A system and method for standardizing plant chlorophyll fluorescence data

Abstract

This invention discloses a system and method for standardizing plant chlorophyll fluorescence data, relating to the fields of plant phenotypic analysis and precision agriculture. The system includes: a chlorophyll fluorescence measurement unit configured to acquire raw fluorescence data from measurement points on the plant canopy; a three-dimensional structural imaging unit configured to acquire a three-dimensional structural model containing the measurement points; and a processing unit configured to: spatially register the coordinates of the measurement points in the measurement system coordinate system with the coordinates in the three-dimensional structural model; determine the incident vector and observation vector based on a pre-calibrated offline measurement system; calculate the normal vector of the leaf surface at the measurement point from the three-dimensional structural model; and construct a standardized model based on the incident vector, observation vector, normal vector, and fill factor; and calculate the raw fluorescence data to generate standardized fluorescence data. This system eliminates interference caused by geometric relationships, obtaining comparable data.

Description

Technical Field

[0001] This invention relates to the fields of plant phenotyping and precision agriculture, and in particular to a system and method for standardizing plant chlorophyll fluorescence data. Background Technology

[0002] Non-contact fluorescence measurement technology for plant photosynthetic efficiency has become an important tool in plant physiological research. However, a core technical challenge in non-contact measurement of plant canopies with complex three-dimensional structures is the comparability of measured values.

[0003] The measured fluorescence signal intensity is not solely determined by the plant's intrinsic physiological state; it is also strongly influenced by the complete spatial geometry at the time of measurement. This geometry is defined by four key components: the direction of the laser source, the spatial orientation of the leaf surface being measured, the observation direction of the fluorescence detector, and the fill factor. Specifically:

[0004] 1. Light source and blade angle: According to Lambert's cosine law, the angle between the excitation beam and the normal to the blade surface (the angle of incidence) determines the effective energy received per unit area of ​​the blade. The larger the angle of incidence, the less energy is received, and the weaker the fluorescence signal.

[0005] 2. Blade and detector angle: When the detector observes the tilted blade surface from a specific angle, the total light flux received will be affected by the change in the projected area.

[0006] 3. Existing non-contact measurements typically assume that the excitation beam falls entirely on the leaf surface. However, in actual measurements, the excitation beam has a certain physical cross-section (e.g., 3-5 mm in diameter). When the measured object is a narrow leaf (such as pine needles or wheat tips), or when the measurement point is located near the leaf edge, the excitation beam often only partially covers the leaf, with the remaining light energy passing through the canopy or into the background. This "partial coverage" phenomenon leads to a significant decrease in the absolute fluorescence intensity received by the detector. Current techniques often cannot distinguish between this signal drop caused by insufficient physical coverage and signal drop caused by plant physiological stress, resulting in serious misjudgments.

[0007] This geometric relationship, consisting of the light source, target, and detector, can change in each measurement, making raw fluorescence data incomparable. A healthy leaf, simply due to an unfavorable measurement angle, may exhibit a lower fluorescence reading than a stressed leaf with an ideal angle, leading to serious misjudgments of plant health. Current technology has failed to provide an effective method for quantitatively standardizing fluorescence data to eliminate this geometric dependence.

[0008] Therefore, there is an urgent need in this field for a technical solution that can standardize fluorescence data measurements. Summary of the Invention

[0009] Based on the technical problems existing in the background art, the present invention proposes a system and method for standardizing plant chlorophyll fluorescence data, eliminating the interference caused by geometric relationships, thereby obtaining comparable data that can truly reflect the intrinsic physiological activities of plants.

[0010] This invention proposes a system for standardizing plant chlorophyll fluorescence data, comprising:

[0011] The chlorophyll fluorescence measurement unit is configured to acquire raw fluorescence data from measurement points on the plant canopy.

[0012] A three-dimensional structural imaging unit is configured to acquire a three-dimensional structural model including the measurement points;

[0013] The processing unit is configured to: spatially register the coordinates of the measurement point in the measurement system coordinate system with the coordinates in the three-dimensional structural model; determine the incident vector and observation vector according to the pre-calibrated offline measurement system; calculate the normal vector of the blade surface at the measurement point from the three-dimensional structural model; construct a standardized model based on the incident vector, observation vector, normal vector, and fill factor; calculate the original fluorescence data to generate standardized fluorescence data; wherein the fill factor is the area of ​​the overlapping region calculated between the projection model and the blade surface at the measurement point, and determines the proportion of the overlapping region area to the total cross-sectional area of ​​the beam.

[0014] Furthermore, the chlorophyll fluorescence measurement unit includes a laser light source and a fluorescence detector;

[0015] The excitation beam emitted by the laser source is guided and focused onto the measurement point on the plant canopy;

[0016] The fluorescence detector receives the fluorescence signal returned from the measurement point to obtain raw fluorescence data.

[0017] Furthermore, the chlorophyll fluorescence measurement unit employs a pulse amplitude modulation method for non-contact measurement.

[0018] Furthermore, the three-dimensional structure imaging unit and the chlorophyll fluorescence measurement unit are installed side by side and their coordinate systems are calibrated.

[0019] Furthermore, the three-dimensional structure imaging unit is a radar or a structured light camera.

[0020] Furthermore, in the processing unit, based on the pre-calibrated offline measurement system, the incident vector from the laser source of the chlorophyll fluorescence measurement unit to the measurement point, and the observation vector from the measurement point to the fluorescence detector of the chlorophyll fluorescence measurement unit are determined.

[0021] Furthermore, the angle between the incident vector and the normal vector is defined as the incident angle, and the angle between the observation vector and the normal vector is defined as the observation angle;

[0022] The correction factor of the standardized model is inversely proportional to the cosine of the incident angle, or...

[0023] The correction factor is inversely proportional to the cosine of the incident angle and directly proportional to the cosine of the observed angle.

[0024] Furthermore, the standardized fluorescence data is proportional to the correction factor.

[0025] Furthermore, the process of constructing the projection model is as follows: based on the three-dimensional structural model and the preset geometric parameters of the excitation beam, a projection model of the excitation beam in three-dimensional space is constructed;

[0026] The processing unit uses the reciprocal of the fill factor to perform compensation calculations on the original fluorescence data, thereby eliminating measurement errors caused by insufficient excitation area.

[0027] Furthermore, the standardized fluorescence data is proportional to the reciprocal of the fill factor.

[0028] A method for standardizing plant chlorophyll fluorescence data, comprising:

[0029] Raw fluorescence data and a three-dimensional structural model including the measurement points were obtained from measurement points on the plant canopy.

[0030] Spatial registration is performed between the coordinates of the measurement point in the measurement system coordinate system and the coordinates in the three-dimensional structural model.

[0031] The incident vector and observation vector are determined based on the pre-calibrated offline measurement system, and the normal vector of the blade surface at the measurement point is calculated from the three-dimensional structural model.

[0032] Based on the projection model of the excitation beam in three-dimensional space, calculate the area of ​​the overlapping region between the projection model and the blade surface at the measurement point, and determine the proportion of the overlapping region area to the total cross-sectional area of ​​the beam, which is denoted as the fill factor at the measurement point.

[0033] A standardized model based on the incident vector, observation vector, normal vector, and fill factor is constructed to calculate the original fluorescence data and generate standardized fluorescence data.

[0034] The advantages of the system and method for standardizing plant chlorophyll fluorescence data provided by this invention are: it provides a system that can eliminate measurement differences caused by changes in the relative geometric relationship between the light source, leaf, and detector; it can standardize fluorescence data measurements, eliminate interference caused by geometric relationships, and obtain comparable data that truly reflects the intrinsic physiological activity of plants, thereby enabling accurate quantitative comparison of plant chlorophyll fluorescence data. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the system structure of the present invention;

[0036] Figure 2 A schematic diagram defining the geometric parameters of a single measurement point;

[0037] Wherein, 1-laser source, 2-fluorescence detector, 3-three-dimensional structure imaging unit, 4-processing unit, 5-plant canopy, N-normal vector of leaf surface, L-incident vector, V-observation vector. - Angle of incidence - Observation angle. Detailed Implementation

[0038] The technical solution of the present invention will now be described in detail through specific embodiments. Many specific details are set forth in the following description to provide a thorough understanding of the invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0039] like Figure 1 and Figure 2 As shown, the present invention proposes a system for standardizing plant chlorophyll fluorescence data, comprising:

[0040] The chlorophyll fluorescence measurement unit is configured to acquire raw fluorescence data from measurement points on the plant canopy.

[0041] The three-dimensional structural imaging unit 3 is configured to acquire a three-dimensional structural model containing the measurement points;

[0042] Processing unit 4 is configured to: spatially register the coordinates of the measurement point in the measurement system coordinate system with the coordinates in the three-dimensional structural model; determine the incident vector and observation vector according to the pre-calibrated offline measurement system; calculate the normal vector of the blade surface at the measurement point from the three-dimensional structural model; construct a standardized model based on the incident vector, observation vector, normal vector, and fill factor; calculate the original fluorescence data to generate standardized fluorescence data; wherein the fill factor is the area of ​​the overlapping region calculated between the projection model and the blade surface at the measurement point, and determines the proportion of the overlapping region area to the total cross-sectional area of ​​the beam.

[0043] This embodiment provides a method to eliminate measurement differences caused by changes in the relative geometric relationship between the laser light source 1, the leaf, and the fluorescence detector 2. It can standardize fluorescence data measurements, eliminate interference from geometric relationships, and thus obtain comparable data that truly reflects the intrinsic physiological activity of plants. This enables precise quantitative comparison of plant chlorophyll fluorescence data.

[0044] It should also be noted that the standardized model in this embodiment can be constructed based on the incident vector, normal vector, and fill factor, or it can be based on the incident vector, observation vector, normal vector, and fill factor. A standardized model including the observation vector is the most thorough correction method.

[0045] The beneficial effects of this embodiment are as follows:

[0046] 1. Improved data comparability: By standardizing the readings at each measurement point using complete geometric relationships, systematic differences introduced by variations in leaf orientation, light source, and detector relative position are eliminated, enabling measurement results to more accurately reflect the plant's internal physiological state.

[0047] 2. Enhanced measurement accuracy: Standardized fluorescence data makes it possible to make precise quantitative comparisons between different leaves and different plants within complex canopies, avoiding misjudgments caused by geometric factors.

[0048] 3. Enhanced application value: This invention elevates non-contact fluorescence measurement from a tool dependent on ideal measurement conditions to a scientific tool capable of reliable quantitative analysis in complex three-dimensional scenarios.

[0049] In this embodiment, the laser light source 1, fluorescence detector 2, three-dimensional structure imaging unit 3, and processing unit 4 are integrated into a measuring head, which is used to measure the measurement points on the plant canopy 5.

[0050] The chlorophyll fluorescence measurement unit in the measuring head employs a non-contact measurement method. The excitation beam emitted by the laser source 1 is guided and focused onto the measurement point on the plant canopy 5. When the laser irradiates the leaf, the light energy is absorbed by the chlorophyll molecules within the leaf, causing electrons to transition to higher energy levels. Electrons in high energy levels are unstable and quickly return to the ground state by releasing energy. A portion of this energy is released in the form of light, which is chlorophyll fluorescence. A key characteristic is that the wavelength of the fluorescence is always longer (but with lower energy) than the wavelength of the excitation light. For example, when excited by a 650nm red laser, chlorophyll will emit far-red and red fluorescence with center wavelengths around 730nm and 680nm, respectively.

[0051] Fluorescence detector 2 is responsible for receiving the fluorescence signal returned from the measurement point. This unit can use pulse amplitude modulation (PAM) technology to measure various fluorescence parameters such as Fo, Fm, Fs, and Fm.

[0052] The three-dimensional structure imaging unit and the chlorophyll fluorescence measurement unit are installed side by side and have undergone rigorous coordinate system calibration. In this embodiment, the three-dimensional structure imaging unit can be a lidar or a structured light camera. It can quickly acquire high-density three-dimensional point cloud data of the plant canopy surface, thereby constructing a detailed three-dimensional structure model.

[0053] In this embodiment, the processing unit 4 is a computer connected to the measuring head via a data cable or wirelessly. The processing unit 4 determines, based on a pre-calibrated offline measuring system, the incident vector from the laser source of the chlorophyll fluorescence measuring unit to the measuring point, the observation vector from the measuring point to the fluorescence detector of the chlorophyll fluorescence measuring unit, and the fill factor.

[0054] In this embodiment, a fill factor is introduced into the standardized model. The processing unit is configured to: construct a projection model of the excitation beam in three-dimensional space based on the three-dimensional structural model and the preset excitation beam geometric parameters; calculate the area of ​​the overlapping region between the projection model and the blade surface at the measurement point, and determine the proportion of the overlapping region area to the total cross-sectional area of ​​the beam, which is denoted as the fill factor; the processing unit uses the reciprocal of the fill factor to perform compensation calculations on the original fluorescence data.

[0055] It should be noted that the purpose of spatially registering the coordinates of the measurement points in the measurement system coordinate system with the coordinates in the three-dimensional structural model is to establish a unified geometric reference system in order to accurately calculate the geometric relationship between light and the leaf surface, thereby accurately physical correcting the chlorophyll fluorescence data.

[0056] The specific purposes of spatial registration are (a1) to (a4):

[0057] (a1) Determining the precise leaf orientation: The three-dimensional structural model provides three-dimensional spatial information for each measurement point in the plant canopy. Through registration, the system can accurately determine which leaf corresponds to measurement point A in the three-dimensional model, and the normal vector of that leaf in space (i.e., the direction vector perpendicular to the leaf surface). The leaf orientation (normal vector) is a key factor affecting illumination and observation angle.

[0058] (a2) Calculate the true incident and observation angles: The intensity of the fluorescence signal strongly depends on the incident angle of the light and the observation angle of the detector. The incident angle is the angle at which the excitation beam illuminates the blade surface, and the observation angle is the angle at which the fluorescence detector "sees" the blade surface. Only by knowing the precise three-dimensional position of the measurement point and the blade normal vector can the accurate incident and observation vectors be calculated based on the positions of the laser source and the detector, and thus the angles between them and the normal vector can be obtained.

[0059] (a3) Determining the effective coverage area of ​​the light spot: This is the basis for identifying the "edge effect." By spatially registering the projection path of the laser beam with the fine three-dimensional grid of the plant canopy, the system can perform three-dimensional Boolean operations to accurately determine whether the excitation light spot falls completely on the leaf surface. For measurement points located at leaf edges, leaf tips, or holes, registration allows the system to calculate the actual projected area of ​​the light spot on the leaf (i.e., the fill factor). This allows for the identification of signal loss caused by "misalignment" or "light leakage".

[0060] (a4) Eliminating variations caused by geometric effects: Under natural conditions, even the same leaf will show different fluorescence values ​​if it is illuminated by laser at different angles or observed from different angles. This difference is not entirely due to changes in the plant's physiological state, but is largely caused by geometric effects. Spatial registration is the foundation for subsequent standardized model calculations. The purpose of this model is to eliminate these geometric effects, so that fluorescence data can more accurately reflect the plant's physiological state, thereby making data measured at different times, locations, and angles comparable.

[0061] It is worth noting that in this embodiment, standardized fluorescence data is obtained by multiplying the reciprocals of the correction factor and the filler factor by the original fluorescence data. The fundamental reason for this is that the original fluorescence data is a "mixed signal" influenced by multiple factors; it contains both desired plant physiological information and unwanted measurement geometric effects. These interferences mainly come from three aspects: (b1) and (b3).

[0062] (b1) Incident angle effect (difference in light energy density);

[0063] When a fixed power laser beam is at a vertical angle ( When the light spot is irradiated, the area of ​​the light spot is the smallest, and the energy (energy density) received per unit area of ​​the leaf is the highest.

[0064] When the laser is at an angle (e.g.) When the laser beam is irradiated on the same leaf, the light spot is elongated into an ellipse, increasing its area. Although the total laser power remains constant, the energy received per unit area of ​​the leaf decreases (compared to...). (Proportional)

[0065] Therefore, even if the physiological state of the leaves is exactly the same, the fluorescence signal excited by oblique illumination will be weaker than that under vertical illumination. This difference is not caused by physiological reasons, but by purely geometric and physical reasons.

[0066] (b2) Observation angle effect (difference in detector field of view coverage area);

[0067] Similarly, the detector's observation angle ( Changes in the angle of inclination ( ) can also introduce geometric interference. When observing the blade surface at >0°, the actual physical area of ​​the blade covered by the detector's field of view (FOV) or a single pixel will increase, and the area change relationship is approximately as follows: ,in, This represents the physical area actually covered by the detector's field of view on the blade surface. It is the projected area of ​​the detector's field of view on a plane perpendicular to the observation direction.

[0068] This means that during tilted observations, a single reading actually aggregates fluorescence energy over a larger physical area on the leaf. Without standardization, this artificially inflated signal (relative to the vertical observation unit area) due to the increased observation area could be mistaken for enhanced plant physiological activity. Therefore, it is necessary to introduce a standardization mechanism. The relevant correction factor normalizes the data from different observation angles to the fluorescence intensity per unit physical area, thereby eliminating the measurement bias caused by geometric projection.

[0069] (b3) Spot edge effect (incomplete filling of the excitation spot);

[0070] Existing measurements typically assume that the excitation spot falls entirely on the leaf surface. However, when measuring narrow leaves (such as needles), leaf edges, or damaged leaves (such as areas with insect holes), the laser spot often only partially covers the leaf (i.e., the fill factor). <1), the remaining light energy is directed to the background. This will directly lead to a significant proportional decrease in the total amount of fluorescence signal received by the detector. This signal attenuation is entirely determined by the "overlap ratio between the light spot and the leaf" and is unrelated to the plant's photosynthetic efficiency; failure to eliminate it will result in serious misjudgment.

[0071] According to (b1) to (b3), the following drawbacks exist if the raw fluorescence data are not corrected: In natural plant canopies, leaf orientation (normal vector) varies greatly, and the light angle, observation angle, and effective coverage ratio of the excitation spot on the leaf also differ. Therefore, directly comparing raw fluorescence data measured at two different angles is like comparing leaf temperature measured under shade and under direct sunlight—lacking comparability and leading to misleading conclusions.

[0072] The purpose of calibration using a standardized model is to transform incomparable "raw fluorescence signals" measured under different geometric conditions into a comparable "standardized fluorescence signal" that only reflects the intrinsic physiological characteristics of the leaf.

[0073] For the first type of correction factor This technology is specifically designed to compensate for differences in light energy density caused by the angle of incidence; it "equivalents" obliquely incident light to perpendicular incident light. If the energy density is only half that of perpendicular light when incident at a 60° angle (because cos(60°) = 0.5), then the correction factor C = 1 / 0.5 = 2, which will double the measured fluorescence signal, thereby offsetting the signal attenuation caused by the oblique illumination. This achieves "light geometric normalization." Now, regardless of the angle from which the laser shines, the standardized fluorescence data obtained are as if measured under "ideal perpendicular illumination," making data from different times and different canopy locations comparable in terms of illumination conditions.

[0074] For the second type of correction factor Simultaneously compensates for the incident angle effect and the observation angle effect, among which, Used to correct incident energy density, as described above. This is used to correct for differences in the physical area covered by the detector's field of view. When making oblique observations, the actual physical area projected onto the blade surface by the detector's field of view (or a single pixel) increases (approximately the same as when making vertical observations). This means that the raw data actually aggregates fluorescence energy over a larger physical area on the leaf, resulting in a higher reading relative to the unit area. By multiplying by... (Value ≤ 1) can offset the area gain effect introduced by the field of view expansion, and accurately normalize the data to the fluorescence intensity per unit physical area.

[0075] The fill factor for the laser spot edge effect specifically compensates for signal loss caused by the laser spot not completely covering the leaf (e.g., hitting the leaf edge or holes). When edge overflow occurs, the effective excitation area is smaller than the theoretical spot cross-section, resulting in a decrease in the total fluorescence signal. For example, if only half of the laser spot falls on the leaf (…), the signal loss is significant. = 0.5), the original signal will be halved; at this point, multiplying by 1 / 0.5 = 2 will bring the signal back to the level of "full-spot excitation". This achieves "excitation effectiveness normalization".

[0076] That is, when edge overflow of the light spot is detected ( When <1), the reciprocal of this factor is used to compensate for the original fluorescence data, thereby eliminating the measurement error caused by insufficient excitation area. This solves the problem in existing methods where the signal decrease caused by insufficient physical coverage is indistinguishable from that caused by plant physiological stress, leading to serious misjudgments.

[0077] The second type of end-to-end normalization, combining correction and filler factors, is the most thorough correction method. It generates standardized fluorescence data that simulates the ideal geometry of "vertical illumination and vertical observation." This makes the data comparable not only at different illumination angles but also at different observation angles (e.g., using detectors at different installation angles or observing from the side of the canopy), different leaf morphologies (e.g., needle-like versus broad-leaved), and different leaf integrity (e.g., insect-eaten versus healthy leaves), ensuring that the measurement results truly reflect the plant's physiological state rather than geometric projection errors.

[0078] Using the above correction factor ( or ) and fill factor ( The advantages of performing data correction are (c1) to (c3):

[0079] (c1) Improved data accuracy and reliability: It eliminates noise introduced by non-physiological factors (angle), allowing fluorescence data to more accurately reflect the photosynthetic physiological state of plants (such as the efficiency of photosystem II, electron transport rate, etc.).

[0080] (c2) Achieve data comparability:

[0081] Spatial comparability: It allows for fair comparison of the physiological state of leaves in different orientations (e.g., top horizontal leaves and side vertical leaves), different leaf morphologies (e.g., needle leaves and broad leaves), and different degrees of integrity (e.g., insect-eaten leaves and healthy leaves) in the canopy;

[0082] Time comparability: It can accurately track the physiological dynamics of the same leaf or the same canopy at different points in time (as the angle of the sun changes continuously);

[0083] System comparability: Enables cross-comparison and integrated analysis of data obtained from different laboratories and different measuring devices (as long as they follow the same standardization principles).

[0084] (c3) Enabling High-Throughput Phenotyping: In modern agriculture and high-throughput plant phenotyping research, mobile platforms or drones equipped with sensors are often used to scan a large number of plants. The leaf orientation of plants is random, and standardized models can ensure that the fluorescence features extracted from massive amounts of data are purely physiological differences, greatly improving the efficiency and accuracy of screening superior varieties.

[0085] As an example, the standardized method of this embodiment is as follows:

[0086] Preparation phase: System calibration;

[0087] Before performing the measurement steps, a one-time offline calibration of the measurement system is required to establish the geometric relationships of each unit in a unified spatial coordinate system. This calibration process aims to calculate the internal parameters of each sensor and the external parameters of their relative positions. The specific implementation steps are (d1) to (d3):

[0088] (d1) Establishing the world coordinate system and setting calibration references:

[0089] Introduce a planar calibration plate with known high-precision geometric features (e.g., a checkerboard calibration plate or a dot array calibration plate) and place it within the system's field of view.

[0090] The calibration plate is scanned using the 3D structural imaging unit 3. The processing unit 4 identifies the point cloud plane or feature corner points of the calibration plate and constructs a 3D Cartesian coordinate system, which is defined as the world coordinate system (C). World For example, the calibration plate plane can be defined as the Z=0 plane, and a corner point of the calibration plate can be defined as the origin (0,0,0).

[0091] (d2) Calibrate the fluorescence detector:

[0092] Two-dimensional images of the calibration plate were acquired using fluorescence detector 2.

[0093] Processing unit 4 extracts the pixel coordinates of feature points (such as checkerboard corners) in the image. Because these feature points are in the world coordinate system The three-dimensional coordinates (X, Y, Z) are known (based on the physical dimensions of the calibration plate). Processing unit 4 uses the PnP (Perspective-n-Point) algorithm or a similar camera calibration algorithm to calculate the intrinsic parameter matrix (including focal length, principal point coordinates, and distortion coefficients) and extrinsic parameter matrix of the fluorescence detector 2. (Includes rotation matrix R and translation vector T).

[0094] Based on the above parameters, establish a coordinate system from the detector pixel coordinate system to the world coordinate system. The back projection model determines the geometric reference of the observation vector V (i.e., the direction from the camera optical center to the spatial ray corresponding to any pixel).

[0095] (d3) Calibrate the laser optical path;

[0096] Since the laser source 1 is fixedly installed in this embodiment, its optical path is a fixed straight line in space. To determine the equation of this straight line, a "multi-position spot fitting method" is used:

[0097] Location A Acquisition: Laser source 1 is activated, projecting its beam onto the calibration plate to form a light spot. Fluorescence detector 2 captures the image, and processing unit 4 extracts the pixel coordinates of the center of the light spot. Combining the known detector parameters from step (d2) with the current calibration plate in the world coordinate system The absolute coordinates of the light spot in three-dimensional space are calculated using the plane equation. .

[0098] Location B Acquisition: Move the calibration plate back and forth along the optical axis to another position, and repeat the above process to obtain the three-dimensional coordinates of the laser spot at another depth. .

[0099] Ray fitting: Processing unit 4 utilizes and By performing spatial straight line fitting, the laser beam in the world coordinate system can be obtained. Three-dimensional ray equations (in Let be a point on the straight line. It is a direction vector. From Along direction The distance traveled can be obtained from the three-dimensional coordinates. This equation precisely describes the incident vector. The physical path.

[0100] After the above calibration is completed, the system will output the calibration parameters (detector intrinsic and extrinsic parameters matrix). Laser ray equation The solidified storage is then placed in processing unit 4, and subsequently proceeds to the following real-time measurement process:

[0101] S1. Synchronous data acquisition;

[0102] When the system measures a point on the target plant canopy 5, the chlorophyll fluorescence measurement unit and the three-dimensional structure imaging unit 3 work synchronously or sequentially within a very short time interval to acquire the original fluorescence data of that point and the three-dimensional point cloud model of the entire canopy, respectively.

[0103] S2, Data Spatial Registration;

[0104] Since the positional relationships of each unit within the measuring head have been pre-calibrated, processing unit 4 can accurately map the fluorescence measurement points to a corresponding spatial coordinate point in the three-dimensional point cloud model through coordinate transformation. superior.

[0105] S3. Complete geometric relationships are determined;

[0106] After registration is completed, for this fluorescence measurement point, processing unit 4 will perform the following calculations, such as... Figure 2 As shown:

[0107] First, based on the pre-calibrated offline measurement system, the incident vector L from the laser source 1 to the measurement point and the observation vector V from the measurement point to the fluorescence detector 2 are determined.

[0108] Secondly, based on the position of the measurement point and its neighboring points in the three-dimensional model, the normal vector N of the blade micro-surface where the measurement point is located is calculated.

[0109] Finally, based on these three vectors, key geometric parameters, such as the angle of incidence, are calculated. (The angle between the incident vector L and the normal vector N) and the observation angle (The angle between the observation vector V and the normal vector N).

[0110] S4. Calculation of light spot edge effect;

[0111] Because the laser beam has a certain physical cross-section (e.g., a diameter of 3-5 mm), when the measurement point is located at the blade edge, near a hole, or on a slender blade, the excitation spot may not completely cover the blade surface. To eliminate the error caused by this "off-center" or "partial hit" error, processing unit 4 performs the following calculations:

[0112] Based on the laser beam path straight line calibrated in step (d3) and the laser beam diameter, a cylindrical model representing the laser beam is generated in three-dimensional space. This beam model is then subjected to a spatial Boolean intersection operation with the blade's three-dimensional fine mesh to calculate the effective area of ​​the laser beam actually projected onto the blade surface. Define the fill factor. The effective area is the sum of the theoretical total cross-sectional area of ​​the beam at the measurement distance. The ratio:

[0113] .

[0114] When the light spot falls entirely inside the leaf. = 1; 0 < 1 when edge overflow occurs. <1.

[0115] S5. Apply standardized models;

[0116] Processing unit 4 applies a standardized model to the raw fluorescence data. Calculations are performed. This normalized model aims to compensate for signal strength variations caused by the complete measurement geometry. A preferred normalized model is... It can be represented as:

[0117] ;

[0118] in, It is a standardized function, namely the correction factor, which is at least the angle of incidence. functions, such as That is, the correction factor is inversely proportional to the cosine of the incident angle to compensate for the difference in effective light energy density caused by leaf tilt. In a more precise standardized model, It can also be the observation angle. A function to compensate for the projection effect from the detector's perspective. A more complete physical model is... The correction factor is inversely proportional to the cosine of the incident angle and directly proportional to the cosine of the observed angle. Fill factor It is used to compensate for signal loss caused by the light spot not completely covering the leaf.

[0119] S6. Generate standardized output;

[0120] After calculation, the system outputs standardized fluorescence data. This data can be directly compared with other standardized measurement data, or assigned as an attribute to corresponding points in the 3D structural model for subsequent analysis.

[0121] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A system for standardizing plant chlorophyll fluorescence data, characterized in that, include: The chlorophyll fluorescence measurement unit is configured to acquire raw fluorescence data from measurement points on the plant canopy. A three-dimensional structural imaging unit is configured to acquire a three-dimensional structural model including the measurement points; The processing unit is configured to: spatially register the coordinates of the measurement point in the measurement system coordinate system with the coordinates in the three-dimensional structural model; determine the incident vector and observation vector according to the pre-calibrated offline measurement system; calculate the normal vector of the blade surface at the measurement point from the three-dimensional structural model; construct a standardized model based on the incident vector, observation vector, normal vector, and fill factor; calculate the original fluorescence data to generate standardized fluorescence data; wherein the fill factor is the area of ​​the overlapping region calculated between the projection model and the blade surface at the measurement point, and determines the proportion of the overlapping region area to the total cross-sectional area of ​​the beam.

2. The system according to claim 1, characterized in that, The chlorophyll fluorescence measurement unit includes a laser light source and a fluorescence detector; The excitation beam emitted by the laser source is guided and focused onto the measurement point on the plant canopy; The fluorescence detector receives the fluorescence signal returned from the measurement point to obtain raw fluorescence data.

3. The system according to claim 1, characterized in that, The chlorophyll fluorescence measurement unit uses a pulse amplitude modulation method for non-contact measurement.

4. The system according to claim 1, characterized in that, The three-dimensional structural imaging unit and the chlorophyll fluorescence measurement unit are installed side by side and their coordinate systems are calibrated.

5. The system according to claim 1, characterized in that, The three-dimensional structure imaging unit is a radar or a structured light camera.

6. The system according to claim 1, characterized in that, In the processing unit, based on the pre-calibrated offline measurement system, the incident vector from the laser source of the chlorophyll fluorescence measurement unit to the measurement point, and the observation vector from the measurement point to the fluorescence detector of the chlorophyll fluorescence measurement unit are determined.

7. The system according to claim 1, characterized in that, The angle between the incident vector and the normal vector is defined as the incident angle, and the angle between the observation vector and the normal vector is defined as the observation angle. The correction factor of the standardized model is inversely proportional to the cosine of the incident angle, or... The correction factor is inversely proportional to the cosine of the incident angle and directly proportional to the cosine of the observed angle.

8. The system according to claim 7, characterized in that, The standardized fluorescence data is proportional to the correction factor.

9. The system according to claim 1, characterized in that, The process of constructing the projection model is as follows: based on the three-dimensional structural model and the preset geometric parameters of the excitation beam, a projection model of the excitation beam in three-dimensional space is constructed. The processing unit uses the reciprocal of the fill factor to perform compensation calculations on the original fluorescence data.

10. A method for standardizing plant chlorophyll fluorescence data, characterized in that, include: Raw fluorescence data and a three-dimensional structural model including the measurement points were obtained from measurement points on the plant canopy. Spatial registration is performed between the coordinates of the measurement point in the measurement system coordinate system and the coordinates in the three-dimensional structural model. The incident vector and observation vector are determined based on the pre-calibrated offline measurement system, and the normal vector of the blade surface at the measurement point is calculated from the three-dimensional structural model. Based on the projection model of the excitation beam in three-dimensional space, calculate the area of ​​the overlapping region between the projection model and the blade surface at the measurement point, and determine the proportion of the overlapping region area to the total cross-sectional area of ​​the beam, which is denoted as the fill factor at the measurement point. A standardized model based on the incident vector, observation vector, normal vector, and fill factor is constructed to calculate the original fluorescence data and generate standardized fluorescence data.

Images