Plant growth vigor analysis method based on illumination

Abstract

The invention provides a plant growth vigor analysis method based on illumination. The method comprises the following steps: shooting a plurality of crop plant pictures at different visual angles; synthesizing a three-dimensional plant image by using a three-dimensional reconstruction technology; simulating sunlight irradiation conditions at different moments according to the actual longitude and latitude information and geographical orientation of the plant; and analyzing, calculating and recording the total projection area of the leaves in each projection shooting image, and calculating to obtain an average illumination area according to the sunlight intensity at different moments so as to reflect the influence of illumination on the plant growth vigor.

Description

Technical Field

[0001] The present invention relates to the field of leaf area detection, and in particular to a method for analyzing the growth trend of plants based on light. Background Art

[0002] Leaf area is an important indicator reflecting the growth and development, physiology and ecology of crops, and is also a factor that must be referred to in the research of crop cultivation, genetic breeding, etc. Rapid and accurate measurement of plant leaf area can effectively improve the research efficiency. Since the leaf shapes of the same kind of plants are similar, one of the existing leaf area detection methods is to use a handheld leaf length detector to detect the leaf length, and then through the area conversion formula, substitute the measured long-axis dimension of the leaf to obtain the leaf area.

[0003] In traditional research on analyzing the growth trend of plants based on light, it mainly reflects the growth trend of plants by calculating the total area of all leaves. However, due to the different leaf shapes of different plants, and the differences in the positions, angles and densities of the leaves, these factors all affect the area of the finally received light. And because the angle of sunlight changes constantly during the day, therefore, the single statistics of the total leaf area cannot accurately reflect the total amount of sunlight received, and thus cannot accurately reflect the growth trend of plants. Summary of the Invention

[0004] The present invention provides a method for analyzing the growth trend of plants based on light, which solves the problem that the traditional method of counting the total leaf area of plants cannot accurately reflect the influence of light on the growth trend of plants.

[0005] To solve the above technical problems, the technical solution adopted by the present invention is: A method for analyzing the growth trend of plants based on light: Take multiple pictures of the crop plant from different perspectives, preferably take at least three pictures with pairwise perpendicular perspectives from the top and two vertical sides. When taking pictures from the top, the leaf length detection device calibrates the geographical orientation and records the longitude and latitude information; Analyze the pictures of the same crop from different perspectives, mark the leaves that appear in multiple pictures, and display them on the display screen of the leaf length detection device; The operator selects one of the marked leaves, which can be selected by button or touch screen operation. When the selection is completed, the leaf length detection device records the leaf. The operator uses the leaf length detection device to measure the actual size of the leaf, and the leaf length detection device automatically enters the leaf size data, and uses this leaf scale data as the scale reference for each picture; Save the set of pictures and the actually detected leaf scale data, and perform data association; Input the taken pictures and leaf detection data into a computer for analysis; Analyze and calculate the positions, angles, and sizes of each leaf in the captured images, and synthesize multiple captured images into a three-dimensional plant image. Multiple photos are all included. For example, the leaves inside are discarded. The more shooting perspectives there are, the closer the three-dimensional plant image is to the real situation; According to the actual latitude and longitude information and geographical orientation of the plant, perform light simulation. In the virtual space, a plurality of azimuth virtual cameras are evenly distributed circumferentially in the upper space of the three-dimensional plant image, and each azimuth virtual camera performs projection shooting to obtain projection shooting images; Analyze, calculate, and record the total projected area of the leaves in each projection shooting image, denoted as the azimuth leaf total area. This process requires finding the outermost boundary of the leaf distribution and removing the gap areas where all leaves do not overlap; According to the real light situation of the plant without clouds, since the sunlight intensity distribution changes with time during the day, establish a light intensity gain coefficient according to different light angles. The equivalent light area of the plant at a single azimuth is denoted as the azimuth light area, and the azimuth light area is equal to the product of each light intensity gain coefficient and each azimuth leaf total area. The average light area of the plant during a day is the cumulative value of each azimuth equivalent light area divided by the number of azimuth virtual cameras.

[0006] In the preferred solution, it includes a leaf image recognition method: When plant leaves overlap each other, calculate the total projected area of the leaves by image recognition technology and remove the gap part. This process involves multiple steps, including image acquisition, preprocessing, leaf recognition and segmentation, projected area calculation, etc. The specific steps are as follows: Image acquisition: Use a high-resolution camera or scanner to shoot or scan plant leaves from different angles to ensure that each leaf is clearly visible in at least one perspective.

[0007] Shooting from multiple angles helps to distinguish overlapping leaves in subsequent processing.

[0008] Image preprocessing: Perform denoising, grayscale conversion, and binarization on the acquired images.

[0009] Denoising can use a median filter or a Gaussian filter to reduce the noise in the image.

[0010] Convert the color image to a grayscale image, and then apply an appropriate threshold to binarize the image to distinguish the leaves (foreground) and the background.

[0011] Leaf recognition and segmentation: Use machine learning algorithms such as deep learning convolutional neural networks (CNNs) to recognize and segment the leaves in the image.

[0012] For overlapping leaves, their respective contours can be recognized through a trained neural network model.

[0013] Projected area calculation: Based on the lighting model, simulate the illumination of light rays from different directions and calculate the projected area of the leaf in a specific direction. Considerations should be given to the orientation and curvature of the leaf, as well as the direction of the light rays.

[0014] Overlay analysis and void removal: By analyzing the front-back relationship and overlapping situation among the leaves, determine the occluded parts in the projection. Then, subtract these void parts from the total projected area to obtain the actual effective projected area of the leaf.

[0015] Through the above steps, the total projected area of the mutually occluding plant leaves can be calculated relatively accurately, and the void parts caused by leaf overlap can be effectively removed. This method combines modern image processing technology and the lighting model, improving the accuracy of leaf area calculation.

[0016] In a preferred embodiment, the leaf length detection device is a handheld leaf length detection device. The handheld leaf length detection device includes a holding frame body. A clamping plate is provided on the holding frame body. One end of the clamping plate is hinged to the holding frame body, and the other end of the clamping plate can be opened. A through gap is provided between the clamping plate and the holding frame body, and the through gap is used for placing the leaf. A rotatable friction roller is provided on the side of the clamping plate close to the holding frame body. One end of the friction roller is provided with a friction wheel driving motor with an encoder. A camera is also provided on the holding frame body.

[0017] In a preferred embodiment, a concave cavity is provided inside the clamping plate. A bearing seat is provided in the concave cavity of the clamping plate. A rotating shaft is also provided. One end of the rotating shaft is connected to the friction roller. A torsion sensor is also provided. The other end of the rotating shaft is connected to the friction wheel driving motor through the torsion sensor. The friction roller protrudes from the opening of the concave cavity of the clamping plate.

[0018] In a preferred embodiment, the clamping plate is provided with an installation slot hole. A telescopic buffer rod is also provided on the side of the clamping plate close to the holding frame body. The buffer rod is slidably sleeved with the installation slot hole. A slidable plug block is provided in the installation slot hole. A screw rod portion is provided on the side of the buffer rod away from the holding frame body. The screw rod portion is threadedly connected to the plug block. An adjusting block is threadedly connected to the end of the installation slot hole away from the holding frame body. A buffer spring is provided between the adjusting block and the plug block.

[0019] In a preferred embodiment, a fixing block is provided on the holding frame body. A shaft rod is provided in the middle of the fixing block. One end of the clamping plate is provided with a hinged sleeve portion, and the hinged sleeve portion is rotatably sleeved with the shaft rod. An ear is provided at the end of the clamping plate away from the friction roller. A limiting rod and a return spring are provided on the ear. A fixing ring is also provided on the side of the fixing block on the holding frame body. The return spring is connected to the fixing ring.

[0020] In a preferred embodiment, a control panel and a display screen are also provided at the holding end of the holding frame body.

[0021] In a preferred embodiment, a storage assembly is provided on the holding frame body. The storage assembly includes a storage groove, and sliding grooves are provided on both sides of the storage groove near the opening. A slidable cover plate is provided in the sliding grooves.

[0022] The beneficial effects of the present invention are as follows: By reflecting the total light area received by the leaf surface of a plant at a certain time through the total projected area, it is more in line with the actual situation of the plant receiving light, and the calculation of the total light area is more accurate; By simulating the light conditions at different times of a day and adjusting the light gain coefficient through geographical information, the light simulation is closer to the actual situation; Through the method of three-dimensional reconstruction, fitting the virtual plant, when simulating the light angle, only the projection diagram needs to be analyzed to obtain the light area at that azimuth angle, which is easier to process and analyze, and there is no need to manually simulate the sunlight angle for multiple azimuth shootings, greatly reducing the labor intensity. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The present invention will be further described below with reference to the drawings and embodiments.

[0024] Figure 1 is a schematic diagram of a method for analyzing the growth trend of plants based on light.

[0025] Figure 2 is a side view of a method for analyzing the growth trend of plants based on light.

[0026] Figure 3 is a structural diagram of the holding frame body.

[0027] Figure 4 is a structural diagram of the clamping plate.

[0028] Figure 5 is a cross-sectional view of the clamping plate.

[0029] Figure 6 is a cross-sectional view of the installation slot hole.

[0030] Figure 7 is a schematic diagram of the three-dimensional reconstruction of the plant.

[0031] Figure 8 is a schematic diagram of the light angle simulation.

[0032] In the figure: holding frame body 1; buffer mechanism 2; installation slot hole 201; buffer rod 202; buffer spring 203; adjusting block 204; screw part 205; sliding plug block 206; blade translation assembly 3; friction wheel drive motor 301; bearing seat 302; torque sensor 303; rotating shaft 304; friction roller 305; control panel 4; fixing block 5; shaft rod 6; clamping plate 7; hinge sleeve part 701; storage assembly 8; storage groove 801; sliding groove 802; cover plate 803; cleaning belt 804; pressing ear 9; limiting rod 10; return spring 11; fixing ring 12; display screen 13; camera 14; through gap 15. Detailed implementation mode

[0033] Example 1: As Figure 1-8 , a method for analyzing the growth trend of plants based on light, is carried out by means of three-dimensional reconstruction. The multi-angle projection of the plant simulates the sunlight during the day. The larger the effective light-receiving area of the plant, the larger the area capable of photosynthesis, and the better the growth of the plant. The specific method is as follows: 1. Image acquisition Multi-angle shooting: Use a high-resolution camera to take at least three photos with pairwise perpendicular perspectives from the top and two vertical sides to ensure that the leaves are captured from all angles. This can obtain images of the plant from different perspectives, making it easier to distinguish overlapping leaves in subsequent processing; Geographic information recording: When shooting from the top, use a leaf length detection device to calibrate the geographic orientation and record the longitude and latitude information.

[0034] 2. Image preprocessing Denoising: Use a median filter or a Gaussian filter for denoising to reduce the noise in the image and improve the image quality.

[0035]

[0036] is the original image, is the filtered image, and * represents the convolution operation; Grayscale conversion: Convert the color image to a grayscale image for subsequent image processing; ; where R, G, and B represent the pixel values of the red, green, and blue channels respectively; Binarization: Binarize the image by setting a threshold to distinguish the leaves (foreground) from the background; .

[0037] 3. Leaf recognition and segmentation Deep learning method: Use a convolutional neural network (CNN) for leaf recognition. The CNN can automatically extract features from the image and learn how to use these features for leaf classification and segmentation.

[0038]

[0039] where is the prediction result, is the input image, is the model parameter; Contour extraction: Extract the contours of the leaves from the binarized image for further processing; 。

[0040] 4. Projection Area Calculation Lighting Simulation: Assume the lighting direction and lighting intensity , through lighting simulation, we can calculate the projection area of the leaf in a specific direction, taking into account the orientation and curvature of the leaf, as well as the direction of the light.

[0041]

[0042] where is the leaf contour, is the normal vector, is the infinitesimal area element; Light Intensity Gain Coefficient: Calculate the light intensity gain coefficient according to the change of the sun's position during the day where and are the azimuth angle and altitude angle;

[0043] where α and β are the altitude angle and azimuth angle of the sun respectively.

[0044] 5. Overlay Analysis and Gap Removal Calculate the effective projection area: Subtract the gaps between the leaves from the total projection area, excluding the gaps caused by leaf overlap;

[0045] where is the area of the gaps between the leaves.

[0046] Gap Area Calculation: For each gap region, calculate its area;

[0047] where M is the number of gap regions, is the th gap region area; Equivalent Illumination Area: For each azimuth angle, calculate the equivalent illumination area, which takes into account the influence of the light intensity gain coefficient;

[0048] Average Illumination Area: For a complete day, calculate the average illumination area. This step comprehensively considers the change of light intensity at different times of the day;

[0049] where N is the number of virtual cameras.

[0050] Embodiment 2: As Figure 1-6 disclosed in, a method for analyzing the growth trend of plants based on light includes a holding frame 1, a clamping plate 7 is arranged on the holding frame 1, one end of the clamping plate 7 is hinged to the holding frame 1, the other end of the clamping plate 7 can be opened, a through-gap 15 is arranged between the clamping plate 7 and the holding frame 1, and the through-gap 15 is used for placing leaves. A rotatable friction roller 305 is arranged on one side of the clamping plate 7 close to the holding frame 1, and a friction wheel driving motor 301 with an encoder is arranged at one end of the friction roller 305.

[0051] The friction roller 305 can be multiple or have a relatively wide wheel surface to reduce the contact pressure with the leaves and prevent damage to the leaf surface.

[0052] In a preferred solution, a concave cavity is arranged inside the clamping plate 7, a bearing seat 302 is arranged in the concave cavity of the clamping plate 7, a rotating shaft 304 is also arranged, one end of the rotating shaft 304 is connected to the friction roller 305, a torsion sensor 303 is also arranged, the other end of the rotating shaft 304 is connected to the friction wheel driving motor 301 through the torsion sensor 303, and the friction roller 305 protrudes from the concave cavity opening of the clamping plate 7.

[0053] The friction wheel driving motor 301 is a reduction motor, and the surface of the friction roller 305 is provided with patterns to increase the friction force on the leaves.

[0054] In a preferred solution, the clamping plate 7 is provided with an installation slot hole 201, a telescopic buffer rod 202 is also arranged on the side of the clamping plate 7 close to the holding frame 1, the buffer rod 202 is slidably sleeved with the installation slot hole 201, a slidable sliding plug block 206 is arranged in the installation slot hole 201, a screw rod part 205 is arranged on the side of the buffer rod 202 away from the holding frame 1, the screw rod part 205 is threadedly connected with the sliding plug block 206, and an adjusting block 204 connected by threads is also arranged at one end of the installation slot hole 201 away from the holding frame 1. A buffer spring 203 is arranged between the adjusting block 204 and the sliding plug block 206.

[0055] The installation slot hole 201 is a stepped hole, and the end with the buffer rod 202 has a smaller opening, which can stop the sliding plug block 206 to prevent it from coming out.

[0056] Both ends of the buffer spring 203 respectively abut against the groove parts of the adjusting block 204 and the sliding plug block 206. Rotating the adjusting block 204 can adjust the compression amount of the buffer spring 203 and change the contact pressure of the friction roller 305 on the leaves.

[0057] In a preferred embodiment, the holding frame body 1 is provided with a fixing block 5. A shaft rod 6 is provided in the middle of the fixing block 5. One end of the clamping plate 7 is provided with a hinge sleeve portion 701. The hinge sleeve portion 701 is rotatably sleeved on the shaft rod 6. A pressing ear 9 is provided at the end of the clamping plate 7 away from the friction roller 305. A limiting rod 10 and a return spring 11 are provided on the pressing ear 9. A fixing ring 12 is also provided on one side of the fixing block 5 on the holding frame body 1. The return spring 11 is connected to the fixing ring 12.

[0058] In a preferred embodiment, a control panel 4 and a display screen 13 are further provided at the holding end of the holding frame body 1.

[0059] The control panel 4 can be provided with a plurality of operation buttons. The control panel 4 and the display screen 13 can also be integrated into a touch screen.

[0060] The operation method is as follows: First, adjust the extension amount of the buffer rod 202 and the compression amount of the buffer spring 203 in advance to ensure appropriate buffer force and stop position for the clamping plate 7. When holding the end of the holding frame body 1, press the pressing ear 9 with a finger and compress the return spring 11 until the limiting rod 10 abuts against the holding frame body 1. At this time, the front end of the clamping plate 7 tilts up. Hold the device and extend it to the leaf, so that the clamping plate 7 is above the leaf and the holding frame body 1 is below the leaf. There are marking scale lines on the clamping plate 7. Align the center of the central friction roller 305, align the center of the friction roller 305 with the central axis of the leaf, move the starting point of the friction roller 305 to the root of the leaf, close the clamping plate 7, and the buffer rod 202 buffers the impact force during closing. Under the action of the gravity of the clamping plate 7, the friction roller 305 always presses tightly against the leaf surface, and the buffer spring 203 is compressed to offset part of the gravity, avoiding excessive contact pressure of the friction roller 305 and damaging the leaf surface. Operate the control panel 4 to start the friction wheel drive motor 301. The leaf is pushed to translate by the frictional force of the friction roller 305. At this time, the encoder of the friction wheel drive motor 301 records the number of rotations of the friction roller 305. When the friction roller 305 drives the leaf to escape through the gap 15, the friction roller 305 idles, and the torque sensor 303 detects a sudden change in torque, determining that the leaf has completed the full stroke, and the encoder terminates recording the number of rotations. At this time, convert the recorded number of rotations into a displacement value, which is the length of the longest axis of the leaf. Similarly, this device can also detect the lateral width of the leaf.

[0061] After scaling the length of the leaf, the touch screen can be operated to take images of the plant from multiple angles by the camera 14 for use in image processing, deep learning, and three-dimensional reconstruction of the plant.

[0062] Embodiment 3: A handheld leaf length detector, comprising a holding frame body 1, a buffer mechanism 2, a blade translation assembly 3 and a storage assembly 8. One end of the top of the holding frame body 1 is provided with a control panel 4. Fixed blocks 5 are fixedly installed on both sides of the top of the holding frame body 1. A shaft rod 6 is fixedly installed inside the fixed block 5. A clamping plate 7 is rotatably installed on the outer side of the shaft rod 6. The blade translation assembly 3 includes a friction wheel driving motor 301, a bearing seat 302, an encoder, a rotating shaft 304 and a friction roller 305. The friction wheel driving motor 301 and the bearing seat 302 are respectively fixedly installed at both ends inside the clamping plate 7. The encoder is fixedly connected to the end of the output shaft of the friction wheel driving motor 301. The rotating shaft 304 is fixedly connected to the end of the encoder. The friction roller 305 is fixedly installed on the outer side of the rotating shaft 304.

[0063] The number of the friction rollers 305 is two, and they extend to the outside of the clamping plate 7 through the through holes at the bottom of the clamping plate 7. One end of the rotating shaft 304 is rotatably installed inside the bearing seat 302.

[0064] The holding frame body 1 is the core component of the detector. The holding frame body 1 integrates all electronic components and control systems. A control panel 4 is provided at the top for inputting commands and viewing measurement results. The fixed block 5 is used to support the clamping plate 7. The fixed block 5 is connected to the clamping plate 7 through the shaft rod 6, enabling the clamping plate 7 to rotate around the shaft rod 6 to clamp the blade. Through the setting of the blade translation assembly 3, the automatic measurement of the blade area can be realized, with fast measurement speed, high accuracy, and reduction of manual errors. The friction wheel driving motor 301 provides power to drive the rotation of the rotating shaft 304. The bearing seat 302 supports the rotating shaft 304 and reduces friction and wear during rotation. The encoder records the rotation angle or distance of the rotating shaft 304, and then calculates the number of turns of the friction roller 305. The rotating shaft 304 transmits the power of the friction wheel driving motor 301 to the friction roller 305. The friction roller 305 contacts the blade and rolls, and calculates the length of the blade through the number of turns of its rolling. Through the ingenious layout of the friction roller 305, the adaptability of the device is significantly enhanced, ensuring that it can easily handle and accurately measure blades of various widths, thus expanding the detection range and practicability of the detector.

[0065] An ear 9 is fixedly installed at one end of the clamping plate 7. A limiting rod 10 is fixedly installed at the bottom of the ear 9. A return spring 11 is sleeved on the outer side of the limiting rod 10. A fixed ring 12 is fixedly installed at the top of the holding frame body 1. The end of the return spring 11 is fixedly installed on the fixed ring 12.

[0066] The ear 9 and the limiting rod 10 cooperate with the return spring 11 to provide a clamping force and a reset function for the clamping plate 7, ensuring that the clamping force is moderate and reusable, and realizing the automatic reset of the clamping plate 7 by using the elastic force of the spring, reducing the operation steps of use.

[0067] The buffer mechanism 2 includes a mounting slot 201, a buffer rod 202, and a buffer spring 203. The mounting slot 201 is provided at the bottom of the clamping plate 7. Both the buffer rod 202 and the buffer spring 203 are arranged inside the mounting slot 201. The buffer rod 202 is slidably mounted inside the mounting slot 201, and its top end is provided with an opening. The buffer spring 203 is located inside the buffer rod 202.

[0068] The buffer mechanism 2 can weaken the impact generated when the clamping plate 7 closes. And when the friction roller 305 contacts the blade, a certain impact force will be generated. The buffer rod 202 and the buffer spring 203 can absorb and relieve this part of the impact force, protecting components such as the friction roller 305 and the rotating shaft 304 from damage.

[0069] Specifically, the storage assembly 8 includes a storage slot 801, a sliding groove 802, a cover plate 803, and a cleaning belt 804. The storage slot 801 is provided at one end of the top of the holding frame 1. The sliding groove 802 is provided at one end of the holding frame 1. The cover plate 803 slides inside the sliding groove 802. The cleaning belt 804 is placed inside the storage slot 801. The top of the cover plate 803 is provided with anti-slip patterns. The sliding groove 802 is located at the top of the storage slot 801 and is communicated with the storage slot 801.

[0070] The storage assembly 8 is used to store accessories such as the cleaning belt 804, improving the convenience of use, reducing the loss and damage of the blade. The cover plate 803 is opened or closed in a sliding manner, so as to take and store the blade. The cleaning belt 804 is used to clean the friction roller 305.

[0071] During use, directly press the earpiece 9, which will drive the clamping plate 7 to lift upward, causing the separation between the clamping plate 7 and the holding frame body 1. At the same time, the downward pressing of the earpiece 9 will exert pressure on the return spring 11, causing it to undergo elastic compression. Subsequently, one end of the leaf to be measured is accurately placed within the scanning and measuring area of the holding frame body 1, and it is ensured that the leaf is directly below the friction roller 305. After the placement of the leaf is completed, the pressure on the earpiece 9 needs to be released. Meanwhile, the elastic force stored in the return spring 11 due to compression begins to be released, causing the earpiece 9 and the clamping plate 7 to close and the clamping plate 7 to tightly clamp the leaf. During the closing process of the clamping plate 7, the buffer rod 202 first contacts the contact surface of the holding frame body 1. As the clamping force further increases, the buffer rod 202 slides inward along the direction of the mounting slot hole 201, and during this process, the buffer spring 203 is compressed, thereby alleviating the impact force generated during the clamping process and protecting the device from damage caused by the impact force. Subsequently, press the measurement button on the control panel 4 to activate the transmission system inside the device. After receiving the start signal, the friction wheel drive motor 301 starts to rotate and drives the connected rotating shaft 304 to rotate synchronously. The rotation of the rotating shaft 304 further drives the friction roller 305 to roll on the surface of the leaf, and the leaf moves synchronously along its rolling direction under the push of the friction roller 305. At the same time, the encoder will accurately record the displacement of the friction roller 305 during the entire rolling process in real time. When it is necessary to clean the friction roller 305, push the cover plate 803 to open the storage slot 801, take out the cleaning belt 804, and place the cleaning belt 804 below the friction roller 305. At this time, the cleaning belt 804 moves synchronously along its rolling direction under the push of the friction roller 305 and cleans the impurities on the surface of the friction roller 305 during the movement.

[0072] Through the built-in leaf translation component 3, the leaf length is measured, and a conversion formula can be integrated into the system. By substituting the leaf length, the automatic measurement of the leaf area is achieved. This design not only significantly improves the measurement speed but also greatly enhances the measurement accuracy, reducing the errors that may occur in traditional manual measurement methods. The layout of the dual friction rollers 305 enables the detector to easily handle and accurately measure leaves of various widths, significantly enhancing the adaptability of the device and expanding the detection range of the detector, making it play an important role in the leaf measurement of different plant species and growth stages. Through the setting of the buffer mechanism 2, the impact force can be alleviated when the clamping plate 7 closes and the friction roller 305 contacts the leaf, effectively protecting key components such as the friction roller 305 and the rotating shaft 304 from damage and extending the service life of the device. The design of the storage component 8 takes into account the convenience of use. Through the combination of the storage slot 801, the sliding slot 802, the cover plate 803, and the cleaning belt 804, it provides convenience for storing and accessing accessories. This not only reduces the risk of loss and damage but also makes operations such as cleaning the friction roller 305 simpler and faster.

[0073] The above embodiments are only the preferred technical solutions of the present invention and should not be regarded as limitations on the present invention. The protection scope of the present invention shall be the technical solutions recorded in the claims, including the equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, the equivalent replacement improvements within this scope are also within the protection scope of the present invention.

Claims

1. A method for analyzing plant growth based on light, characterized in that: Take multiple pictures of crop plants from different perspectives; Analyze pictures of the same crop from different perspectives and mark leaves that appear in multiple pictures; Select one of the marked leaves; The blade length detection device records the blade, and the operator uses the blade length detection device to measure the actual size of the blade. The blade length detection device automatically enters the blade size data, and this blade size data is used as the scale reference of each picture; The group of images and the actual detected blade scale data are saved and the data are associated; Analyze and calculate the position, angle and size of each leaf in the photographed image, and synthesize multiple photographed images into a three-dimensional plant image; According to the actual latitude and longitude information and geographic orientation of the plant, a lighting simulation is performed. In the virtual space, multiple azimuth virtual cameras are evenly distributed along the circumference above the three-dimensional plant image. Each azimuth virtual camera performs projection shooting to obtain a projection shooting image. Analyze, calculate and record the total projection area of ​​the blade in each projected image, which is recorded as the total azimuth blade area; The light intensity gain coefficient is established according to the light angle at different time periods, and the equivalent light area of ​​the plant at a single azimuth angle is recorded as the azimuth light area. The azimuth illumination area is equal to the light intensity gain coefficient multiplied by the total area of ​​the blades at each azimuth angle. The average illumination area is equal to the accumulated value of the azimuth equivalent illumination area divided by the number of azimuth virtual cameras.

2. The method for analyzing plant growth potential based on light according to claim 1, characterized in that: Including leaf image recognition method: Image acquisition: Use a high-resolution camera or scanner to photograph or scan plant leaves from different angles to ensure that each leaf is clearly visible in at least one viewing angle; Image preprocessing: denoising, grayscale and binarization of the collected images; Leaf recognition and segmentation: Using machine learning algorithms such as deep learning convolutional neural networks to recognize and segment leaves in images; Projection area calculation: Based on the illumination model, simulate the light irradiation in different directions and calculate the projection area of ​​the leaf in a specific direction; Overlay analysis and gap removal: By analyzing the front-to-back relationship and overlap between the leaves, the blocked part in the projection is determined, and the gap part is subtracted from the overall projection area to obtain the actual leaf projection area.

3. The method for analyzing plant growth based on light according to claim 1 or 2, characterized in that: The length detection device is a handheld blade length detection device, comprising a holding frame (1), a clamping plate (7) being provided on the holding frame (1), one end of the clamping plate (7) being hinged to the holding frame (1), the other end of the clamping plate (7) being openable, a through gap (15) being provided between the clamping plate (7) and the holding frame (1), the through gap (15) being used to place the blade, a rotatable friction roller (305) being provided on one side of the clamping plate (7) close to the holding frame (1), a friction wheel driving motor (301) with an encoder being provided at one end of the friction roller (305), and a camera (14) being further provided on the holding frame (1).

4. The method for analyzing plant growth potential based on light according to claim 3, characterized in that: A concave cavity is provided in the clamping plate (7), a bearing seat (302) is provided in the concave cavity of the clamping plate (7), a rotating shaft (304) is provided, one end of the rotating shaft (304) is connected to a friction roller (305), a torque sensor (303) is provided, the other end of the rotating shaft (304) is connected to a friction wheel driving motor (301) via the torque sensor (303), and the friction roller (305) protrudes from the concave cavity opening of the clamping plate (7).

5. The method for analyzing plant growth potential based on light according to claim 3, characterized in that: The clamping plate (7) is provided with a mounting slot (201), and a retractable buffer rod (202) is also provided on the side of the clamping plate (7) close to the holding frame (1), the buffer rod (202) is slidably sleeved with the mounting slot (201), a slidable sliding plug block (206) is provided in the mounting slot (201), a screw portion (205) is provided on the side of the buffer rod (202) away from the holding frame (1), the screw portion (205) is threadedly connected to the sliding plug block (206), and an adjustment block (204) threadedly connected is also provided at one end of the mounting slot (201) away from the holding frame (1), and a buffer spring (203) is provided between the adjustment block (204) and the sliding plug block (206).

6. The method for analyzing plant growth potential based on light according to claim 3, characterized in that: A holding frame (1) is provided with a fixing block (5), a shaft (6) is provided in the middle of the fixing block (5), a hinged sleeve (701) is provided at one end of the clamping plate (7), the hinged sleeve (701) is rotatably sleeved with the shaft (6), a pressing ear (9) is provided at the end of the clamping plate (7) away from the friction roller (305), a limiting rod (10) and a return spring (11) are provided on the pressing ear (9), and a fixing ring (12) is also provided on one side of the fixing block (5) on the holding frame (1), and the return spring (11) is connected to the fixing ring (12).

7. The method for analyzing plant growth potential based on light according to claim 3, characterized in that: The gripping end of the gripping frame (1) is also provided with a control panel (4) and a display screen (13).

8. The method for analyzing plant growth potential based on light according to claim 3, characterized in that: A storage assembly (8) is provided on the grip frame (1), and the storage assembly (8) comprises a storage slot (801). Slide slots (802) are provided on both sides of the storage slot (801) near the opening, and a slidable cover plate (803) is provided in the slide slot (802).

Images