A strawberry chilling injury risk warning and stress detection method and device based on phenotypic fusion

Abstract

The present invention discloses a strawberry chilling injury risk warning and stress detection method and device based on phenotypic fusion in the field of plant chilling injury detection technology. The method comprises extracting phenotypic spatial distribution characteristics based on the fluorescence parameter phenotypic image and hyperspectral phenotypic image of the leaf to be tested whose internal electrons are in the ground state; calculating the correlation between the leaf pixels of the phenotypic spatial distribution characteristics and the physiological indicators of the leaf to be tested, obtaining the spatial distribution characteristic parameters and inputting them into the XGBoost integrated learning model, predicting the photosynthetic physiological potential index and the relative negative accumulated temperature of the leaf to be tested, and calculating the chilling injury risk index result of the leaf to be tested; determining the strawberry low temperature chilling injury warning and stress degree according to the chilling injury risk index and the photosynthetic physiological potential index. The present invention fully integrates the spatial variation characteristic information in a specific direction with the phenotypic data, and realizes the rapid, non-destructive and quantitative determination of the chilling injury risk warning level and stress degree of the strawberry plant from the leaf scale.

Description

Technical Field

[0001] The invention relates to a strawberry chilling injury risk early warning and stress detection method and device based on phenotype fusion, belonging to the technical field of plant chilling injury detection. Background Art

[0002] Since the 21st century, with the rapid development of optical sensor technology and the widespread popularization of computer technology, phenotypic imaging technology has made significant progress in the field of plant physiological information monitoring. Among them, hyperspectral imaging technology and chlorophyll fluorescence imaging technology have become two important means of plant phenotypic research. By capturing the reflection characteristics of plant leaves in different bands, hyperspectral imaging technology can accurately reflect the changes in leaf structure, pigment content and element content, and then be applied to the inversion of photosynthetic physiological parameters. Chlorophyll fluorescence imaging technology uses the distribution relationship of light energy in the process of plant photosynthesis to monitor photosynthetic physiological changes in real time and sensitively, especially in the detection of early stresses such as drought. In addition, some studies have also attempted to integrate chlorophyll fluorescence dynamics spectral imaging with hyperspectral imaging to obtain more comprehensive plant physiological information.

[0003] Although the existing technology has made many advances in plant phenotypic imaging, there are still some obvious defects. First, most of the current research or inventions are based on two-dimensional domains, and do not fully consider the spatial variation information in specific directions such as parallel main veins and vertical main veins, which limits the in-depth mining and utilization of plant physiological information. Secondly, most studies only use a single phenotypic data source, or the advantages of multi-source data fusion are not fully utilized, which affects the accuracy and comprehensiveness of the monitoring results. In addition, the poor portability of some device platforms limits their application in the wild or in complex environments. Finally, the current technical solutions generally lack quantitative early warning solutions for plant stress potential based on phenotypic fusion, and cannot provide strong support for the early detection and response of plant stress. These defects and deficiencies are problems that need to be urgently solved in the current field of plant stress monitoring.

[0004] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgement or any form of suggestion that the information constitutes the prior art already known to ordinary technicians in this field. Summary of the invention

[0005] The technical problem to be solved by the present invention is: how to fully integrate the spatial variation characteristic information in a specific direction with the phenotypic data to quickly, non-destructively and quantitatively determine the cold injury risk warning and stress degree of strawberry plants at the leaf scale.

[0006] In order to solve the above technical problems, the present invention is implemented by adopting the following technical solutions.

[0007] In a first aspect, the present invention provides a strawberry chilling injury risk warning and stress detection method based on phenotypic fusion.

[0008] The leaf to be tested is placed under dark adaptation conditions so that the electrons inside the leaf to be tested are in the ground state;

[0009] Extracting phenotypic spatial distribution characteristics based on the fluorescence parameter phenotypic image and the hyperspectral phenotypic image of the leaf to be tested;

[0010] The spatial distribution characteristic parameters are obtained by calculating the correlation between the leaf pixels of the phenotypic spatial distribution characteristics and the physiological indexes of the leaves to be measured;

[0011] According to the spatial distribution characteristic parameters and the XGBoost integrated learning model, phenotypic fusion is performed to predict the photosynthetic physiological potential index and relative negative accumulated temperature of the leaves to be tested;

[0012] Calculating the chilling injury risk index result of the leaf to be tested according to the photosynthetic physiological potential index and the relative negative accumulated temperature;

[0013] Determine the strawberry low temperature chilling injury warning and stress degree according to the chilling injury risk index and the photosynthetic physiological potential index;

[0014] The phenotypic spatial distribution characteristics include a one-dimensional line domain parallel to the main vein direction of the leaf to be measured, a one-dimensional line domain perpendicular to the main vein direction of the leaf to be measured, and a two-dimensional surface domain of the leaf to be measured.

[0015] Furthermore, based on the fluorescence parameter phenotype image and the hyperspectral phenotype image of the leaf to be tested, phenotypic spatial distribution characteristics are extracted, including:

[0016] Based on the chlorophyll fluorescence imaging measurement method, the basic fluorescence parameter phenotype image of the leaf to be measured and the chlorophyll fluorescence parameters and chlorophyll fluorescence parameters are collected and calculated to obtain the fluorescence parameter phenotype image of the leaf to be measured; the fluorescence parameter phenotype image includes a photochemical quenching coefficient image, a non-photochemical quenching coefficient image, a photosynthetic system PSⅡ actual fluorescence quantum efficiency image and a non-photochemical quenching fluorescence quantum yield image;

[0017] Based on the spectral index imaging measurement method, the reflectance images of the three bands of blue, green and near-infrared of the leaves to be measured are collected to construct a hyperspectral phenotypic image of the leaves to be measured; the hyperspectral phenotypic image includes a blue light near-infrared normalized difference hyperspectral phenotypic image and a green light near-infrared normalized difference hyperspectral phenotypic image;

[0018] Extracting a one-dimensional line domain parallel to the main vein direction of the leaf to be tested from the photochemical quenching coefficient image, the non-photochemical quenching coefficient image, and the photosynthetic system PSⅡ actual fluorescence quantum efficiency image of the leaf to be tested respectively;

[0019] Extracting a one-dimensional line domain perpendicular to the main vein direction of the leaf to be tested from the green light near-infrared normalized difference hyperspectral phenotype image of the leaf to be tested;

[0020] The two-dimensional surface area of ​​the leaf to be tested is extracted from the non-photochemical quenching fluorescence quantum yield image of the leaf to be tested and the blue light near-infrared normalized difference hyperspectral phenotype image respectively.

[0021] Furthermore, the spatial distribution characteristic parameters are obtained by calculating the correlation between the leaf pixels of the phenotypic spatial distribution characteristics and the physiological indicators of the leaves to be measured, including:

[0022] Respectively calculating the texture characteristic parameters of the leaf pixels of the fluorescence parameter phenotype image and the hyperspectral phenotype image of the leaf to be tested under the phenotypic spatial distribution characteristics;

[0023] According to the correlation between the texture characteristic parameters and the maximum net photosynthetic rate, leaf relative conductivity and chlorophyll content, the spatial distribution characteristic parameters are obtained; wherein the maximum net photosynthetic rate, leaf relative conductivity and chlorophyll content are physiological indicators of the leaves to be tested.

[0024] Further, the texture feature parameters include contrast, angular second-order moment, texture entropy and moment of inertia;

[0025] The calculation expression of the contrast is:

[0026] ;

[0027] Where CON represents the contrast of the phenotypic spatial distribution characteristics, i represents the pixel index of the i-th grayscale value of the phenotypic spatial distribution characteristics, j represents the pixel index of the j-th grayscale value of the phenotypic spatial distribution characteristics, and P(i,j) represents the probability value of the pixel index of the grayscale value (i,j) in the gray level co-occurrence matrix GLCM of the phenotypic spatial distribution characteristics;

[0028] The calculation expression of the angular second-order moment is:

[0029] ;

[0030] In the formula, ASM represents the angular second moment of the phenotypic spatial distribution characteristics;

[0031] The calculation expression of the texture entropy is:

[0032] ;

[0033] Where TENT represents the texture entropy of the phenotypic spatial distribution characteristics, It represents the logarithmic function with base 2. ε has no practical meaning and is a very small constant to ensure that the logarithmic operation is not 0. Its value is 10. -10 ;

[0034] The calculation expression of the moment of inertia is:

[0035] ;

[0036] Where INEM represents the moment of inertia of the spatial distribution characteristics of the phenotype.

[0037] Furthermore, the method for determining strawberry low temperature chilling injury warning and stress degree according to the chilling injury risk index and the photosynthetic physiological potential index comprises:

[0038] When strawberry low temperature chilling damage has not occurred, the strawberry low temperature chilling damage warning is determined according to the chilling damage risk index result, including:

[0039] When the chilling injury risk index result is ≥3, it is determined that there is no chilling injury to strawberries;

[0040] When 2≤chilling injury risk index result<3, strawberry low temperature chilling injury blue warning is determined;

[0041] When 1≤chilling injury risk index result<2, a yellow warning of strawberry low temperature chilling injury is determined;

[0042] When 0≤chilling injury risk index result<1, strawberry low temperature chilling injury red warning is determined;

[0043] When the chilling injury risk index result is less than 0, it is determined that strawberry chilling injury has occurred;

[0044] When strawberry low temperature chilling injury has occurred, the degree of chilling injury of strawberry plants is determined according to the photosynthetic physiological potential index, including:

[0045] When the photosynthetic physiological potential index is ≥3.5, it is determined that the strawberry plants have no chilling damage;

[0046] When the photosynthetic physiological potential index is 2.8≤<3.5, the strawberry plants are judged to be slightly damaged by chilling injury;

[0047] When 1.5≤photosynthetic physiological potential index<2.8, the strawberry plants were judged to be moderately damaged by chilling injury;

[0048] When the photosynthetic physiological potential index is ≤1.5, the strawberry plants are judged to have suffered severe chilling damage.

[0049] In a second aspect, the present invention provides a strawberry chilling injury risk warning and stress device based on phenotypic fusion, including a device body, a dark adaptation chamber detachably connected to the device body, and a collector dark adaptation chamber connection module, for implementing the method described in the first aspect;

[0050] Wherein, the device body comprises:

[0051] The acquisition and transmission interface 1 is connected to the acquisition and analysis storage module 3 and is used to transmit the collected data of the fluorescence parameter phenotype image and the hyperspectral phenotype image of the leaf to be tested to the USB communication interface of the external device;

[0052] The power interface 2 is a plug-in interface connected to the power module 4 and is used to transmit power;

[0053] The acquisition analysis storage module 3 is connected to the acquisition transmission interface 1 at the top and respectively connected to the imaging acquisition module 7 and the artificial light source 9 at the bottom, and is used to control the program for acquiring the fluorescence parameter phenotype image and the hyperspectral phenotype image, and temporarily store the data of the acquired fluorescence parameter phenotype image and the hyperspectral phenotype image of the leaf to be tested;

[0054] The power module 4 is connected to the power interface 2 at the top and connected to the acquisition analysis storage module 3, the imaging acquisition module 7 and the artificial light source 9 at the bottom, for providing power;

[0055] The main body shell 6 is connected to the dark adaptation chamber at the bottom and is used to accommodate and protect the device body;

[0056] The imaging acquisition module 7 includes a multispectral CMOS image sensor and a filter wheel, which is connected to the acquisition analysis storage module 3 at the top and the imaging lens 8 at the bottom, and is connected to the power module 4 through a power line; it is used to acquire fluorescence parameter phenotype images and hyperspectral phenotype images;

[0057] An imaging lens 8, connected to the imaging acquisition module 7 at the top, for adjusting the focal length and imaging magnification;

[0058] The artificial light source 9 includes a red light unit, a blue light unit, a white light unit, and a far-infrared unit, wherein the red light unit is used as a measuring light source; the blue light unit is used as a saturated pulse light source; the white light unit is used as an actinic light source; and the far-infrared unit is used as a measuring light source for chlorophyll fluorescence parameters. The artificial light source 9 is respectively connected to the collection and analysis storage module 3 and the power supply module 4, and is in a concentric circular structure with a hollow interior. The imaging lens 8 is installed in the hollow part of the concentric circle to ensure that the light can be evenly irradiated on the leaf to be measured, and the imaging lens 8 is used for imaging collection;

[0059] The collector dark adaptation chamber connection module 10 is located between the bottom of the main housing 6 and the top of the upper opening of the dark adaptation chamber, and is raised at the edge of the dark adaptation chamber, so as to be magnetically connected to the device body when the device body is directly inserted into the dark adaptation chamber;

[0060] The dark adaptation chamber is used to place the leaf to be tested completely or partially under dark adaptation conditions so that the electrons inside the leaf to be tested are in the ground state.

[0061] Furthermore, expansion bayonet ports 5 are distributed on four sides of the upper portion of the main housing 6 for connecting external auxiliary equipment.

[0062] Furthermore, the red light unit adopts a red light source of a LED lamp bead array with a wavelength of 650nm; the blue light unit adopts a blue light source of a LED lamp bead array with a wavelength of 450nm; the white light unit adopts a white light source of a LED lamp bead array with a wavelength of 400~700nm; and the far-infrared unit adopts a far-infrared light source of a LED lamp bead array with a wavelength of 760nm.

[0063] Furthermore, the dark adaptation chamber comprises:

[0064] The dark adaptation chamber main body shell 11 is used to accommodate and protect the internal space to form a dark adaptation environment;

[0065] Dark adaptation shades 12 are located on both sides of the top of the dark adaptation chamber and are hard multi-piece translational structures used to block external light;

[0066] The dark-adaptation shade curtain exposed handle 13 is connected to the lowest shade curtain in the dark-adaptation shade curtain 12, exposed outside the dark-adaptation chamber, and used to operate the opening and closing of the dark-adaptation shade curtain;

[0067] The dark-adaptive shade curtain moving guide rail 14 is connected to the dark-adaptive shade curtain 12 and the exposed handle 13 of the dark-adaptive shade curtain, and is used to fix the moving track of the exposed handle 13 of the dark-adaptive shade curtain so that the dark-adaptive shade curtain is evenly unfolded.

[0068] Furthermore, the dark adaptation chamber further comprises:

[0069] The blade holder 15 is an axis-rotating transparent plastic pressing sheet, located on both sides of the upper edge of the dark adaptation chamber base 16, and is used to lay and fix the blade to be tested;

[0070] The dark adaptation chamber base 16 is detachably connected to the bottom of the dark adaptation chamber main body shell 11 by magnetic attraction, and is used to support and fix the dark adaptation chamber main body shell 11;

[0071] The dark adaptation bin base main body connection module 17 is located at the bottom of the dark adaptation bin main body shell 11 and the top around the dark adaptation bin base 16, and is used to connect the dark adaptation bin main body shell 11 and the dark adaptation bin base 16 to ensure the light sealing at the connection between the dark adaptation bin main body shell 11 and the dark adaptation bin base.

[0072] Compared with the prior art, the present invention has the following beneficial effects:

[0073] 1. The method provided by the present invention uses fluorescent parameter phenotypic images and hyperspectral phenotypic imaging as data sources to extract phenotypic spatial distribution characteristics, obtains spatial distribution characteristic parameters based on the correlation between phenotypic spatial distribution characteristics and physiological indicators of the leaves to be tested, and inputs the XGBoost ensemble learning model to fuse multi-source phenotypic spatial distribution characteristics, thereby realizing the quantification of strawberry cold injury risk warning and stress detection, and solving the technical problem of insufficient utilization of phenotypic data advantage fusion.

[0074] 2. The present invention realizes a pure phenotypic quantitative early warning of the risk of strawberry low temperature damage by mining the key parameters that are most sensitive to changes in plant photosynthetic physiology with low temperature stress from the spatial distribution characteristics of multi-source phenotypes of the leaves to be tested, thereby solving the technical problem of the lack of a quantitative early warning scheme for plant stress potential based on phenotypic fusion.

[0075] 3. The present invention uses the one-dimensional line domain parallel to the main vein direction of the leaf to be measured, the one-dimensional line domain perpendicular to the main vein direction of the leaf to be measured and the two-dimensional surface domain of the leaf to be measured as key features for risk warning and stress detection, thereby solving the technical problem of the lack of spatial variation feature information in a specific direction in the prior art.

[0076] 4. The present invention constructs a strawberry chilling injury risk warning and stress detection device with phenotype fusion including an imaging acquisition module, an artificial light source, and a dark adaptation chamber based on the principle of lightweight, thereby solving the technical problem of poor portability of the existing phenotypic detection device platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 It is a flow chart of a strawberry chilling injury risk warning and stress detection method based on phenotypic fusion provided by an embodiment of the present invention;

[0078] Figure 2 It is a schematic diagram of the accuracy verification results of a strawberry chilling injury risk warning and stress detection method based on phenotypic fusion provided in an embodiment of the present invention;

[0079] Figure 3 It is a schematic diagram of the three-dimensional structure of a strawberry chilling injury risk warning and stress detection device based on phenotypic fusion provided in an embodiment of the present invention;

[0080] Figure 4It is a cross-sectional view of the internal structure of a strawberry chilling injury risk warning and stress detection device based on phenotype fusion provided by an embodiment of the present invention;

[0081] Figure 5 is a bottom view of the artificial light source provided by an embodiment of the present invention;

[0082] Figure 6 It is a schematic diagram of the three-dimensional structure of the dark adaptation chamber provided by an embodiment of the present invention under the condition that the main shell of the dark adaptation chamber and the base of the dark adaptation chamber are separated;

[0083] Figure 7 is an example diagram of phenotypic distribution characteristics of a one-dimensional line domain parallel to the main vein direction of the leaf to be tested provided by an embodiment of the present invention;

[0084] Figure 8 is an example diagram of phenotypic distribution characteristics of a one-dimensional line domain perpendicular to the main vein direction of the leaf to be tested provided by an embodiment of the present invention;

[0085] Fig. 9 is an example diagram of phenotypic distribution characteristics of a two-dimensional surface area of ​​a leaf to be tested provided by an embodiment of the present invention;

[0086] Figure markings: 1. acquisition and transmission interface; 2. power interface; 3. acquisition, analysis and storage module; 4. power module; 5. expansion bayonet; 6. main housing; 7. imaging acquisition module; 8. imaging lens; 9. artificial light source; 10. collector dark adaptation bin connection module; 11. dark adaptation bin main housing; 12. dark adaptation shade; 13. dark adaptation shade exposed handle; 14. dark adaptation shade moving guide rail; 15. blade holder; 16. dark adaptation bin base; 17. dark adaptation bin base main connection module. DETAILED DESCRIPTION

[0087] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0088] The term "and / or" is only a description of the association relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " generally indicates that the related objects are in an "or" relationship.

[0089] Example 1

[0090] like Figure 1As shown, this embodiment introduces a strawberry chilling injury risk warning and stress detection method based on phenotypic fusion, which is characterized by comprising:

[0091] Step 1: Place the leaves to be tested of the strawberry plant under dark adaptation conditions so that the electrons inside the leaves to be tested are in the ground state.

[0092] Dark adaptation can open the electron gates of the tested leaves of the strawberry plant and keep the internal electrons in the ground state.

[0093] First, the leaves to be tested were kept still for 30 minutes under dark adaptation conditions, and then chlorophyll fluorescence imaging was performed to obtain a fluorescence parameter phenotype image based on the chlorophyll fluorescence imaging measurement.

[0094] Step 2: Extracting phenotypic spatial distribution characteristics based on the fluorescence parameter phenotypic image and the hyperspectral phenotypic image of the leaf to be tested.

[0095] The phenotypic spatial distribution characteristics include a one-dimensional line domain parallel to the main vein direction of the leaf to be measured, a one-dimensional line domain perpendicular to the main vein direction of the leaf to be measured, and a two-dimensional surface domain of the leaf to be measured.

[0096] In this embodiment, a one-dimensional line domain parallel to the main vein direction of the leaf to be measured is defined as: a straight line segment parallel to the main vein of the leaf to be measured, not overlapping with the main vein, passing through the leaf, and having two end points located inside the leaf, such as Figure 7 shown.

[0097] In this embodiment, a one-dimensional line domain perpendicular to the main vein direction of the leaf to be measured is defined as: a straight line segment perpendicular to the main vein of the leaf to be measured, intersecting with the main vein at the widest part of the leaf, passing through the leaf, and having two end points located inside the leaf, such as Figure 8 shown.

[0098] In this embodiment, the two-dimensional surface area of ​​the blade to be tested is defined as: the two-dimensional surface area is located inside the blade and basically covers the entire blade, and is a plane area surrounded by an irregular nearly circular shape, such as Fig. 9 shown.

[0099] Based on the fluorescence parameter phenotype image and the hyperspectral phenotype image of the leaf to be tested, extracting phenotypic spatial distribution characteristics, including:

[0100] Step 2.1: Based on the chlorophyll fluorescence imaging measurement method, the basic fluorescence parameter phenotype image and the chlorophyll fluorescence parameter of the leaf to be measured are collected and calculated to obtain the fluorescence parameter phenotype image of the leaf to be measured;

[0101] Chlorophyll fluorescence imaging mainly measures five parameters: initial fluorescence Fo, maximum fluorescence Fm, initial fluorescence Fo' under light adaptation, maximum fluorescence Fm' under light adaptation and steady-state fluorescence Fs.

[0102] Based on the chlorophyll fluorescence imaging method, the steps for collecting basic fluorescence parameter phenotype images and fluorescence imaging parameters are as follows:

[0103] (1) Open 10 Observe the initial fluorescence Fo image of the leaf to be measured with the measuring light, and record the initial fluorescence Fo parameters;

[0104] (2) Open 6000 Observe the maximum fluorescence Fm image of the leaf to be tested with saturated pulse light and record the maximum fluorescence Fm parameters;

[0105] (3) Open 600 The actinic light makes the leaves to be tested adapt to light;

[0106] (4) Open 6000 again Observe the maximum fluorescence Fm' image of the tested leaf in the light adaptation state with saturated pulse light, and record the maximum fluorescence Fm' parameters in the light adaptation state;

[0107] (5) Repeat steps (1) to (4) according to the set cycle;

[0108] (6) Then open 10 Observe the initial fluorescence Fo' image of the leaves under light adaptation with far-infrared light, and record the initial fluorescence Fo' parameters under light adaptation;

[0109] (7) Finally open 600 The steady-state fluorescence Fs image of the tested leaves was collected using actinic light, and the steady-state fluorescence Fs parameters were recorded.

[0110] Based on the collected basic fluorescence parameter phenotype images and fluorescence imaging parameters, a fluorescence parameter image is obtained by calculation, wherein the fluorescence parameter phenotype image includes a photochemical quenching coefficient image, a non-photochemical quenching coefficient image, an actual fluorescence quantum efficiency image of photosynthetic system PSⅡ and a non-photochemical quenching fluorescence quantum yield image.

[0111] The calculation expression of the photochemical quenching coefficient image is:

[0112] qP=1-(Fs-Fo') / (Fm'-Fo');

[0113] Where qP represents the photochemical quenching coefficient image, and Fs represents the steady-state fluorescence.

[0114] The calculation expression of the non-photochemical quenching coefficient image is:

[0115] NPQ=Fm / Fm'-1;

[0116] Where NPQ represents the non-photochemical quenching coefficient image.

[0117] The calculation expression of the actual fluorescence quantum efficiency image of the photosynthetic system PSⅡ is:

[0118] Y(Ⅱ)=(Fm'-Fs) / Fm';

[0119] Where Y(Ⅱ) represents the actual fluorescence quantum efficiency image of photosynthetic system PSⅡ.

[0120] The calculation expression of the non-photochemical quenching fluorescence quantum yield image is:

[0121] Y(NO)=1 / (NPQ+1+(qP·Fo' / Fs)(Fm / Fo-1));

[0122] Where Y(NO) represents the non-photochemical quenching fluorescence quantum yield image.

[0123] Step 2.2: After the fluorescence parameter image is acquired, based on the spectral index imaging measurement method, the reflectance images of the blue, green and near-infrared bands of the leaves to be tested are collected to construct a hyperspectral phenotype image of the leaves to be tested.

[0124] The hyperspectral phenotype image includes a blue light near-infrared normalized difference hyperspectral phenotype image and a green light near-infrared normalized difference hyperspectral phenotype image.

[0125] Among them, the calculation expression of the blue light near-infrared normalized difference hyperspectral phenotype image is:

[0126] NDVInb=(NIR-B) / (NIR+B);

[0127] Wherein, NDVInb represents the blue-light near-infrared normalized difference hyperspectral phenotypic image, NIR represents the near-infrared band reflectance spectrum phenotypic image, and B represents the blue-light band reflectance spectrum phenotypic image.

[0128] Among them, the calculation expression of the green light near-infrared normalized difference hyperspectral phenotype image is:

[0129] NDVIng=(NIR-G) / (NIR+G);

[0130] Where NDVIng represents the green light near-infrared normalized difference hyperspectral phenotypic image, and G represents the green light band reflectance spectrum phenotypic image.

[0131] Step 2.2: Extract phenotypic spatial distribution characteristics.

[0132] A one-dimensional line domain parallel to the main vein direction of the leaf to be tested is extracted from the photochemical quenching coefficient image qP, the non-photochemical quenching coefficient image NPQ, and the photosynthetic system PSⅡ actual fluorescence quantum efficiency image Y(Ⅱ) of the leaf to be tested respectively.

[0133] A one-dimensional line domain perpendicular to the main vein direction of the leaf to be tested is extracted from the green light near-infrared normalized difference hyperspectral phenotype image NDVIng of the leaf to be tested.

[0134] The two-dimensional surface area of ​​the leaf to be tested is extracted from the non-photochemical quenching fluorescence quantum yield image Y(NO) and the blue light near-infrared normalized difference hyperspectral phenotype image NDVInb of the leaf to be tested respectively.

[0135] Step 3: Obtaining spatial distribution characteristic parameters by calculating the correlation between the leaf pixels of the phenotypic spatial distribution characteristics and the physiological indicators of the leaves to be measured, including:

[0136] Respectively calculating the texture characteristic parameters of the leaf pixels of the fluorescence parameter phenotype image and the hyperspectral phenotype image of the leaf to be tested under the phenotypic spatial distribution characteristics;

[0137] According to the correlation between the texture characteristic parameter and the maximum net photosynthetic rate, leaf relative electrical conductivity and chlorophyll content, the spatial distribution characteristic parameter is obtained.

[0138] Wherein, the texture feature parameters include contrast, angular second-order moment, texture entropy and moment of inertia;

[0139] The calculation expression of the contrast is:

[0140] ;

[0141] Where CON represents the contrast of the phenotypic spatial distribution characteristics, i represents the pixel index of the i-th grayscale value of the phenotypic spatial distribution characteristics, j represents the pixel index of the j-th grayscale value of the phenotypic spatial distribution characteristics, and P(i,j) represents the probability value of the pixel index of the grayscale value (i,j) in the gray level co-occurrence matrix GLCM of the phenotypic spatial distribution characteristics;

[0142] The calculation expression of the angular second-order moment is:

[0143] ;

[0144] In the formula, ASM represents the angular second moment of the phenotypic spatial distribution characteristics;

[0145] The calculation expression of the texture entropy is:

[0146] ;

[0147] Where TENT represents the texture entropy of the phenotypic spatial distribution characteristics, It represents the logarithmic function with base 2. ε has no practical meaning and is a very small constant to ensure that the logarithmic operation is not 0. Its value is 10.-10 ;

[0148] The calculation expression of the moment of inertia is:

[0149] ;

[0150] Where INEM represents the moment of inertia of the spatial distribution characteristics of the phenotype.

[0151] According to the calculation results of the correlation between the texture characteristic parameters and the maximum net photosynthetic rate, leaf relative conductivity and chlorophyll content, it can be known that the texture entropy of the non-photochemical quenching coefficient image in the direction parallel to the main vein of the leaf to be tested and the contrast of the green light near-infrared normalized difference hyperspectral phenotype image in the direction perpendicular to the main vein of the leaf to be tested have the best correlation with the maximum net photosynthetic rate Pmax, the moment of inertia of the non-photochemical quenching fluorescence quantum yield image in the two-dimensional surface domain of the leaf to be tested and the angular second-order moment of the blue light near-infrared normalized difference hyperspectral phenotype image in the two-dimensional surface domain of the leaf to be tested have the best correlation with the leaf relative conductivity REC, and the texture entropy of the photochemical quenching coefficient image in the direction parallel to the main vein of the leaf to be tested and the angular second-order moment of the photosynthetic system PSⅡ actual fluorescence quantum efficiency image in the direction parallel to the main vein of the leaf to be tested have the best correlation with the chlorophyll content Chl.

[0152] Therefore, non-photochemical quenching coefficient image, green light near-infrared normalized difference hyperspectral phenotype image, non-photochemical quenching fluorescence quantum yield image, blue light near-infrared normalized difference hyperspectral phenotype image, photochemical quenching coefficient image and photosynthetic system PSⅡ actual fluorescence quantum efficiency image are selected as the spatial distribution characteristic parameters of the present invention.

[0153] Step 4: According to the spatial distribution characteristic parameters and the XGBoost integrated learning model, phenotypic fusion is performed to predict the photosynthetic physiological potential index and relative negative accumulated temperature of the leaves to be tested.

[0154] The present invention uses the chilling injury risk index CDRI and the photosynthetic physiological potential index PPPI as quantitative evaluation basis for strawberry chilling injury risk warning and stress detection, respectively, and is the synergistic fusion inversion target of spatial distribution characteristic parameters.

[0155] The tree-based ensemble machine learning model is a machine learning algorithm that uses decision trees as basic units and improves prediction performance by integrating multiple decision trees. Compared with the neural network model, the tree-based ensemble machine learning model is more advantageous in this embodiment because the tree-based ensemble machine learning model can not only effectively avoid the overfitting problem, but also has better interpretability.

[0156] Therefore, in order to improve the accuracy of strawberry chilling risk warning and stress detection, this implementation preferably uses the integrated learning model XGBoost as a feature fusion. In the XGBoost integrated learning model, the spatial distribution feature parameters are used as input factors. Based on the method provided by the present invention, this implementation verifies the accuracy of strawberry chilling risk warning and stress detection. The verification results are as follows: Figure 2 As shown, Figure 2 (a) is the verification result of the chilling damage risk index CDRI, Figure 2 (b) is the verification result of photosynthetic physiological potential index PPPI.

[0157] It is worth noting that the validation sample is independent of the training sample and the test sample. In this embodiment, the ratio between the validation sample, the training sample and the test sample is set to 1:2:5.

[0158] Step 5: Calculate the chilling injury risk index result of the leaf to be tested according to the photosynthetic physiological potential index and the relative negative accumulated temperature.

[0159] Maximum net photosynthetic rate The chilling damage risk index CDRI is calculated by introducing the relative negative accumulated temperature RNAT on the basis of the photosynthetic physiological potential index PPPI.

[0160] Among them, the calculation expression of photosynthetic physiological potential index PPPI is:

[0161] ;

[0162] ;

[0163] ;

[0164] ;

[0165] In the formula, Represents the first component of the photosynthetic physiological potential index PPPI, Represents the second component of the photosynthetic physiological potential index PPPI, Represents the third component of the photosynthetic physiological potential index PPPI, represents the maximum net photosynthetic rate, REC represents the relative electrical conductivity of leaves, Indicates the chlorophyll content.

[0166] Among them, the calculation expression of relative negative accumulated temperature RNAT is:

[0167] ;

[0168] In the formula, n represents the total number of accumulated hours. Indicates the temperature under the suitable environment CK, taking 20℃, Indicates the temperature in the low temperature environment LT at time t.

[0169] Among them, the calculation expression of the chilling damage risk index CDRI is:

[0170] ;

[0171] Step 6: Determine the strawberry low temperature damage warning and stress level.

[0172] The method for determining strawberry low temperature chilling injury warning and stress degree according to the chilling injury risk index and the photosynthetic physiological potential index comprises:

[0173] When strawberry low temperature chilling damage has not occurred, the strawberry low temperature chilling damage warning is determined according to the chilling damage risk index result, including:

[0174] When the chilling injury risk index result is ≥3, it is determined that there is no chilling injury to strawberries;

[0175] When 2≤chilling injury risk index result<3, strawberry low temperature chilling injury blue warning is determined;

[0176] When 1≤chilling injury risk index result<2, a yellow warning of strawberry low temperature chilling injury is determined;

[0177] When 0≤chilling injury risk index result<1, strawberry low temperature chilling injury red warning is determined;

[0178] When the chilling injury risk index result is less than 0, it is determined that strawberry chilling injury has occurred;

[0179] When strawberry low temperature chilling injury has occurred, the degree of chilling injury of strawberry plants is determined according to the photosynthetic physiological potential index, including:

[0180] When the photosynthetic physiological potential index is ≥3.5, it is determined that the strawberry plants have no chilling damage;

[0181] When the photosynthetic physiological potential index is 2.8≤<3.5, the strawberry plants are judged to be slightly damaged by chilling injury;

[0182] When 1.5≤photosynthetic physiological potential index<2.8, the strawberry plants were judged to be moderately damaged by chilling injury;

[0183] When the photosynthetic physiological potential index is ≤1.5, the strawberry plants are judged to have suffered severe chilling damage.

[0184] Example 2

[0185] like Figure 3 , Figure 4 As shown, based on the same inventive concept as Example 1, the present invention introduces a strawberry cold injury risk warning and stress device based on phenotypic fusion, characterized in that it includes a device body, a dark adaptation chamber detachably connected to the device body, and a collector dark adaptation chamber connection module, which is used to implement the method described in Example 1.

[0186] The device body includes: an acquisition and transmission interface 1, a power interface 2, an acquisition and analysis storage module 3, a power module 4, an expansion bayonet 5, a main body shell 6, an imaging acquisition module 7, an imaging lens 8, and an artificial light source 9.

[0187] The acquisition and transmission interface 1 is connected to the acquisition and analysis storage module 3 and is used to transmit the acquired data of the fluorescence parameter phenotype image and the hyperspectral phenotype image of the leaf to be tested to the USB communication interface of the external device.

[0188] The power interface 2 is a plug-in interface connected to the power module 4 and is used to transmit power.

[0189] In this embodiment, the acquisition and transmission interface 1 and the power interface 2 are both located at the top of the device body, the acquisition and transmission interface 1 is a single USB 2.0 port, and the power interface 2 is a two-phase power adapter.

[0190] The acquisition, analysis and storage module 3 is connected to the acquisition and transmission interface 1 at the top and to the imaging acquisition module 7 and the artificial light source 9 at the bottom, respectively, and is used to control the program for acquiring fluorescence parameter phenotype images and hyperspectral phenotype images, and temporarily store the data of the acquired fluorescence parameter phenotype images and hyperspectral phenotype images of the leaves to be tested.

[0191] In this embodiment, the acquisition and analysis storage module 3 receives the acquisition signal input by the acquisition and transmission interface 1, is responsible for regulating the coordinated work of the imaging acquisition module 7 and the artificial light source 9, acquires fluorescence imaging images and hyperspectral phenotypic images under different light sources and different filters, extracts the leaf pixels of the phenotypic spatial distribution characteristics according to the preset standard placement position of the leaves to be tested, and automatically calculates the spatial distribution characteristic parameters, and further inverts the photosynthetic physiological potential index PPPI and the relative negative accumulated temperature RNAT through the pre-trained XGBoost integrated learning model, and predicts the cold damage risk index CDRI, and stores the cold damage risk index CDRI result together with the imaging data of the fluorescence parameter phenotypic image and the hyperspectral phenotypic image, the physiological potential index PPPI, the relative negative accumulated temperature RNAT, the spatial distribution characteristic parameters and other intermediate data into the device cache, and outputs through the acquisition and transmission interface 1.

[0192] The power module 4 is connected to the power interface 2 at the top and to the acquisition, analysis and storage module 3, the imaging acquisition module 7 and the artificial light source 9 at the bottom, for providing power.

[0193] In this embodiment, the power supply module 4 is responsible for modulating the current and voltage, maintaining the operation of the acquisition and analysis storage module 3, the imaging acquisition module 7, and the artificial light source 9, and is equipped with a small battery pack to complete the current observation under unexpected breakpoint conditions and break again after the storage is completed.

[0194] Among them, expansion bayonet sockets 5 are distributed on the four sides of the upper part of the main shell of the device for connecting external auxiliary equipment.

[0195] In this embodiment, the expansion bayonet 5 is a “concave” shaped metal component, and the auxiliary device can be pushed in and fixed from the left or right side of the expansion bayonet 5 .

[0196] The main body shell 6 is connected to the dark adaptation chamber at the bottom, and is used to accommodate and protect the device body, specifically: the bottom is connected to the dark adaptation chamber main body shell 11, and is used to accommodate and protect the acquisition analysis storage module 3, the power module 4, the imaging acquisition module 7, the imaging lens 8, and the artificial light source 9;

[0197] In this embodiment, the main shell 6 is made of polycarbonate, the inner side is a diffuse reflection and light absorption material, and the outer side is a frosted material.

[0198] The imaging acquisition module 7 includes a multispectral CMOS image sensor and a filter wheel, which is connected to the acquisition analysis storage module 3 at the top and to the imaging lens 8 at the bottom, and is also connected to the power module 4 through a power line; it is used to acquire fluorescence parameter phenotype images and hyperspectral phenotype images.

[0199] In this embodiment, the imaging acquisition module 7 uses a 16-megapixel CMOS image sensor with a resolution of 4000×4000 and an effective size of 20 microns for each pixel. The spectral coverage of the CMOS image sensor is wide, from 400 nanometers to 800 nanometers, and is suitable for a variety of spectral imaging needs. In addition, this embodiment is also equipped with an 8-bit filter wheel at the connection between the imaging acquisition module 7 and the imaging lens 8, which includes multiple bands such as fluorescence, blue light, green light, near infrared and far infrared, so that the imaging acquisition module can flexibly select the required spectral band for imaging.

[0200] The imaging lens 8 is connected to the imaging acquisition module 7 at the top and is used to adjust the focal length and imaging magnification.

[0201] In this embodiment, the aperture of the imaging lens 8 is F2.2 and the focal length is 5 mm. It is installed below the imaging acquisition module 7 and in the hollow area of ​​the artificial light source 9 and can be removed and replaced as needed.

[0202] The artificial light source 9 includes a red light unit, a blue light unit, a white light unit, and a far infrared unit, wherein the red light unit is used as a measuring light source; the blue light unit is used as a saturated pulse light source; the white light unit is used as an actinic light source; and the far infrared unit is used as a measuring light source for chlorophyll fluorescence parameters. The artificial light source 9 is respectively connected to the collection and analysis storage module 3 and the power supply module 4 on the top, and is in a concentric circular structure with a hollow interior, such as Figure 5 As shown; the imaging lens 8 is installed in the hollow of the concentric circles to ensure that the light can be evenly irradiated on the blade to be measured, and the imaging lens 8 is used to collect images.

[0203] Among them, in this embodiment, the red light unit adopts a red light source of a 650nm LED lamp bead array; the blue light unit adopts a blue light source of a 450nm LED lamp bead array; the white light unit adopts a white light source of a 400~700nm LED lamp bead array; the far-infrared unit adopts a far-infrared light source of a 760nm LED lamp bead array.

[0204] Among them, the collector dark adaptation chamber connection module 10 is located at the bottom of the device main body shell and the top of the upper opening of the dark adaptation chamber, and is raised at part of the edge of the dark adaptation chamber, and is used to be magnetically connected to the device body when the device body is directly inserted.

[0205] In this embodiment, the portion of the device body in the collector dark adaptation chamber connection module 10 that contacts and docks with the dark adaptation chamber is light-absorbing cotton covered with a magnet.

[0206] The dark adaptation chamber is used to place the leaf to be tested completely or partially under dark adaptation conditions so that the electrons inside the leaf to be tested are in the ground state.

[0207] The dark adaptation warehouse comprises: a dark adaptation warehouse main body shell 11, a dark adaptation blackout curtain 12, an exposed handle of the dark adaptation blackout curtain 13, a dark adaptation blackout curtain moving guide rail 14, a blade holder 15, a dark adaptation warehouse base 16, and a dark adaptation warehouse base main body connection module 17.

[0208] The main body shell 11 of the dark adaptation chamber is used to accommodate and protect the internal space to form a dark adaptation environment.

[0209] In this embodiment, the dark adaptation warehouse main body shell 11 is made of polycarbonate, the inner side is diffuse reflection and light absorption material, and the outer side is smooth and highly reflective material. The dark adaptation warehouse base 16 and the dark adaptation warehouse main body shell 11 are split and connected by magnetic attraction.

[0210] The dark-adaptation blackout curtains 12 are located on both sides of the top of the dark-adaptation chamber and are hard multi-piece translational structures for blocking external light.

[0211] In this embodiment, the dark-adaptive blackout curtain 12 is made of polycarbonate, and the surface is covered with a thin layer of light-absorbing cotton.

[0212] The exposed handle 13 of the dark-adapting blackout curtain is connected to the lowest blackout curtain in the dark-adapting blackout curtain 12, exposed outside the dark-adapting chamber, and is used to operate the opening and closing of the dark-adapting blackout curtain.

[0213] In this embodiment, the exposed handle 13 of the dark-adaptive shade is made of polycarbonate.

[0214] The dark-adaptive shade curtain moving guide rail 14 is connected to the dark-adaptive shade curtain 12 and the exposed handle 13 of the dark-adaptive shade curtain, and is used to fix the moving track of the exposed handle 13 of the dark-adaptive shade curtain so that the dark-adaptive shade curtain is evenly unfolded.

[0215] In this embodiment, the dark-adaptive shade curtain moving guide rail 14 and its inner side are covered with a thin layer of light-absorbing cotton to prevent external light from entering.

[0216] Wherein, the dark adaptation chamber further comprises:

[0217] The blade holder 15 is an axis-rotating transparent plastic pressing sheet, located on both side edges of the upper part of the dark adaptation chamber base 16, and is used to lay and fix the blade to be tested.

[0218] In this embodiment, the blade holder 15 is composed of an end point shaft and a transparent blade strip. The end point shaft is fixed at a corner point of the dark adaptation chamber base 16 . The fixing range of the two blade holders 15 distributed at the diagonal corner points covers the entire dark adaptation chamber base 16 .

[0219] The dark adaptation chamber base 16 is detachably connected to the bottom of the dark adaptation chamber body 11 by magnetic attraction. Figure 6 As shown; used to support and fix the dark adaptation warehouse main body shell 11.

[0220] In this embodiment, the surface of the dark adaptation chamber base 16 is made of diffuse reflection and light absorption material.

[0221] The dark adaptation warehouse base main body connection module 17 is located at the bottom of the dark adaptation warehouse main body shell 11 and the top around the dark adaptation warehouse base 16. Both sides are made of magnet-covered light-absorbing cotton material. It is used to connect the dark adaptation warehouse main body shell 11 and the dark adaptation warehouse base 16 to ensure the light sealing at the connection between the dark adaptation warehouse main body shell 11 and the dark adaptation warehouse base 16.

[0222] The specific functions of the above components are implemented by referring to the relevant contents of the method in Example 1 and will not be elaborated here.

[0223] Example 3

[0224] In combination with Example 1 and Example 2, this example introduces the experimental steps of a strawberry chilling injury risk warning and stress detection method based on phenotypic fusion, including:

[0225] (1) Dark adaptation steps:

[0226] The dark adaptation chamber base 16 is attached to the back of the blade to be tested, and the blade holder 15 is used to firmly press the edge of the blade to be tested to ensure that the blade to be tested is tightly attached to the dark adaptation chamber base 16 .

[0227] The dark adaptation chamber base 16 on which the blade to be tested is fixed is connected to the dark adaptation chamber main body shell 11 by magnetic attraction.

[0228] The exposed handles 13 of the dark adaptation shade curtain are moved from both sides to the middle to completely close the dark adaptation shade curtain 12, thereby providing a dark adaptation environment for the leaves to be tested for 30 minutes.

[0229] (2) Preparation for imaging acquisition:

[0230] After the dark adaptation is completed, the imaging acquisition module 7 is aligned and inserted into the collector dark adaptation chamber connection module 10, and is automatically adsorbed and fixed by magnetic attraction.

[0231] The exposed handle 13 of the dark-adaptive shading curtain is moved from the middle to both sides to retract the dark-adaptive shading curtain 12 to expose the blade to be tested.

[0232] Turn on the power of the entire strawberry chilling risk warning and stress detection system, and connect the external device through the acquisition transmission interface 1 to prepare for fluorescence imaging acquisition.

[0233] (3) Fluorescence imaging acquisition:

[0234] First, the measuring light on the main housing 11 of the dark adaptation chamber is turned on, and the initial fluorescence Fo image is synchronously acquired using the imaging acquisition module 7 .

[0235] Then, the saturation pulse light is turned on, and the imaging acquisition module 7 is used to synchronously acquire the maximum fluorescence Fm image.

[0236] The actinic light is turned on to make the leaves to be tested adapt to light, and then the saturation pulse light is turned on again, and the maximum fluorescence Fm' image in the light adaptation state is synchronously collected by the imaging acquisition module 7.

[0237] After repeating the above steps according to a preset cycle, the far-infrared light is turned on, and the imaging acquisition module 7 is used to synchronously acquire the initial fluorescence Fo' image in the light adaptation state.

[0238] Finally, the actinic light is turned on again, and the imaging acquisition module 7 is used to synchronously acquire the steady-state fluorescence Fs image.

[0239] Based on the collected fluorescence parameter phenotype images, the photochemical quenching coefficient qP image, the non-photochemical quenching coefficient NPQ image, the actual fluorescence quantum efficiency Y(Ⅱ) image of the photosynthetic system PSⅡ and the non-photochemical quenching fluorescence quantum yield Y(NO) image were calculated.

[0240] (4) Reflectivity image acquisition:

[0241] The white light LED array in the artificial light source 9 is turned on to collect reflectance images of the leaves to be tested in the three wavebands of blue, green and near infrared.

[0242] After analysis by the acquisition, analysis and storage module 3, a blue light near-infrared normalized difference hyperspectral phenotype image NDVInb and a green light near-infrared normalized difference hyperspectral phenotype image NDVIng are obtained.

[0243] (5) Image and data processing:

[0244] The acquisition, analysis and storage module 3 automatically generates pixel positions in a one-dimensional line domain and a two-dimensional surface domain parallel to and perpendicular to the main veins according to the direction of the main veins of the leaf to be measured.

[0245] Automatically discard redundant parts of the collected data, retain specific images such as photochemical quenching coefficient image, non-photochemical quenching coefficient image, PSⅡ actual fluorescence quantum efficiency image parallel to the main vein direction in the one-dimensional line domain data, retain the green light near-infrared normalized difference spectral index image perpendicular to the main vein direction of the one-dimensional line domain data, and retain the non-photochemical quenching fluorescence quantum yield image and blue light near-infrared normalized difference spectral index image Two-dimensional surface image data.

[0246] The texture characteristic parameters of the fluorescence parameter phenotype image and the spectral index image under the phenotype spatial distribution characteristics are calculated, including texture entropy, angular second-order moment, moment of inertia and contrast.

[0247] According to the correlation between texture characteristic parameters and maximum net photosynthetic rate, leaf relative conductivity and chlorophyll content, the spatial distribution characteristic parameters were obtained.

[0248] (6) Model prediction:

[0249] The acquisition, analysis and storage module 3 automatically substitutes the spatial distribution characteristic parameters obtained by analysis into the pre-deployed XGBoost integrated learning model to predict the photosynthetic physiological potential index PPPI and the relative negative accumulated temperature RNAT.

[0250] Based on the photosynthetic physiological potential index PPPI and the relative negative accumulated temperature RNAT, the chilling damage risk index CDRI was calculated.

[0251] (7) Data storage and output:

[0252] The acquisition, analysis and storage module 3 stores the final cold damage risk index CDRI result, as well as the imaging data of the fluorescence parameter phenotype image, the hyperspectral phenotype image, the photosynthetic physiological potential index PPPI, the relative negative accumulated temperature RNAT, the spatial distribution characteristic parameters and other intermediate data into the device cache, and outputs them to the external device via the acquisition and transmission interface 1 for further analysis or recording.

[0253] (8) Cold damage warning and stress level determination:

[0254] Based on the results of the chilling injury risk index CDRI and the photosynthetic physiological potential index PPPI, the low temperature chilling injury warning and stress level of strawberry are comprehensively determined.

[0255] In summary, the strawberry chilling risk warning and stress detection method based on phenotypic fusion proposed by the present invention realizes the comprehensive extraction of the phenotypic spatial distribution characteristics of the leaves to be tested by utilizing the fluorescence parameter phenotypic image and the hyperspectral phenotypic image of the leaves to be tested under dark adaptation conditions. The phenotypic spatial distribution characteristics cover the physiological variation information of the leaves in the direction parallel to the main veins, perpendicular to the main veins, and in the two-dimensional domain. In addition, the present invention further obtains the spatial distribution characteristic parameters with physiological significance by calculating the correlation between the phenotypic spatial distribution characteristics and the three key plant photosynthetic physiological potential indicators of the maximum net photosynthetic rate, the relative conductivity of the leaves, and the chlorophyll content. It not only newly explores the variation distribution characteristics of the physiology of the leaves to be tested in different specific directions, but also makes the utilization of the leaf scale phenotypic information more comprehensive than the prior art. Therefore, the method provided by the present invention realizes the full fusion of the spatial variation characteristic information in a specific direction with the phenotypic data, so as to quickly, non-destructively and quantitatively determine the chilling risk warning and stress degree of strawberry plants from the leaf scale.

[0256] The present invention uses spatial distribution characteristic parameters as training data, and combined with the advanced XGBoost ensemble learning model, can achieve accurate prediction of leaf photosynthetic physiological potential index and relative negative accumulated temperature. This innovative method not only provides a scientific basis for the assessment of strawberry low temperature chilling risk, but also successfully achieves a purely phenotypic quantitative forecast of crop low temperature disaster risk by introducing relative negative accumulated temperature, further improving the accuracy and practicality of the prediction.

[0257] The present invention further deduces the chilling injury risk index result of the leaves through the calculated photosynthetic physiological potential index and relative negative accumulated temperature of the leaves to be tested, which not only directly reflects the degree of damage of the leaves under low temperature stress, but also can comprehensively determine the chilling injury warning and stress degree of strawberry plants in combination with the photosynthetic physiological potential index. Compared with the traditional method, the present invention not only improves the accuracy and reliability of chilling injury risk assessment, but also its portability, high efficiency and real-time performance also provide strong support for the sustainable development of the strawberry planting industry. At the same time, the method has a wide range of applications, not only applicable to strawberries, but also can be extended to other crops, providing a new technical approach for the prevention and control of low temperature disasters in agricultural production.

[0258] At the same time, the present invention also provides a strawberry cold injury risk warning and stress detection device based on phenotypic fusion. The device has a compact structure, strong mobility, and is easy to carry. It is equipped with a detachable dark adaptation chamber, which allows simultaneous dark adaptation during multi-target observation. Compared with other devices, the dark adaptation efficiency before observation is higher.

[0259] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0260] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0261] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0262] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0263] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A strawberry chilling injury risk warning and stress detection method based on phenotypic fusion, characterized in that: include: The leaf to be tested is placed under dark adaptation conditions so that the internal electrons of the leaf to be tested are in the ground state; Based on the fluorescence parameter phenotype image and the hyperspectral phenotype image of the leaf to be tested, extracting phenotypic spatial distribution characteristics, including: Based on the chlorophyll fluorescence imaging measurement method, the basic fluorescence parameter phenotype image and the chlorophyll fluorescence parameter image of the leaf to be measured are collected for calculation to obtain the fluorescence parameter phenotype image of the leaf to be measured; the fluorescence parameter phenotype image includes a photochemical quenching coefficient image, a non-photochemical quenching coefficient image, a photosynthetic system PSⅡ actual fluorescence quantum efficiency image and a non-photochemical quenching fluorescence quantum yield image; Based on the spectral index imaging measurement method, the reflectance images of the three bands of blue, green and near-infrared of the leaves to be measured are collected to construct a hyperspectral phenotypic image of the leaves to be measured; the hyperspectral phenotypic image includes a blue light near-infrared normalized difference hyperspectral phenotypic image and a green light near-infrared normalized difference hyperspectral phenotypic image; Extracting a one-dimensional line domain parallel to the main vein direction of the leaf to be tested from the photochemical quenching coefficient image, the non-photochemical quenching coefficient image, and the photosynthetic system PSⅡ actual fluorescence quantum efficiency image of the leaf to be tested respectively; Extracting a one-dimensional line domain perpendicular to the main vein direction of the leaf to be tested from the green light near-infrared normalized difference hyperspectral phenotype image of the leaf to be tested; Extracting the two-dimensional surface area of ​​the leaf to be tested from the non-photochemical quenching fluorescence quantum yield image and the blue light near-infrared normalized difference hyperspectral phenotype image of the leaf to be tested respectively; The spatial distribution characteristic parameters are obtained by calculating the correlation between the leaf pixels of the phenotypic spatial distribution characteristics and the physiological indexes of the leaves to be measured; According to the spatial distribution characteristic parameters and the XGBoost integrated learning model, phenotypic fusion is performed to predict the photosynthetic physiological potential index and relative negative accumulated temperature of the leaves to be tested; Calculating the chilling injury risk index result of the leaf to be tested according to the photosynthetic physiological potential index and the relative negative accumulated temperature; Determine the strawberry low temperature chilling injury warning and stress degree according to the chilling injury risk index and the photosynthetic physiological potential index; The phenotypic spatial distribution characteristics include a one-dimensional line domain parallel to the main vein direction of the leaf to be measured, a one-dimensional line domain perpendicular to the main vein direction of the leaf to be measured, and a two-dimensional surface domain of the leaf to be measured.

2. The strawberry chilling injury risk early warning and stress detection method based on phenotypic fusion according to claim 1, characterized in that: The method of obtaining the spatial distribution characteristic parameters by calculating the correlation between the leaf pixels of the phenotypic spatial distribution characteristics and the physiological indicators of the leaves to be measured includes: Respectively calculating the texture characteristic parameters of the leaf pixels of the fluorescence parameter phenotype image and the hyperspectral phenotype image of the leaf to be tested under the phenotypic spatial distribution characteristics; According to the correlation between the texture characteristic parameters and the maximum net photosynthetic rate, leaf relative conductivity and chlorophyll content, the spatial distribution characteristic parameters are obtained; wherein the maximum net photosynthetic rate, leaf relative conductivity and chlorophyll content are physiological indicators of the leaves to be tested.

3. The strawberry chilling injury risk early warning and stress detection method based on phenotypic fusion according to claim 2, characterized in that: The texture feature parameters include contrast, angular second-order moment, texture entropy and moment of inertia; The calculation expression of the contrast is: ； Where CON represents the contrast of the phenotypic spatial distribution characteristics, i represents the pixel index of the i-th grayscale value of the phenotypic spatial distribution characteristics, j represents the pixel index of the j-th grayscale value of the phenotypic spatial distribution characteristics, and P(i,j) represents the probability value of the pixel index of the grayscale value (i,j) in the gray level co-occurrence matrix GLCM of the phenotypic spatial distribution characteristics; The calculation expression of the angular second-order moment is: ； In the formula, ASM represents the angular second moment of the phenotypic spatial distribution characteristics; The calculation expression of the texture entropy is: ； Where TENT represents the texture entropy of the phenotypic spatial distribution characteristics, It represents the logarithmic function with base 2. ε has no practical meaning and is a very small constant to ensure that the logarithmic operation is not 0. Its value is 10. -10 ; The calculation expression of the moment of inertia is: ； Where INEM represents the moment of inertia of the spatial distribution characteristics of the phenotype.

4. The strawberry chilling injury risk early warning and stress detection method based on phenotypic fusion according to claim 1, characterized in that: The method for determining strawberry low temperature chilling injury warning and stress degree according to the chilling injury risk index and the photosynthetic physiological potential index comprises: When strawberry low temperature chilling damage has not occurred, the strawberry low temperature chilling damage warning is determined according to the chilling damage risk index result, including: When the chilling injury risk index result is ≥3, it is determined that there is no chilling injury to strawberries; When 2≤chilling injury risk index result<3, strawberry low temperature chilling injury blue warning is determined; When 1≤chilling injury risk index result<2, a yellow warning of strawberry low temperature chilling injury is determined; When 0≤chilling injury risk index result<1, strawberry low temperature chilling injury red warning is determined; When the chilling injury risk index result is less than 0, it is determined that strawberry chilling injury has occurred; When strawberry low temperature chilling injury has occurred, the degree of chilling injury of strawberry plants is determined according to the photosynthetic physiological potential index, including: When the photosynthetic physiological potential index is ≥3.5, it is determined that the strawberry plants have no chilling damage; When the photosynthetic physiological potential index is 2.8≤<3.5, the strawberry plants are judged to be slightly damaged by chilling injury; When 1.5≤photosynthetic physiological potential index<2.8, the strawberry plants were judged to be moderately damaged by chilling injury; When the photosynthetic physiological potential index is ≤1.5, the strawberry plants are judged to have suffered severe chilling damage.

5. A strawberry chilling injury risk warning and stress device based on phenotypic fusion, characterized in that: It comprises a device body, a dark adaptation chamber detachably connected to the device body, and a collector dark adaptation chamber connection module, and is used to implement the method described in any one of claims 1 to 4; Wherein, the device body comprises: A collection and transmission interface (1) connected to the collection and analysis storage module (3) and used to transmit the collected data of the fluorescence parameter phenotype image and the hyperspectral phenotype image of the leaf to be tested to a USB communication interface of an external device; The power interface (2) is a plug-in interface connected to the power module (4) and is used to transmit power; An acquisition analysis storage module (3) connected to the acquisition transmission interface (1) at the top and respectively connected to the imaging acquisition module (7) and the artificial light source (9) at the bottom, for controlling the program for acquiring the fluorescence parameter phenotype image and the hyperspectral phenotype image, and temporarily storing the acquired data of the fluorescence parameter phenotype image and the hyperspectral phenotype image of the leaf to be tested; A power module (4) connected to the power interface (2) at the top and respectively connected to the acquisition and analysis storage module (3), the imaging acquisition module (7) and the artificial light source (9) at the bottom, for providing power; A main body shell (6), connected to the dark adaptation chamber at the bottom, and used to accommodate and protect the device body; An imaging acquisition module (7), comprising a multispectral CMOS image sensor and a filter wheel, connected to the acquisition analysis storage module (3) at the top, connected to the imaging lens (8) at the bottom, and connected to the power module (4) via a power line, for acquiring fluorescence parameter phenotype images and hyperspectral phenotype images; An imaging lens (8) connected to the imaging acquisition module (7) at the top and used to adjust the focal length and imaging magnification; The artificial light source (9) comprises a red light unit, a blue light unit, a white light unit, and a far infrared unit, wherein the red light unit is used as a measuring light source; the blue light unit is used as a saturated pulse light source; the white light unit is used as an actinic light source; and the far infrared unit is used as a measuring light source for initial fluorescence parameters. The artificial light source (9) is in the form of a concentric circular structure with a hollow interior. The imaging lens (8) is installed in the hollow part of the concentric circular structure to ensure that light can be evenly irradiated onto the blade to be measured, and imaging is collected through the imaging lens (8); The collector dark adaptation chamber connection module (10) is located between the bottom of the main housing (6) and the top of the upper opening of the dark adaptation chamber, and is raised at a portion of the edge of the dark adaptation chamber, and is used for magnetically connecting with the device body when the device body is directly inserted into the dark adaptation chamber; The dark adaptation chamber is used to place the leaf to be tested completely or partially under dark adaptation conditions so that the electrons inside the leaf to be tested are in the ground state.

6. The strawberry chilling injury risk warning and stress device based on phenotypic fusion according to claim 5, characterized in that: The four sides of the upper portion of the main housing (6) are also provided with expansion bayonet ports (5) for connecting external auxiliary equipment.

7. The strawberry chilling injury risk warning and coercion device based on phenotypic fusion according to claim 5, characterized in that: The red light unit adopts a red light source of a LED lamp bead array with a wavelength of 650nm; the blue light unit adopts a blue light source of a LED lamp bead array with a wavelength of 450nm; the white light unit adopts a white light source of a LED lamp bead array with a wavelength of 400~700nm; the far-infrared unit adopts a far-infrared light source of a LED lamp bead array with a wavelength of 760nm.

8. The strawberry chilling injury risk warning and coercion device based on phenotypic fusion according to claim 5, characterized in that: The dark adaptation chamber comprises: The main body shell (11) of the dark adaptation chamber is used to accommodate and protect the internal space to form a dark adaptation environment; Dark adaptation blackout curtains (12), located on both sides of the top of the dark adaptation chamber, are hard multi-piece translational structures and are used to block external light; An exposed handle (13) of the dark adaptation blackout curtain, connected to the lowest blackout curtain of the dark adaptation blackout curtain (12), exposed outside the dark adaptation chamber, and used for operating the opening and closing of the dark adaptation blackout curtain; The dark-adaptive shade curtain moving guide rail (14) is connected to the dark-adaptive shade curtain (12) and the exposed handle (13) of the dark-adaptive shade curtain, and is used to fix the moving track of the exposed handle (13) of the dark-adaptive shade curtain, so that the dark-adaptive shade curtain is evenly unfolded.

9. The strawberry chilling injury risk warning and coercion device based on phenotypic fusion according to claim 8, characterized in that: The dark adaptation chamber also includes: The blade holder (15) is an axis-rotating transparent plastic pressing sheet, located on both sides of the upper edge of the dark adaptation chamber base (16), and is used to lay and fix the blade to be tested; A dark adaptation chamber base (16) is detachably connected to the bottom of the dark adaptation chamber main body shell (11) by magnetic attraction, and is used to support and fix the dark adaptation chamber main body shell (11); The dark adaptation chamber base main body connection module (17) is located at the bottom of the dark adaptation chamber main body shell (11) and the top of the dark adaptation chamber base (16) and is used to connect the dark adaptation chamber main body shell (11) and the dark adaptation chamber base (16) to ensure the light tightness of the connection between the dark adaptation chamber main body shell (11) and the dark adaptation chamber base.

Images