) algorithm to obtain the light transmission simulation parameters of the three-layer tissue;

[0011] Step 3: Construct a three-layer tissue light transmission model based on the optical characteristic parameters of the three-layer tissue and the light transmission simulation parameters of the three-layer tissue;

[0012] Step 4: Obtain multiple light source incident angles and detector receiving angles according to a preset method, input the multiple light source incident angles into the three-layer tissue light transmission model to obtain multiple spectral images; input the multiple detector receiving angles into the three-layer tissue light transmission model to obtain a photon number image;

[0013] Step 5: Analyze multiple spectral images based on the degree of light penetration and energy attenuation to determine the optimal light source incident angle; analyze the photon number image based on the number of photons to determine the optimal detector receiving angle.

[0014] Preferably, the three layers of tissue include the peel oil cell layer, the peel sponge layer, and the pulp petals; the optical characteristic parameters include reflectivity, thickness, spot diameter, total reflectivity, total transmittance, average refractive index, and anisotropy coefficient; the light transmission simulation parameters include absorption coefficient, scattering coefficient, reduced scattering coefficient, total reflection flux, and total transmission flux.

[0015] Preferably, step 1 specifically comprises: obtaining a plurality of citrus fruit samples, measuring the optical characteristic parameters of the plurality of citrus fruit samples at a wavelength of 400 to 1100 nm using a single integrating sphere, and taking the average value to obtain the optical characteristic parameters of the three-layer tissue.

[0016] Preferably, step 3 is specifically as follows: simulating a citrus fruit model based on the optical characteristic parameters of the three-layer tissue; simulating the photon flight distance and probability by the Monte Carlo photon tracing method based on the light transmission simulation parameters of the three-layer tissue; constructing a light source model and a detector model; constructing a three-layer tissue light transmission model based on the citrus fruit model, the light source model, the detector model, and the photon flight distance and probability.

[0017] Preferably, in step 4, the preset method is the equal increment method; the range of the incident angle of the light source is 0 to 90°; the range of the receiving angle of the detector is 0 to 40°; the light source and the detector are respectively arranged on both sides of the normal.

[0018] Preferably, the method further includes step 6: setting up a light transmission test platform according to multiple light source incident angles and detector receiving angles, and conducting experiments on the light transmission test platform to verify the accuracy of the three-layer tissue light transmission model.

[0019] At the same time, a system based on the above method is also provided, comprising the following units:

[0020] Sample measurement unit: used to obtain multiple citrus fruit samples and perform measurements to obtain the optical characteristic parameters of the three-layer tissue;

[0021] A simulation parameter acquisition unit is used to process the optical characteristic parameters of the three-layer tissue using an inverse amplification (IAD) algorithm to obtain light transmission simulation parameters of the three-layer tissue;

[0022] Model building unit: used to process the optical characteristic parameters of the three-layer tissue through the inverse multiplication (IAD) algorithm to obtain the light transmission simulation parameters of the three-layer tissue;

[0023] Image acquisition unit: used to obtain multiple light source incident angles and detector receiving angles according to a preset method, input the multiple light source incident angles into the three-layer tissue light transmission model respectively to obtain multiple spectral images; input the multiple detector receiving angles into the three-layer tissue light transmission model to obtain a photon number image;

[0024] Image analysis unit: used to analyze multiple spectral images according to the degree of light penetration and energy attenuation to determine the optimal light source incident angle; analyze the photon number image according to the number of photons to determine the optimal detector receiving angle.

[0025] A non-destructive testing method is also provided, comprising the following steps:

[0026] Step 1: Obtain the optimal light source incident angle and the optimal detector receiving angle;

[0027] Step 2: Set the light source and detector according to the optimal light source incident angle and the optimal detector receiving angle respectively;

[0028] Step 3: Perform near-infrared spectroscopy on the citrus fruits to be tested to obtain near-infrared spectral data;

[0029] Step 4: Process the near-infrared spectral data to obtain the soluble solids content of citrus fruits;

[0030] The optimal light source incident angle is an angle of 30 to 40 degrees with the normal; the optimal detector receiving angle is an angle of 5 to 15 degrees with the normal; the light source and the detector are respectively arranged on both sides of the normal.

[0031] Preferably, the light source is composed of 6 to 8 halogen lamps; the power of the halogen lamp is 130 to 170 W; the distance between the light source and the citrus fruit is 1 to 3 cm; the distance between the detector and the citrus fruit is 0.1 to 1 cm.

[0032] Preferably, step 4 specifically comprises: performing SG smoothing, SNV standard normal transformation, and CARS feature extraction processing on the near-infrared spectral data, and then inputting the data into a trained partial least squares model to obtain the soluble solids content of citrus fruits.

[0033] At the same time, a non-destructive testing device is also provided for implementing the above-mentioned non-destructive testing method for citrus fruits, comprising: a detector, a support platform and a light source; the support platform is arranged between the detector and the light source;

[0034] The light source includes a reflective cup and a plurality of bulbs; the plurality of bulbs are arranged in the reflective cup in an equidistant manner.

[0035] Preferably, the reflective cup is made of aluminum alloy; the bulb is a 150W halogen lamp; the number of bulbs is 7; the distance between the light source and the support platform is 1 to 3 cm; the distance between the detector and the support platform is 0.1 to 1 cm; the incident angle of the light source is 30 to 40° with the normal; the receiving angle of the detector is 5 to 15° with the normal; the light source and the detector are respectively arranged on both sides of the normal.

[0036] Compared with the existing technology, this solution has the following beneficial effects:

[0037] The optimization method of this scheme adjusts the angle of the light source to the optimal light source incident angle to enhance the transmission of near-infrared light in citrus fruits and reduce the attenuation of the near-infrared spectrum. Then, the angle of the detector is adjusted to the optimal detector receiving angle to increase the number of photons received by the detector. The more photons there are, the stronger the near-infrared spectrum signal. Combined with the optimal light source incident angle and the optimal detector receiving angle, the signal strength and quality of the near-infrared spectrum are comprehensively improved, thereby improving the accuracy of non-destructive testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a flow chart of the optimization method of Example 1;

[0039] Figure 2 This is a structural diagram of the three-layer structure of pomelo in Example 1;

[0040] Figure 3 This is a flight distance-probability diagram of the photons transmitted in the peel oil cell layer in Example 1;

[0041] Figure 4 This is a flight distance-probability diagram of the photons transmitted in the peel sponge layer of Example 1;

[0042] Figure 5 The flight distance-probability diagram of the photons transmitted in the pulp petals in Example 1;

[0043] Figure 6Schematic diagram of the three-layer tissue light transmission model of Example 1;

[0044] Figure 7 The spectrum image of Example 1 is simulated with the incident angle of the light source being 0°;

[0045] Figure 8 This is a spectrum image simulated when the incident angle of the light source in Example 1 is 18°;

[0046] Figure 9 This is a spectrum image simulated when the incident angle of the light source in Example 1 is 36°;

[0047] Figure 10 This is a spectrum image simulated when the incident angle of the light source in Example 1 is 54°;

[0048] Figure 11 This is a spectrum image simulated when the incident angle of the light source in Example 1 is 72°;

[0049] Figure 12 The photon number image of Example 1 simulated at different detector receiving angles;

[0050] Figure 13 Schematic diagram of the incident angle of the light source and the receiving angle of the detector in Examples 1, 3, and 4;

[0051] Figure 14 It is a front view of the equipment used in Examples 1, 3, and 4;

[0052] Figure 15 This is the energy intensity image of the experiment with different light source incident angles in Example 1;

[0053] Figure 16 Energy intensity images of different detector receiving angle tests in Example 1;

[0054] Figure 17 is a block diagram of the optimization system of Example 2;

[0055] Figure 18 This is a flow chart of the nondestructive testing method of Example 3;

[0056] Figure 19 This is the result diagram of Example 3 with the light source incident angle of 0° and the detector receiving angle of 0°;

[0057] Figure 20 The result diagram of Example 3 is that the incident angle of the light source is 18° and the receiving angle of the detector is rotated 10° along the Z axis;

[0058] Figure 21 The result diagram of Example 3 is that the incident angle of the light source is 36° and the receiving angle of the detector is rotated 10° along the Z axis;

[0059] Figure 22 The result diagram of Example 3 is that the incident angle of the light source is 36° and the receiving angle of the detector is rotated 10° along the Y axis;

[0060] Figure 23 This is a structural diagram of the light source of Example 4. DETAILED DESCRIPTION

[0061] To make the objectives, technical solutions, and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Generally, the components of the present application described and shown in the drawings herein can be arranged and designed in various different configurations.

[0062] Example 1

[0063] refer to Figure 1-19 A method for setting a light source and a detector for a non-destructive testing method comprises the following steps:

[0064] Step 1: Obtain multiple citrus fruit samples and measure them to obtain the optical property parameters of the three-layer tissue;

[0065] Preferably, the three layers of tissue include the peel oil cell layer, the peel sponge layer, and the pulp petals; the optical characteristic parameters include reflectivity, thickness, spot diameter, total reflectivity, total transmittance, average refractive index, and anisotropy coefficient; the light transmission simulation parameters include absorption coefficient, scattering coefficient, reduced scattering coefficient, total reflection flux, and total transmission flux.

[0066] Preferably, step 1 specifically comprises: obtaining a plurality of citrus fruit samples, measuring the optical characteristic parameters of the plurality of citrus fruit samples at a wavelength of 400 to 1100 nm using a single integrating sphere, and taking the average value to obtain the optical characteristic parameters of the three-layer tissue.

[0067] In this embodiment, the citrus fruit is pomelo, and 30 red-fleshed honey pomeloes are picked from the orchard. The three layers of tissue of the 30 red-fleshed honey pomeloes are peeled off, namely, the peel oil layer, the peel sponge layer, and the flesh segment. Figure 2As shown. Using a single integrating sphere acquisition method, a reflectivity detector (model JY-F03, manufactured by Guangzhou Jingyi Optoelectronics Technology Co., Ltd.) was used to measure the reflectivity of the peel oil coating, peel sponge layer, and pulp petals at wavelengths of 400 to 1100. A universal transmittance detector (model JY-T03, manufactured by Guangzhou Jingyi Optoelectronics Technology Co., Ltd.) was used to measure the total transmittance. A vernier caliper was used to measure the thickness of the peel oil coating, peel sponge layer, and pulp petals. The data measured for 30 red-fleshed pomeloes were averaged, and the reflectivities of the peel oil cell layer, peel sponge layer, and pulp segments were 0.51, 0.66, and 0.70, respectively. The thicknesses of the peel oil cell layer, peel sponge layer, and pulp segments were 2.00 mm, 24.00 mm, and 90.00 mm, respectively. The spot diameter was 2 mm, the total reflectivity was 0.53, and the total transmittance was 0.14. It can be seen from the "Optical Path Simulation and Experiment of Spectral Nondestructive Testing of Pomelo Internal Quality Based on ZEMAX" that the average reflectivities of the peel oil cell layer, peel sponge layer, and pulp segments were 1.80, 1.65, and 1.40, respectively, and the anisotropy coefficients were 0.63, 0.63, and 0.60, respectively.

[0068] Step 2: Process the optical characteristic parameters of the three-layer tissue using the inverse amplification (IAD) algorithm to obtain the light transmission simulation parameters of the three-layer tissue;

[0069] The measured optical characteristic parameters are processed by the inverse amplification (IAD) algorithm to calculate the absorption coefficient, scattering coefficient, reduced scattering coefficient, total reflection flux and total transmission flux. The specific data are shown in Table 1:

[0070] Table 1 Optical transmission simulation parameters

[0071] name Absorption coefficient Scattering coefficient Reduced scattering coefficient Total reflected flux Total transmitted flux Peel oil cell layer 0.49 0.51 0.19 0.07 0.40 Peel sponge layer 0.06 0.11 0.04 0.12 0.40 pulp petals 0.01 0.02 0.01 0.10 0.46

[0072] Step 3: Construct a three-layer tissue light transmission model based on the optical characteristic parameters of the three-layer tissue and the light transmission simulation parameters of the three-layer tissue;

[0073] Preferably, step 3 is specifically as follows: simulating a citrus fruit model based on the optical characteristic parameters of the three-layer tissue; simulating the photon flight distance and probability by the Monte Carlo photon tracing method based on the light transmission simulation parameters of the three-layer tissue; constructing a light source model and a detector model; constructing a three-layer tissue light transmission model based on the citrus fruit model, the light source model, the detector model, and the photon flight distance and probability.

[0074] In this embodiment, MCX software was used to model the grapefruit, light source, and detector. The optical characteristic parameters of the three layers of tissue were input to construct a citrus fruit model. The light source model was set to a Gaussian light source with a weight of 5 and a radius of 20 mm. The radius of the detector model was set to 6 mm. The center of the detector model was kept at the same height as the center of the light source model. The light gate T0 = 0 and the acquisition time D t = 200ms, used for time domain photon migration simulation. The initialization parameters are set to (0, 0, 1, 1), the random seed for light transport simulation is 1648335518 and the number of photons is 10 8 .

[0075] Using MATLAB software, the photon flight distance and probability were obtained through the multi-layer tissue steady-state light MC transmission program. The principle of the MC transmission program is: the external part of the fruit tissue is reflected by the air,

[0076] The relevant formula is:

[0077]

[0078] Among them, R sp is the mirror reflectivity, n1 is the refractive index of air, and n2 is the refractive index of each layer of grapefruit tissue;

[0079]

[0080] Among them, Δs is the photon motion step length, μ α is the absorption coefficient, μ s is the scattering coefficient, ξ is a random number in the range [0,1];

[0081]

[0082] Among them, the weight ω decays every time the photon moves, and when the threshold is lower than 0.00001, the photon movement stops.

[0083] Input the light transmission simulation parameters into the multilayer tissue steady-state light MC transmission program and set the number of outgoing photons to 10 4 , the flight distance and probability of photons in the oil cell layer of the peel are obtained by Monte Carlo photon tracing method, as shown in Figure 3 As shown, the flight distance and probability of the peel sponge layer are as follows Figure 4 As shown, the flight distance and probability in the pulp petal are as follows Figure 5 It is calculated that at 100 mm of tissue, the flight distance of photons in the oil cell layer of the peel is 100 to 300 cm, with a probability of about 0 to 0.004, the flight distance of photons in the spongy layer of the peel is 8 to 83 cm, with a probability of about 0.0025 to 0.14, and the flight distance of photons in the pulp petals is 0 to 33 cm, with a probability of about 0.05 to 0.48.

[0084] According to the set citrus fruit model, light source model, detector model, and photon flight distance and probability, a three-layer tissue light transmission model is constructed to simulate the light transmission of grapefruit, such as Figure 6 shown.

[0085] Step 4: Obtain multiple light source incident angles and detector receiving angles according to a preset method, input the multiple light source incident angles into the three-layer tissue light transmission model to obtain multiple spectral images; input the multiple detector receiving angles into the three-layer tissue light transmission model to obtain a photon number image;

[0086] Preferably, in step 4, the preset method is the equal increment method; the range of the light source incident angle is 0 to 90°; the range of the light source incident angle is 0 to 90°; the range of the detector receiving angle is 0 to 40°; the light source and the detector are respectively arranged on both sides of the normal.

[0087] In this embodiment, the incident angles of the light source are set to 0°, 18°, 36°, 54°, and 72° with the normal line by the equal increment method, and the receiving angle of the detector is 0°. The light source model and the detector model are adjusted and simulated to obtain a spectral image with a light source incident angle of 0°, as shown in FIG. Figure 7 As shown in the figure, the spectrum image of the light source with an incident angle of 18° is as follows: Figure 8 As shown in the figure, the spectrum image with the incident angle of the light source being 36° is as follows: Figure 9 As shown in the figure, the spectrum image of the light source with an incident angle of 54° is as follows: Figure 10 As shown in the figure, the spectrum image of the light source with an incident angle of 72° is as follows: Figure 11 shown.

[0088] By using the equal increment method, the detector is set to rotate 0°, 10°, 20°, 30°, and 40° along the Z axis and 0°, 10°, 20°, 30°, and 40° along the Y axis. The incident angle of the light source is 36° and input into the three-layer tissue light transmission model to obtain the photon number image, as shown in the figure. Figure 12 shown.

[0089] It should be noted that the normal is parallel to the horizontal plane, the light source is irradiated obliquely downward on the grapefruit, the detector rotates along the Z axis, that is, it rotates upward by 0°, 10°, 20°, 30°, and 40°, and the detector rotates along the Y axis, that is, it rotates on the horizontal plane by 0°, 10°, 20°, 30°, and 40°. The positions of the light source incident angle (α) and the detector acceptance angle (β) are as follows Figure 13 shown.

[0090] Step 5: Analyze multiple spectral images based on the degree of light penetration and energy attenuation to determine the optimal light source incident angle; analyze the photon number image based on the number of photons to determine the optimal detector receiving angle.

[0091] In this embodiment, Figure 7-11 The spectral image is analyzed. Figure 10 and Figure 11 In the image, the middle part has higher energy, but it does not penetrate the entire pomelo pulp, which may cause the spectral information to be missed. Figure 7 and Figure 8 Although it penetrates the complete flesh petal, it is severely attenuated in the peel. Figure 9 The energy density is highest in the flesh, and the energy attenuation is minimal after penetrating the oil cell layer of the peel. The short longitudinal distance through the spongy layer of the peel means that the energy attenuation is relatively small, so the optimal light source incident angle is 36°.

[0092] The photon count graph shows that the detector receives the most photons when it is rotated 10° along the Z axis. The higher the number of photons, the stronger the spectral signal. Therefore, the optimal detector receiving angle is a 10° angle with the normal.

[0093] Preferably, the method further includes step 6: setting up a light transmission test platform according to multiple light source incident angles and detector receiving angles, and conducting experiments on the light transmission test platform to verify the accuracy of the three-layer tissue light transmission model.

[0094] In this embodiment, the structure of the optical transmission test platform is as follows: Figure 14 As shown, the light source 1 consists of seven 150W halogen lamps and an aluminum reflector. The light source is rotated a certain angle and then locked. The detector 3 is slid in the hemispherical groove to change its rotation angle and then locked. The optical fiber of the detector is replaced with a fiber-optic probe that penetrates the pomelo fruit tissue to complete the light transmission experiment. A QEpro spectrometer collects near-infrared spectral data. Before acquisition, the instrument is preheated for 10 minutes. After calibration with a black and white plate, the experiment begins. The wavelength range is 400-1100 nm, the integration time is 1000 ms, and the number of averages is 1.

[0095] The fiber optic probe was used to puncture the three layers of pomelo fruit tissue with a thickness of 0 to 60 mm and the incident angle of the light source was adjusted (0°, 18°, 36°, 54° and 72° respectively) to obtain energy intensity images at different light source incident angles, such as Figure 15 As shown in Figure 3, the experimental results are basically consistent with the results of the three-layer tissue light transmission model.

[0096] Keeping the incident angle of the light source unchanged, adjust the rotation angle of the detector to 0°, 10°, 20°, 30°, 40° along the Z axis and 0°, 10°, 20°, 30°, 40° along the Y axis to obtain the energy intensity image, such as Figure 16 The experimental results are basically consistent with the results of the three-layer tissue light transmission model.

[0097] Example 2

[0098] refer to Figure 17 , a system based on the above method includes the following units:

[0099] Sample measurement unit: used to obtain multiple citrus fruit samples and perform measurements to obtain the optical characteristic parameters of the three-layer tissue;

[0100] A simulation parameter acquisition unit is used to process the optical characteristic parameters of the three-layer tissue using an inverse amplification (IAD) algorithm to obtain light transmission simulation parameters of the three-layer tissue;

[0101] Model building unit: used to process the optical characteristic parameters of the three-layer tissue through the inverse multiplication (IAD) algorithm to obtain the light transmission simulation parameters of the three-layer tissue;

[0102] Image acquisition unit: used to obtain multiple light source incident angles and detector receiving angles according to a preset method, input the multiple light source incident angles into the three-layer tissue light transmission model respectively to obtain multiple spectral images; input the multiple detector receiving angles into the three-layer tissue light transmission model to obtain a photon number image;

[0103] Image analysis unit: used to analyze multiple spectral images according to the degree of light penetration and energy attenuation to determine the optimal light source incident angle; analyze the photon number image according to the number of photons to determine the optimal detector receiving angle.

[0104] In this embodiment, the specific workflow of the optimization system is as follows: the sample measurement unit measures the optical characteristic parameters of the peel oil cell layer, peel sponge layer, and pulp segments of multiple citrus fruit samples through corresponding detection equipment and takes the average value to obtain the optical characteristic parameters of the three-layer tissue, and then inputs the optical characteristic parameters of the three-layer tissue into the simulation parameter acquisition unit.

[0105] After receiving the optical characteristic parameters of the three-layer tissue, the simulation parameter acquisition unit processes the optical characteristic parameters of the three-layer tissue through an inverse amplification (IAD) algorithm to obtain the light transmission simulation parameters of the three-layer tissue, and then inputs the light transmission simulation parameters of the three-layer tissue into the model construction unit.

[0106] The model building unit uses MCX software to model the grapefruit, light source, and detector. The pulp of the citrus fruit is represented by a sphere, and the peel oil cell layer and the peel sponge layer are composed of spherical shells of different thicknesses. The optical characteristic parameters of the three layers of tissue are input to construct the citrus fruit model. The light source model is set to a Gaussian light source with a weight of 5 and a radius of 20 mm. The radius of the detector model is set to 6 mm. The center of the detector model is kept at the same height as the center of the light source model. The light gate T0 = 0 and the acquisition time D t = 200ms, used for time domain photon migration simulation. The initialization parameters are set to (0, 0, 1, 1), the random seed for light transport simulation is 1648335518 and the number of photons is 10 8 Then input the light transmission simulation parameters into the multilayer tissue steady-state light MC transmission program and set the number of emitted photons to 10. 4 Through the Monte Carlo photon tracing method, the flight distance and probability of photons in the oil cell layer of the peel are obtained. Then, based on the citrus fruit model, light source model, detector model and photon flight distance and probability, a three-layer tissue light transmission model is constructed and input into the image acquisition unit.

[0107] The image acquisition unit sets different light source incident angles and detector receiving angles through the equal increment method, and inputs different light source incident angles and detector receiving angles into the three-layer tissue light transmission model input through the control variable method to obtain photon number images and multiple spectral images, and input them into the image analysis unit.

[0108] The image analysis unit analyzes multiple spectral images based on the principle that higher light penetration and lower energy attenuation indicate better spectral image quality, obtaining the best-quality spectral image and determining the optimal light source incident angle corresponding to the best-quality spectral image. Based on the principle that a higher number of photons indicates a stronger spectral signal, the detector receiving angle corresponding to the highest number of photons in the photon number image is determined, thereby determining the optimal detector receiving angle.

[0109] Example 3

[0110] refer to Figure 18-22 , a nondestructive testing method comprising the following steps:

[0111] Step 1: Obtain the optimal light source incident angle and the optimal detector receiving angle;

[0112] The optimal light source incident angle and the optimal detector receiving angle are obtained by the optimization method of Example 1. The optimal light source incident angle is an angle of 36° with the normal; the optimal detector receiving angle is an angle of 10° with the normal.

[0113] Step 2: Set the light source 1 and detector 3 according to the optimal light source incident angle and the optimal detector receiving angle respectively;

[0114] In this embodiment, the light source 1 and the detector 3 in the nondestructive testing equipment are set. The parameters of the light source 1 are set as follows: the incident angle of the light source is 36° with the normal line, the light source 1 is composed of 7 halogen lamps, the power of the halogen lamps is 150W, and the distance between the light source 1 and the citrus fruit is 2 cm;

[0115] The parameters of the detector 3 are set as follows: the rotation angle is 10° with the normal line, and the distance between the detector 3 and the citrus fruit is 0.5 cm.

[0116] Step 3: Perform near-infrared spectroscopy on the citrus fruits to be tested to obtain near-infrared spectral data;

[0117] In this embodiment, the citrus fruits to be tested are placed in an adjusted non-destructive testing device for testing to obtain near-infrared spectrum data.

[0118] Step 4: Process the near-infrared spectral data to obtain the soluble solids content of citrus fruits;

[0119] Preferably, step 4 specifically comprises: performing SG smoothing, SNV standard normal transformation, and CARS feature extraction processing on the near-infrared spectral data, and then inputting the data into a trained partial least squares model to obtain the soluble solids content of citrus fruits.

[0120] In this embodiment, the partial least squares model is trained as follows: near-infrared spectral data of 30 grapefruits are collected, the soluble solids content of the pulp of the 30 grapefruits is measured using a digital optical refractometer, the near-infrared spectral data is subjected to SG smoothing, SNV standard normal transformation, and CARS feature extraction processing, and then modeling is performed using the partial least squares method. The near-infrared spectral data and the soluble solids content are divided into a training set and a prediction set, with the ratio of the training set to the prediction set being 3:1, to achieve model training.

[0121] After obtaining the near-infrared spectral data of the citrus fruit to be tested, the near-infrared spectral data is processed by SG smoothing, SNV standard normal transformation, and CARS feature extraction, and then input into the trained partial least squares model to obtain the soluble solids content of the citrus fruit.

[0122] In order to further illustrate that the optimal light source incident angle and the optimal detector receiving angle can improve the detection accuracy, the near-infrared spectral data of 30 grapefruits with different light source incident angles and detector receiving angles and the corresponding grapefruit pulp soluble solids content were collected. The near-infrared spectral data with different light source incident angles and detector receiving angles and the pulp soluble solids content were used to form a training set and a prediction set respectively. The R 2 (coefficient of determination) and RMSEC (root mean square error of calibration) and R of the prediction set 2 (coefficient of determination) and RMSEP (root mean square error of prediction), R 2 The higher the value, the lower the RMSEC and RMSEP values, which indicates that the detection accuracy is higher.

[0123] Among them, the incident angle of the light source is 0°, the receiving angle of the detector is 0°, and the R 2 and RMSEC, R of the prediction set 2 The results of RMSEP are as follows Figure 19 As shown in the figure, the incident angle of the light source is 18°, the detector receiving angle is rotated 10° along the Z axis, and the R 2 and RMSEC, R of the prediction set 2 The results of RMSEP are as follows Figure 20 As shown in the figure, the incident angle of the light source is 36°, the detector receiving angle rotates 10° around the Z axis, and the R 2 and RMSEC, R of the prediction set 2 The results of RMSEP are as follows Figure 21 As shown in the figure, the incident angle of the light source is 36°, the detector receiving angle is rotated 10° along the Y axis, and the R 2 and RMSEC, R of the prediction set 2 The results of RMSEP are as follows Figure 22 As shown; it can be obtained that when the incident angle of the light source is 36° and the detector receiving angle rotates 10° along the Z axis, the R 2 and RMSEC, R of the prediction set 2 The values ​​of RMSEP and training set are the highest. 2 and RMSEC were 0.89 and 0.25 respectively, and the R 2 and RMSEP are 0.84 and 0.38 respectively, which proves that the detection accuracy can be improved by setting the optimal light source incident angle and the optimal detector receiving angle.

[0124] Example 4

[0125] refer to Figure 13 、 Figure 23A nondestructive testing device for citrus fruits, used to implement the above-mentioned nondestructive testing method for citrus fruits, comprises: a detector 3, a support platform 2 and a light source 1; the support platform 2 is arranged between the detector 3 and the light source 1;

[0126] The light source 1 includes a reflective cup 11 and a plurality of bulbs 12 ; the plurality of bulbs 12 are arranged in an equidistant manner in the reflective cup 11 .

[0127] Preferably, the reflective cup 11 is made of aluminum alloy; the bulb 12 is a 150W halogen lamp; the number of the bulbs 12 is 7; the distance between the light source 1 and the support platform 2 is 1 to 3 cm; the distance between the detector 3 and the support platform 2 is 0.1 to 1 cm; the incident angle of the light source 1 is 30 to 40° with the normal; the rotation angle of the detector 3 is 5 to 15° with the normal; the light source and the detector are respectively arranged on both sides of the normal.

[0128] In this embodiment, the light source 1 comprises an aluminum reflector 11 and seven bulbs 12. The seven bulbs 12 are arranged equidistantly within the reflector 11; in other words, the distance between adjacent bulbs 12 is equal. The normal of the light source 1 to the citrus fruit is parallel to the horizontal plane, and the incident angle of the light source 1 on the citrus fruit is 36° with the normal (rotated 36° downward). The distance between the light source 1 and the support platform 2 is 2 cm. The detector 3 is configured so that the receiving angle of the detector 3 is 10° with the normal (rotated 10° upward), and the distance between the detector 3 and the support platform 2 is 0.1 cm. Citrus fruit is placed on the support platform 2, and the light source 1 is turned on to illuminate the citrus fruit. The detector 3 receives the light signal transmitted through the citrus fruit and converts the light signal into an electrical signal, which is transmitted to the QEpro spectrometer to collect the near-infrared spectrum, which is then analyzed and detected using a partial least squares model.

[0129] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

Claims

1. A method for setting a light source and a detector for a non-destructive testing method, characterized in that: The steps include: Step 1: Obtain multiple citrus fruit samples and measure them to obtain the optical property parameters of the three-layer tissue; Step 2: Process the optical characteristic parameters of the three-layer tissue using the inverse amplification (IAD) algorithm to obtain the light transmission simulation parameters of the three-layer tissue; Step 3: Construct a three-layer tissue light transmission model based on the optical characteristic parameters of the three-layer tissue and the light transmission simulation parameters of the three-layer tissue; Step 4: Obtain multiple light source incident angles and detector receiving angles according to a preset method, input the multiple light source incident angles into the three-layer tissue light transmission model to obtain multiple spectral images; input the multiple detector receiving angles into the three-layer tissue light transmission model to obtain a photon number image; Step 5: Analyze multiple spectral images based on the degree of light penetration and energy attenuation to determine the optimal light source incident angle; analyze the photon number image based on the number of photons to determine the optimal detector receiving angle.

2. The method for setting up a light source and a detector for a nondestructive testing method according to claim 1, characterized in that: The three layers of tissue include the peel oil cell layer, the peel sponge layer, and the pulp petals; the optical characteristic parameters include reflectivity, thickness, spot diameter, total reflectivity, total transmittance, average refractive index, and anisotropy coefficient; the light transmission simulation parameters include absorption coefficient, scattering coefficient, reduced scattering coefficient, total reflection flux, and total transmission flux.

3. The method for setting up a light source and a detector for a nondestructive testing method according to claim 1, wherein: The step 1 specifically comprises: obtaining a plurality of citrus fruit samples, measuring the optical characteristic parameters of the plurality of citrus fruit samples at wavelengths of 400 to 1100 nm using a single integrating sphere, and taking an average value to obtain the optical characteristic parameters of the three-layer tissue.

4. The method for setting up a light source and a detector for a nondestructive testing method according to claim 1, wherein: Step 3 specifically comprises: simulating a citrus fruit model based on the optical characteristic parameters of the three-layer tissue; simulating the photon flight distance and probability using the Monte Carlo photon tracing method based on the light transmission simulation parameters of the three-layer tissue; constructing a light source model and a detector model; and constructing a three-layer tissue light transmission model based on the citrus fruit model, the light source model, the detector model, and the photon flight distance and probability.

5. The method for setting up a light source and a detector for a nondestructive testing method according to claim 1, wherein: In step 4, the preset method is the equal increment method; the range of the incident angle of the light source is 0 to 90°; the range of the receiving angle of the detector is 0 to 40°; the light source and the detector are respectively arranged on both sides of the normal.

6. The method for setting up a light source and a detector for a nondestructive testing method according to claim 1, wherein: The method further includes step 6: setting up a light transmission test platform according to multiple light source incident angles and detector receiving angles, and conducting experiments on the light transmission test platform to verify the accuracy of the three-layer tissue light transmission model.

7. A system according to the method of claims 1 to 6, characterized in that The following units are included: Sample measurement unit: used to obtain multiple citrus fruit samples and perform measurements to obtain the optical characteristic parameters of the three-layer tissue; A simulation parameter acquisition unit is used to process the optical characteristic parameters of the three-layer tissue using an inverse amplification (IAD) algorithm to obtain light transmission simulation parameters of the three-layer tissue; Model building unit: used to process the optical characteristic parameters of the three-layer tissue through the inverse multiplication (IAD) algorithm to obtain the light transmission simulation parameters of the three-layer tissue; Image acquisition unit: used to obtain multiple light source incident angles and detector receiving angles according to a preset method, input the multiple light source incident angles into the three-layer tissue light transmission model respectively to obtain multiple spectral images; input the multiple detector receiving angles into the three-layer tissue light transmission model to obtain a photon number image; Image analysis unit: used to analyze multiple spectral images according to the degree of light penetration and energy attenuation to determine the optimal light source incident angle; analyze the photon number image according to the number of photons to determine the optimal detector receiving angle.

8. A non-destructive testing method, characterized in that: The following steps are involved: Step 1: Obtain the optimal light source incident angle and the optimal detector receiving angle; Step 2: Set the light source and detector according to the optimal light source incident angle and the optimal detector receiving angle respectively; Step 3: Perform near-infrared spectroscopy on the citrus fruits to be tested to obtain near-infrared spectral data; Step 4: Process the near-infrared spectral data to obtain the soluble solids content of citrus fruits; The optimal light source incident angle is an angle of 30 to 40 degrees with the normal; the optimal detector receiving angle is an angle of 5 to 15 degrees with the normal; the light source and the detector are respectively arranged on both sides of the normal.

9. The fruit non-destructive testing method according to claim 8, characterized in that: The light source consists of 6 to 8 halogen lamps; the power of the halogen lamps is 130 to 170 W; the distance between the light source and the citrus fruits is 1 to 3 cm; the distance between the detector and the citrus fruits is 0.1 to 1 cm.

10. The fruit non-destructive testing method according to claim 8, characterized in that: The step 4 specifically includes: performing SG smoothing, SNV standard normal transformation, and CARS feature extraction on the near-infrared spectral data, and then inputting the data into a trained partial least squares model to obtain the soluble solids content of citrus fruits.

11. A non-destructive testing device, characterized in that: A method for nondestructive testing of citrus fruits according to any one of claims 8 to 10, comprising: a detector, a support platform, and a light source; the support platform is arranged between the detector and the light source; The light source includes a reflective cup and a plurality of bulbs; the plurality of bulbs are arranged in the reflective cup in an equidistant manner.

12. The nondestructive testing device according to claim 11, characterized in that: The reflective cup is made of aluminum alloy; the bulb is a 150W halogen lamp; the number of bulbs is 7; the distance between the light source and the support platform is 1 to 3 cm; the distance between the detector and the support platform is 0.1 to 1 cm; the incident angle of the light source is 30 to 40 degrees with the normal; the receiving angle of the detector is 5 to 15 degrees with the normal; the light source and the detector are respectively arranged on both sides of the normal.

Images