A cherry sugar content nondestructive testing device based on multispectral imaging technology and application

Abstract

The application discloses a cherry sugar content nondestructive detection equipment based on a multispectral imaging technology and application, which comprises a detection table, a conveying mechanism, dark box modules, a tray, a rotating mechanism and a lifting mechanism.

Description

Technical Field

[0001] This invention relates to the field of non-destructive testing technology for agricultural products, specifically to a non-destructive testing device and application for cherry sugar content based on multispectral imaging technology. Background Technology

[0002] As a high-value fruit, the internal quality of cherries (especially sugar content and soluble solids content) is a key indicator that determines their market value and consumer preference.

[0003] Traditional methods for sugar content testing mainly rely on handheld refractometers for puncture measurement. This method has drawbacks such as being destructive, inefficient, and having poor sample representativeness, and cannot meet the needs of modern orchard precision management and automated sorting lines.

[0004] Near-infrared spectroscopy has been used for non-destructive testing of sugar content in fruits, but it is mostly used for single-point or small-area measurements. For fruits like cherries, which are small in size and may have uneven sugar content distribution, single-point measurements are not representative enough and have limited throughput. Multispectral imaging technology can simultaneously acquire spatial information and spectral information of multiple characteristic wavelengths of a sample, offering a potential solution to this problem.

[0005] However, existing fruit inspection equipment based on multispectral imaging typically suffers from the following problems: insufficient correction for diffuse reflectance intensity differences caused by the curved shape of cherries, affecting model accuracy; lack of a mechanism for simultaneously acquiring and fusing analysis with cherry physiological characteristics (such as size and color), limiting the accuracy of the prediction model; and low system integration, making it difficult to directly embed into production lines for real-time online inspection. Therefore, there is an urgent need to design a non-destructive testing device and application for cherry sugar content based on multispectral imaging technology to solve these problems. Summary of the Invention

[0006] The purpose of this invention is to provide a non-destructive testing device and application for cherry sugar content based on multispectral imaging technology, so as to solve the above-mentioned shortcomings in the prior art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A non-destructive testing device for cherry sugar content based on multispectral imaging technology includes a testing stage.

[0009] A conveying mechanism is installed on both sides of the top of the testing station for conveying cherry samples;

[0010] A dark box module is fixed above the detection station, and the conveying mechanism passes through its interior. The dark box module is used to provide a controllable and uniform detection light environment and shield external stray light interference.

[0011] A tray, in which cherry samples are placed, is positioned above the conveyor mechanism;

[0012] A rotating mechanism is disposed below the detection table and is used to rotate the tray without contact.

[0013] A lifting mechanism is installed on top of the dark box module;

[0014] An elastic transmission component, one end of which is connected to a lifting mechanism, the lifting mechanism being used to drive the elastic transmission component to move up and down in multiple stages;

[0015] A door assembly is disposed on both sides of the dark box module and is used to open and close the openings on both sides of the dark box module. The lifting mechanism drives the door assembly to move up and down through an elastic transmission component.

[0016] An active lighting module is located inside the dark box module and is fixedly installed at the bottom center of the elastic transmission component. When the elastic transmission component moves up and down, it drives the active lighting module to move up and down again.

[0017] The detectors are multiple and installed in multiple locations inside the dark box module, and are used to detect cherry samples on a tray inside the dark box module.

[0018] As a preferred technical solution provided by the present invention, the conveying mechanism includes a frame fixed to the bottom of the testing table, support arms fixedly installed on both sides of the frame, a power roller provided at the top of the support arm, a conveyor belt provided on the two power rollers, and baffles located on both sides of the conveyor belt fixedly installed on both sides of the testing table.

[0019] As a preferred technical solution provided by the present invention, the dark box module is a closed structure with openings on both sides, and it is made of light-blocking material.

[0020] As a preferred technical solution provided by the present invention, the tray includes a substrate, a transparent cover is fixedly installed on the top periphery of the substrate, and the top of the transparent cover is an open structure. A mounting groove is provided at the bottom center of the substrate, and a magnetic block is fixed inside the mounting groove.

[0021] As a preferred technical solution provided by the present invention, the rotating mechanism includes a drive motor fixedly installed on the outer wall of the bottom of the testing platform, and the output end of the drive motor is connected to a transmission box. An energized coil is connected to the upper part of the transmission box, and the upper surface of the energized coil is flush with the upper surface of the testing platform.

[0022] As a preferred technical solution provided by the present invention, the lifting mechanism includes an electric cylinder fixedly installed on one side of the top of the darkroom module, and a connecting arm is fixedly connected to the upper end of the electric cylinder. A hollow guide column parallel to the electric cylinder is fixedly installed at the top center of the darkroom module, and a lifting rod with its top end fixed to the connecting arm is slidably inserted into the guide column. The bottom end of the lifting rod extends through to the upper part of the interior of the darkroom module.

[0023] As a preferred technical solution provided by the present invention, the elastic transmission component includes an extension arm fixed to the lifting rod, the end of the extension arm is provided with a vertically distributed guide rod that can be raised and lowered, and a spring located below the extension arm is sleeved on the outside of the guide rod.

[0024] As a preferred technical solution provided by the present invention, the door assembly includes side rails fixedly installed on both sides of the openings on both sides of the dark box module, and a sealing strip is provided inside the side rails. A door panel is inserted between the two side rails from top to bottom, and the top of the door panel is fixedly connected to the bottom of the guide rod.

[0025] As a preferred technical solution provided by the present invention, the detector includes a multispectral imaging module and a control and computing processing module. The filter array equipped with it covers at least the visible light and near-infrared bands that are sensitive to the cherry sugar content characteristics, and is a characteristic wavelength in the range of -nm, which is used to simultaneously acquire multispectral image cubes of cherry samples at multiple discrete wavelengths.

[0026] The control and computing module includes a main control unit, an image acquisition card, and a computing unit. The main control unit coordinates and controls the timing of lighting, rotation, transmission, and image acquisition. The computing unit has built-in algorithm software for image processing, feature extraction, and sugar content prediction.

[0027] The active lighting module uses a ring-shaped LED uniform light source.

[0028] A non-destructive testing device for cherry sugar content based on multispectral imaging technology, as described above, includes the following steps:

[0029] S1. System Calibration: Use a standard reflectance whiteboard to calibrate the reflectance of the multispectral imaging system;

[0030] S2. Sample loading: Place the cherries to be tested in a single layer and column inside the tray, and use the conveyor mechanism to transport the tray to the inside of the dark box module and place it there;

[0031] S3. Image Synchronous Acquisition: The detector is activated, the rotating mechanism drives the tray to rotate and makes the cherry rotate at a constant speed. When the cherry rotates to multiple equally divided angles, the multispectral imaging module synchronously triggers the acquisition of a set of multispectral images.

[0032] S4. Image Processing and Feature Fusion: The computing unit performs 3D reconstruction and reflectivity correction on the acquired image sequences and extracts fusion feature vectors.

[0033] S5. Sugar content prediction: Input the fused feature vector into the pre-stored sugar content prediction model to calculate the predicted sugar content value of the cherry.

[0034] In the above technical solution, the present invention provides a non-destructive testing device and application for cherry sugar content based on multispectral imaging technology, which has the following beneficial effects:

[0035] (1) By setting up a dark box module, a controllable and uniform detection light environment is provided to shield external stray light interference. Rotational multi-angle imaging is used, which is equivalent to sampling most of the surface of a single cherry. The obtained sugar content information is the "overall average" rather than a single point measurement, and the result is more representative.

[0036] (2) The detector, composed of a multispectral imaging module and a control and computational processing module, effectively eliminates the influence of cherry surface curvature on spectral measurement through three-dimensional surface reconstruction and reflectivity correction. It also combines multi-dimensional features such as spectrum, color, and morphology to significantly improve the accuracy of the sugar content prediction model.

[0037] (3) The overall equipment has a compact structure and automated process. It is combined with the conveying mechanism, the testing table and the dark box module. The lifting mechanism drives the elastic transmission parts and the door components to move, thereby realizing the opening and closing of the dark box module and realizing the assembly line rapid detection of cherries, which meets the requirements of industrial sorting efficiency. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0039] Figure 1 This is a three-dimensional view of the overall structure of a non-destructive testing device for cherry sugar content based on multispectral imaging technology and its application embodiment according to the present invention.

[0040] Figure 2 This is a structural front view of a cherry sugar content non-destructive testing device and its application embodiment based on multispectral imaging technology according to the present invention.

[0041] Figure 3 This is a schematic diagram of the conveying mechanism structure provided by the present invention, which is a non-destructive testing device for cherry sugar content based on multispectral imaging technology and its application embodiment.

[0042] Figure 4This is a schematic diagram of the tray and rotating mechanism structure provided in the present invention, which is a non-destructive testing device for cherry sugar content based on multispectral imaging technology and its application embodiment.

[0043] Figure 5 This is a schematic diagram of the lifting mechanism and elastic transmission component provided in the non-destructive testing equipment for cherry sugar content based on multispectral imaging technology and its application embodiment of the present invention.

[0044] Figure 6 This is a schematic diagram of the active illumination module and the detection device structure provided by the present invention, which is a non-destructive testing device for cherry sugar content based on multispectral imaging technology and its application embodiment.

[0045] Figure 7 This is a schematic diagram of the installation structure of the tray and rotating mechanism below the dark box module, provided by the present invention, which is a non-destructive testing device for cherry sugar content based on multispectral imaging technology and its application embodiment.

[0046] Figure 8 This is a flowchart illustrating a non-destructive testing device for cherry sugar content based on multispectral imaging technology and its application embodiments according to the present invention.

[0047] 1. Testing table; 2. Conveying mechanism; 21. Frame; 22. Support arm; 23. Baffle; 24. Conveyor belt; 25. Power roller; 3. Dark box module; 4. Tray; 41. Base plate; 42. Transparent cover; 43. Magnetic block; 5. Rotating mechanism; 51. Drive motor; 52. Transmission box; 53. Energized coil; 6. Lifting mechanism; 61. Electric cylinder; 62. Connecting arm; 63. Guide column; 64. Lifting rod; 7. Elastic transmission component; 71. Extension arm; 72. Spring; 73. Guide rod; 8. Door assembly; 81. Side rail; 82. Sealing strip; 83. Door panel; 9. Active lighting module; 10. Detector. Detailed Implementation

[0048] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0049] like Figure 1-8 As shown in the figure, an embodiment of the present invention provides a non-destructive testing device for cherry sugar content based on multispectral imaging technology, including a testing stage 1.

[0050] In one embodiment of the present invention, a conveying mechanism 2 is provided on both sides of the top of the testing platform 1 for conveying cherry samples. The conveying mechanism 2 includes a frame 21 fixed to the bottom of the testing platform 1. Support arms 22 are fixedly installed on both sides of the frame 21. A power roller 25 is provided at the top of the support arm 22. A conveyor belt 24 is provided on the two power rollers 25. Baffles 23 located on both sides of the conveyor belt 24 are fixedly installed on both sides of the testing platform 1. When the motor of the power roller 25 is started, the power roller 25 can drive the conveyor belt 24 to rotate. The upper edge of the conveyor belt 24 is slidably connected to the upper surface of the testing platform 1.

[0051] In another embodiment of the present invention, the dark box module 3 is fixed above the detection stage 1, and the conveying mechanism 2 passes through its interior. The dark box module 3 is used to provide a controllable and uniform detection light environment to shield external stray light interference. The dark box module 3 is a closed structure with openings on both sides and is made of light-blocking material.

[0052] In another embodiment of the present invention, a tray 4 is provided, and cherry samples are placed inside the tray 4. The tray 4 is placed above the conveying mechanism 2. The tray 4 includes a base plate 41. A transparent cover 41 is fixedly installed on the top periphery of the base plate 41, and the top of the transparent cover 41 is an open structure. A mounting groove is provided at the bottom center of the base plate 41, and a magnetic block 43 is fixed inside the mounting groove. The magnetic block 43 is used in conjunction with an energized coil 53. When the conveyor belt 24 is in the middle, the energized coil 52 can be used to attract the magnetic block 43.

[0053] In one embodiment of the present invention, a rotating mechanism 5 is disposed below the detection platform 1. The rotating mechanism 5 is used to rotate the tray 4 without contact. The rotating mechanism 5 includes a drive motor 51 fixedly installed on the bottom outer wall of the detection platform 1. The output end of the drive motor 51 is connected to a transmission box 52. An energized coil 53 is connected to the upper part of the transmission box 52. The upper surface of the energized coil 53 is flush with the upper surface of the detection platform 1. By activating the energized coil 53, the magnetic block 43 can be attracted. At this time, the drive motor 51 is activated again to drive the energized coil 53 to rotate through the transmission box 53, thereby driving the tray 4 above the conveyor belt 24 to rotate inside the dark box module 3.

[0054] In another embodiment of the present invention, a lifting mechanism 6 is installed on the top of the dark box module 3. The lifting mechanism 6 includes an electric cylinder 61 fixedly installed on one side of the top of the dark box module 3, and a connecting arm 62 is fixedly connected to the upper end of the electric cylinder 61. A hollow guide column 63 parallel to the electric cylinder 61 is fixedly installed at the center of the top of the dark box module 3. A lifting rod 64 with its top end fixed to the connecting arm 62 is slidably inserted into the guide column 63. The bottom end of the lifting rod 64 extends through the upper part of the interior of the dark box module 3. When the electric cylinder 61 is activated, the connecting arm 62 can be driven to move up and down. The connecting arm 62 can drive the lifting rod 64 to move up and down synchronously inside the guide column 63. The lifting of the lifting rod 64 can not only drive the elastic transmission component 7 to move up and down, but also, after the elastic transmission component 7 has descended to the position, under its elastic action, as the lifting rod 64 continues to descend, it can drive the active lighting module 9 at the bottom of the lifting rod 64 to descend, thereby realizing supplementary lighting for the cherry samples on the tray 4.

[0055] In another embodiment of the present invention, the elastic transmission member 7 has one end connected to the lifting mechanism 6, which is used to drive the elastic transmission member 7 to move up and down in multiple stages. The elastic transmission member 7 includes an extension arm 71 fixed to the lifting rod 64. The end of the extension arm 71 is provided with a vertically distributed guide rod 73 that can be raised and lowered. A spring 72 located below the extension arm 71 is sleeved on the outside of the guide rod 73. After the lifting rod 64 drives the elastic transmission member 7 to descend, the elastic transmission member 7 drives the door assembly 8 to descend and close on the side of the dark box module 3. As the lifting rod 64 continues to descend, the spring 72 is compressed.

[0056] In one embodiment of the present invention, a door assembly 8 is disposed on both sides of the dark box module 3 and is used to open and close the openings on both sides of the dark box module 3. The lifting mechanism 6 drives the door assembly 8 to move up and down through the elastic transmission member 7. The door assembly 8 includes side rails 81 fixedly installed on both sides of the openings on both sides of the dark box module 3, and a sealing strip 82 is provided inside the side rails 81. A door panel 83 is inserted between the two side rails 81 from top to bottom. The top of the door panel 83 is fixedly connected to the bottom end of the guide rod 73. The opening and closing of both sides of the dark box module 3 is realized by the door panel 83 sliding inside the side rails 81.

[0057] In another embodiment of the present invention, the active lighting module 9 is disposed inside the upper part of the dark box module 3, and the active lighting module 9 is fixedly installed at the bottom center of the elastic transmission member 7. When the elastic transmission member 7 rises and falls, it drives the active lighting module 9 to move up and down a second time. The active lighting module 9 adopts a ring-shaped LED uniform light source. The emission spectrum of the LED light source includes six characteristic wavelengths: 630nm, 680nm, 780nm, 850nm, 900nm, and 940nm.

[0058] In another embodiment of the present invention, detector 10, comprising multiple detectors installed at various locations within the dark box module 3, is used to detect cherry samples on tray 4 inside the dark box module 3. Detector 10 includes a multispectral imaging module and a control and computation processing module. The core of the multispectral imaging module is a multispectral camera. The filter array equipped with the multispectral imaging module covers at least the visible and near-infrared bands sensitive to cherry sugar content characteristics, specifically 6-10 characteristic wavelengths within the 600-1000nm range, for simultaneously acquiring multispectral image cubes of cherry samples at multiple discrete wavelengths.

[0059] The control and computing module includes a main control unit, an image acquisition card, and a computing unit. The main control unit coordinates and controls the timing of lighting, rotation, transmission, and image acquisition. The computing unit has built-in algorithm software for image processing, feature extraction, and sugar content prediction.

[0060] The algorithm process includes:

[0061] Image preprocessing and surface reconstruction: Align multi-band images, perform 3D point cloud reconstruction on rotated sequence images, and obtain information on the size, shape and surface curvature of cherries.

[0062] Reflectivity correction: Using the reconstructed three-dimensional surface model, optical correction is performed on the difference in incident-reflection angles of pixels in the original multispectral image caused by surface geometry, and the corrected surface reflectivity image is calculated.

[0063] Feature extraction and fusion: Extract the average reflectance, reflectance distribution variance, texture features, etc. for each feature wavelength from the corrected reflectance image; at the same time, extract color features (such as R, G, B components, hue H) from the visible light band image; and extract morphological features such as size, volume, and sphericity from the 3D model.

[0064] Sugar content prediction model: The extracted multispectral features, color features and morphological features are used as input variables, and the sugar content of cherry samples measured by standard methods is used as the output variable. A sugar content prediction model is established using machine learning algorithms (such as partial least squares regression PLSR, support vector machine regression SVR or lightweight neural network) and embedded in the computing unit.

[0065] Real-time prediction and decision-making: For a new input cherry multispectral image sequence, the above process is run in real time, and the predicted sugar content is output through the sugar content prediction model.

[0066] An application such as a non-destructive testing device for cherry sugar content based on multispectral imaging technology includes the following steps:

[0067] S1. System Calibration: Use a standard reflectance whiteboard to calibrate the reflectance of the multispectral imaging system;

[0068] S2, Sample loading: Place the cherries to be tested in a single layer and row inside the tray 4, and use the conveying mechanism 2 to transport the tray 4 to the inside of the dark box module 3 and keep it there;

[0069] S3. Image synchronous acquisition: Detector 10 is started, and the rotating mechanism 5 drives the tray 4 to rotate, making the cherry rotate at a constant speed. When the cherry rotates to multiple equally divided angles, the multispectral imaging module synchronously triggers the acquisition of a set of multispectral images.

[0070] S4. Image Processing and Feature Fusion: The computing unit performs 3D reconstruction and reflectivity correction on the acquired image sequences and extracts fusion feature vectors.

[0071] S5. Sugar content prediction: Input the fused feature vector into the pre-stored sugar content prediction model to calculate the predicted sugar content value of the cherry.

[0072] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A non-destructive testing device for cherry sugar content based on multispectral imaging technology, comprising a testing stage (1), characterized in that, The conveying mechanism (2) is set on both sides of the top of the detection station (1) for conveying cherry samples; The dark box module (3) is fixed above the detection stage (1), and the conveying mechanism (2) passes through its interior. The dark box module (3) is used to provide a controllable and uniform detection light environment to shield external stray light interference. The tray (4) contains cherry samples placed inside it, and the tray (4) is placed above the conveying mechanism (2). A rotating mechanism (5) is disposed below the detection table (1) and is used to rotate the tray (4) without contact. A lifting mechanism (6) is installed on top of the dark box module (3); Elastic transmission component (7), one end of which is connected to lifting mechanism (6), which is used to drive the elastic transmission component (7) to move up and down in multiple stages; Door assembly (8), the door assembly (8) is disposed on both sides of the dark box module (3) and is used to open and close the openings on both sides of the dark box module (3). The lifting mechanism (6) drives the door assembly (8) to move up and down through the elastic transmission member (7). An active lighting module (9) is located inside the dark box module (3) and is fixedly installed at the bottom center of the elastic transmission member (7). When the elastic transmission member (7) is raised or lowered, it drives the active lighting module (9) to move up and down. Detectors (10), there are multiple detectors (10) installed in multiple locations inside the dark box module (3), the detectors (10) are used to detect cherry samples on trays (4) inside the dark box module (3).

2. The cherry sugar content non-destructive testing device based on multispectral imaging technology according to claim 1, characterized in that, The conveying mechanism (2) includes a frame (21) fixed at the bottom of the testing table (1), with support arms (22) fixedly installed on both sides of the frame (21), a power roller (25) provided at the top of the support arm (22), a conveyor belt (24) provided on the two power rollers (25), and baffles (23) located on both sides of the conveyor belt (24) fixedly installed on both sides of the testing table (1).

3. The cherry sugar content non-destructive testing device based on multispectral imaging technology according to claim 1, characterized in that, The dark box module (3) is a closed structure with openings on both sides and is made of light-blocking material.

4. The cherry sugar content non-destructive testing device based on multispectral imaging technology according to claim 1, characterized in that, The tray (4) includes a base plate (41), a transparent cover (41) is fixedly installed on the top periphery of the base plate (41), and the top of the transparent cover (41) is an open structure. A mounting groove is provided at the bottom center of the base plate (41), and a magnetic block (43) is fixed inside the mounting groove.

5. The cherry sugar content non-destructive testing device based on multispectral imaging technology according to claim 4, characterized in that, The rotating mechanism (5) includes a drive motor (51) fixedly installed on the bottom outer wall of the test platform (1), and the output end of the drive motor (51) is connected to a transmission box (52). An energized coil (53) is connected to the top of the transmission box (52), and the upper surface of the energized coil (53) is flush with the upper surface of the test platform (1).

6. The cherry sugar content non-destructive testing device based on multispectral imaging technology according to claim 1, characterized in that, The lifting mechanism (6) includes an electric cylinder (61) fixedly installed on one side of the top of the dark box module (3), and a connecting arm (62) is fixedly connected to the upper end of the electric cylinder (61). A hollow guide column (63) parallel to the electric cylinder (61) is fixedly installed at the center of the top of the dark box module (3), and a lifting rod (64) with its top end fixed to the connecting arm (62) is slidably inserted inside the guide column (63). The bottom end of the lifting rod (64) extends through to the upper part of the interior of the dark box module (3).

7. The cherry sugar content non-destructive testing device based on multispectral imaging technology according to claim 6, characterized in that, The elastic transmission component (7) includes an extension arm (71) fixed to the lifting rod (64). The end of the extension arm (71) is provided with a vertically distributed guide rod (73) that can be raised and lowered. A spring (72) located below the extension arm (71) is sleeved on the outside of the guide rod (73).

8. The cherry sugar content non-destructive testing device based on multispectral imaging technology according to claim 7, characterized in that, The door assembly (8) includes side rails (81) fixedly installed on both sides of the openings on both sides of the dark box module (3), and a sealing strip (82) is provided inside the side rails (81). A door panel (83) is inserted between the two side rails (81) from top to bottom, and the top of the door panel (83) is fixedly connected to the bottom of the guide rod (73).

9. The cherry sugar content non-destructive testing device based on multispectral imaging technology according to claim 1, characterized in that, The detector (10) includes a multispectral imaging module and a control and computing processing module. The filter array equipped with it covers at least the visible light and near-infrared bands that are sensitive to the cherry sugar content characteristics, which are 6-10 characteristic wavelengths in the range of 600-1000nm, and is used to simultaneously acquire multispectral image cubes of cherry samples at multiple discrete wavelengths. The control and computing module includes a main control unit, an image acquisition card, and a computing unit. The main control unit coordinates and controls the timing of lighting, rotation, transmission, and image acquisition. The computing unit has built-in algorithm software for image processing, feature extraction, and sugar content prediction. The active lighting module (9) uses a ring-shaped LED uniform light source.

10. A non-destructive testing device for cherry sugar content based on multispectral imaging technology as described in any one of claims 1-9, characterized in that, Includes the following steps: S1. System Calibration: Use a standard reflectance whiteboard to calibrate the reflectance of the multispectral imaging system; S2, Sample loading: Place the cherries to be tested in a single layer and column inside the tray (4), and use the conveying mechanism (2) to transport the tray (4) to the inside of the dark box module (3) and stop there; S3, Image Synchronous Acquisition: Start the detector (10), the rotating mechanism (5) drives the tray (4) to rotate and make the cherry rotate at a constant speed. When the cherry rotates to multiple equally divided angles, the multispectral imaging module synchronously triggers the acquisition of a set of multispectral images. S4. Image Processing and Feature Fusion: The computing unit performs 3D reconstruction and reflectivity correction on the acquired image sequences and extracts fusion feature vectors. S5. Sugar content prediction: Input the fused feature vector into the pre-stored sugar content prediction model to calculate the predicted sugar content value of the cherry.

Images