Plant living body flavonol imaging detection method based on multi-excitation chlorophyll fluorescence imaging technology

Abstract

The invention relates to the technical field of plant living body nondestructive testing, and discloses a plant living body flavonol imaging detection method based on a multi-excitation light chlorophyll fluorescence imaging technology, which comprises the following steps: on the basis of the multi-excitation light chlorophyll fluorescence imaging technology, using a multi-excitation light chlorophyll fluorescence hyperspectral imaging function of a FluorTron multifunctional hyperspectral imaging system to carry out multi-excitation light chlorophyll fluorescence hyperspectral imaging on the basis of the multi-excitation light chlorophyll fluorescence imaging technology; collecting chlorophyll fluorescence spectrum images of the ginkgo leaves in a non-invasive and high-throughput manner; then, carrying out de-noising processing and fluorescence spectrum analysis on the spectral image by using FluorVision hyperspectral fluorescence imaging analysis software; and finally, analyzing Flav indexes of the flavonol model by using FluorVision hyperspectral fluorescence imaging analysis software, and making a graph. Compared with a traditional flavonol detection method, the method has the advantages of high throughput, non-contact, non-damage, digitization, visualization, in-vivo imaging detection and the like, in addition, a multi-excitation chlorophyll fluorescence spectrum imaging technology is adopted, the chlorophyll content, the anthocyanin content, the plant N element state and the like can be detected and analyzed at the same time, and the method is suitable for large-scale popularization and application. Therefore, the method has important application value and prospect in the fields of plant flavonol and anthocyanin high-throughput in-vivo detection, medicinal plant phenotype analysis, secondary metabolite imaging analysis, non-destructive rapid detection of traditional Chinese medicinal materials, in-vivo detection of functional components of fruits and vegetables, postharvest biology, remote sensing and the like.

Description

Technical Field

[0001] The invention relates to the technical field of non-destructive detection of living plants, and in particular to a method for imaging and detecting flavonols in living plants based on multi-excitation light chlorophyll fluorescence imaging technology. Background Art

[0002] Flavonoids such as flavonols are important components of plant secondary metabolites. They play an important role in plant growth and development and in protecting against external damage. They have antioxidant effects and absorb ultraviolet rays to protect chlorophyll in mesophyll cells. Therefore, they are called "sunscreen" of plants. At the same time, flavonols not only play a role in protecting plants from external environmental stress in plants, but also have a variety of physiologically active functions and health benefits for the human body. They are one of the hot topics in current medical and food science research. Traditional flavonoid detection methods require extraction, separation and content determination through a variety of physical and chemical methods. For example, high performance liquid chromatography HPLC (Dong et al., 2005) uses the different distribution coefficients of different substances in the stationary phase and the mobile phase to separate them through chromatography. Flavonol compounds are separated in the chromatographic column according to their chemical properties, and then quantified by ultraviolet detector-high performance liquid chromatography UV-Vis (Solovchenko et al., 2001). The specific method is as follows: 1) Select an appropriate retention column and mobile phase, set the composition and flow rate of the mobile phase; 2) Take an appropriate amount of ginkgo leaf sample, crush and extract it to obtain a mixture of flavonoids; 3) Add the above mixture to the HPLC equipment and enter the chromatographic column through the sample injection port; 4) In the chromatographic column, flavonoids will gradually separate under the action of the mobile phase according to their polarity and mutual force; 5) Use a detector to detect the peak shape, peak height, peak area and other parameters of different components in the sample, so as to quantitatively determine the total flavonoid content in the ginkgo leaf. This type of method requires pretreatment, crushing, extraction, separation and concentration of samples such as leaves, which is complicated, time-consuming, laborious, and requires sample destruction, and cannot meet the needs of rapid, high-throughput, non-destructive, and in vivo detection. How to quickly and non-destructively detect the flavonol content of plants such as Chinese medicinal materials, fruits and vegetables by high-throughput in vivo imaging is of great significance for studying the temporal and spatial distribution of flavonol content (such as different seasons, different leaves or different parts of leaves), when to harvest the highest content, under what conditions the highest content is (such as different lighting conditions - photobiological research), in vivo quality detection and identification, high-throughput detection of functional components from different sources and origins, etc. However, there are currently no mature instruments and equipment on the market for rapid, non-destructive, non-contact, high-throughput in vivo imaging measurements of flavonols.

[0003] Based on the phenomenon that chlorophyll absorbs red light, flavonols absorb ultraviolet light, and anthocyanins absorb green light, the French company Force-a developed and produced the instrument Dualex (Goulas et al., 2004; Muller et al., 2013) for measuring the flavonol content index Flav using an algorithm model (3). This instrument achieves the purpose of non-destructive in vivo measurement of flavonols, but because it does not have spectral imaging capabilities, it cannot achieve the goals of non-contact, high-throughput, and visualization. Summary of the invention

[0004] 1. Technical issues to be resolved In view of the shortcomings of the prior art, the present invention provides a plant living flavonol imaging detection method based on multi-excitation light chlorophyll fluorescence imaging technology, which has the advantages of high throughput, non-contact, non-destructive, digitized, visualized, and living imaging detection compared to traditional flavonol detection methods. In addition, the multi-excitation light chlorophyll fluorescence imaging technology can also simultaneously detect and analyze chlorophyll content, anthocyanin content (need to be configured with green excitation light) and plant N status, etc., so that it has the advantage of potential application value in the fields of medicinal plant phenotypic analysis, secondary metabolite imaging analysis, non-destructive rapid detection of Chinese medicinal materials, etc., and solves the problem that the flavonoid compound detection method in the prior art needs to be extracted, separated and determined by multiple physical and chemical methods. These methods often require pretreatment, crushing, extraction, separation and concentration of samples such as leaves, and the operation is complicated, time-consuming, labor-intensive, and requires destruction of samples.

[0005] (II) Technical solution To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a method for imaging and detecting flavonols in living plants based on multi-excitation chlorophyll fluorescence imaging technology, which uses FluorTron® multifunctional hyperspectral imaging analysis equipment to collect fluorescence signals. The system is equipped with a system for collecting fluorescence spectrum images of plant leaves; FluorVision hyperspectral fluorescence imaging analysis software for processing and analyzing fluorescence spectrum images; and ultraviolet and red light sources for exciting chlorophyll fluorescence. Therefore, it can simultaneously have (reflected light) hyperspectral imaging analysis, chlorophyll fluorescence hyperspectral imaging analysis, and UV-excited biofluorescence (UV-MCF) hyperspectral imaging analysis functions to fully decode biofluorescence. This patent is based on multi-excitation chlorophyll fluorescence hyperspectral imaging analysis. It can not only perform chlorophyll fluorescence imaging analysis, but also obtain its hyperspectral data cube and then analyze its spectral characteristics (spectral-resolved), so that the two-dimensional imaging analysis of chlorophyll fluorescence is upgraded to the level of hyperspectral imaging analysis (up to hundreds of spectral latitudes), filling the gap that the chlorophyll fluorescence imaging instrument products on the market can only analyze its fluorescence dynamics but cannot analyze chlorophyll fluorescence in the spectral dimension (spectral-resolved).

[0006] A method for imaging and detecting flavonols in living plants based on multi-excitation chlorophyll fluorescence imaging technology, comprising the following steps: 1) Use FluorTron® multifunctional hyperspectral imaging system to collect fluorescence spectral images of Ginkgo biloba leaves in a non-destructive and high-throughput manner; 2) Use FluorVision hyperspectral fluorescence imaging analysis software to perform denoising and fluorescence spectrum analysis on the collected spectral images; 3) FluorVision hyperspectral fluorescence imaging analysis software was used to analyze the flavonol model Flav index and generate images, thereby calculating the ginkgo leaf flavonol index and the distribution of flavonol content on the leaf surface.

[0007] Preferably, the chlorophyll fluorescence intensity excited by ultraviolet light is inversely proportional to the flavonol concentration in the epidermis of the ginkgo leaves, while the chlorophyll fluorescence intensity excited by red light is not affected by the flavonol concentration. By comparing the chlorophyll fluorescence signals under two different light excitations, according to the Beer-Lambert law, the flavonol index FLAV can be obtained, which is characterized in that in the step 1), the specific process is to collect ginkgo leaves with good sunlight irradiation conditions (light group) and ginkgo leaves in the shade all year round (shade group), adapt to indoor light for 15 minutes, and then place the sample on the FluorTron® system sample tray, put it into the FluorTron® system mainframe box to make it in a completely dark environment, and then turn on the ultraviolet light and 620nm red light excitation light sources respectively, control the FluorTron® system sample tray to automatically move to the multifunctional hyperspectral imaging workstation, and collect the chlorophyll fluorescence hyperspectral data cube excited by ultraviolet light and red light in turn; In addition, when acquiring images, in order to obtain more accurate sample fluorescence signal data for easy analysis and comparison, a dark reference image is acquired at the same time as the sample spectral image, that is, an image acquired in total darkness to represent the dark current noise of the camera.

[0008] Preferably, in step 2), the collected spectral image is denoised using FluorVision hyperspectral fluorescence imaging analysis software, which mainly includes dark current removal, threshold segmentation, noise filtering, background masking and other steps. The dark current removal correction formula is as follows: Ic = Iraw - Idark (1) Where Ic represents the corrected fluorescence spectrum signal data, Iraw represents the collected original fluorescence spectrum signal data, and Idark represents the fluorescence spectrum signal data collected in total darkness with the shutter closed.

[0009] Fluorescence spectrum processing includes: using FluorVision hyperspectral fluorescence imaging analysis software to obtain the region of interest ROI, the method is: automatically identify and accurately extract the region of interest (ROI) of the ginkgo leaf through the threshold segmentation method, and distinguish the leaf from the background. Extract the complete leaf ROI from the image after threshold segmentation, calculate the mean fluorescence value of all pixels in the ROI of the same group of samples, and obtain the average fluorescence spectrum curve within the same group. The calculation formula is as follows: (2) Where Flsi represents the average fluorescence spectrum of the i-th channel, ∑Si represents the fluorescence spectrum of the pixel in the i-th channel, and Ni represents the number of pixels in the region of interest of the i-th channel.

[0010] Preferably, in step 3), the principle of using FluorVision hyperspectral fluorescence imaging analysis software to analyze the Flav index of the flavonol model is: solar ultraviolet light irradiation can induce the production of flavonols, thereby causing the accumulation of flavonols on the leaf surface. Both ultraviolet light and visible light (such as red light) can excite chlorophyll fluorescence, but flavonols can absorb ultraviolet light, blocking it from penetrating into the chloroplasts of mesophyll cells to excite fluorescence, while red light is not absorbed by flavonols and can be used as a reference excitation light. Based on this, the ultraviolet and red light dual excitation chlorophyll fluorescence technology (CHlFES) can be used to detect the flavonol content (Agati et al., 2011; Cerovic et al., 2012; Goulas et al., 2004; Julkunen-Tiitto et al., 2014), and its algorithm model is: FLAV=log(FRF_R / FRF_UV)(3) Among them, FRF_R is the far-red fluorescence (740nm) excited by red light, and FRF_UV is the far-red fluorescence excited by ultraviolet light; this index can reflect the flavonol content in the leaf epidermis.

[0011] (III) Beneficial effects Compared with the prior art, the present invention provides a method for imaging and detecting flavonols in living plants based on multi-excitation chlorophyll fluorescence imaging technology, which has the following beneficial effects: 1. This method for detecting flavonols in living plants based on multi-excitation chlorophyll fluorescence imaging technology simplifies the operation process and improves detection efficiency through an automated imaging system. It is suitable for detecting the flavonol content of a variety of plants and has good applicability and promotion value.

[0012] 2. This method of plant flavonol imaging detection based on multi-excitation chlorophyll fluorescence imaging technology can be used in a conveyor belt system to detect large batches of samples at the same time, and the pipeline operation can significantly improve the throughput and efficiency of detection and screening. It has extremely high application value for the rapid screening and analysis of large-scale plant samples.

[0013] 3. This method of plant living flavonol imaging detection based on multi-excitation chlorophyll fluorescence imaging technology can automatically extract continuous fluorescence spectrum data in the 400-1000nm band by using FluorTron® multifunctional imaging technology and FluorVision hyperspectral fluorescence imaging analysis software technology, and calculate the flavonoid index Flav based on it, thus realizing the digital expression of flavonol content. 4. This method of plant living flavonol imaging detection based on multi-excitation chlorophyll fluorescence imaging technology can intuitively and visually display the distribution of flavonols in plants through imaging technology, which helps to better understand its biological functions and mechanisms of action, and is also conducive to evaluating the medicinal value and quality of plants.

[0014] 5. This method of plant living flavonol imaging detection based on multi-excitation light chlorophyll fluorescence imaging technology avoids chemical extraction and physical damage to plant tissues through non-contact imaging detection, has no effect on plant growth, and can continuously and periodically detect the growth status of plants. It is suitable for detection throughout the entire plant life cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 : Schematic diagram of the process of this detection method; Figure 2 :FluorTron® multifunctional hyperspectral imaging analysis system; Figure 3 : Hyperspectral fluorescence imaging data cube; Figure 4 :FluorVision® hyperspectral fluorescence imaging analysis software; Figure 5 :UV-excited chlorophyll fluorescence spectrum curve; Figure 6 :Red light excited chlorophyll fluorescence spectrum curve; Figure 7 : Flavonol index of sample leaves; Figure 8 : Distribution diagram of flavonol content. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] See also Figure 1-8, a method for imaging and detecting flavonols in living plants based on multi-excitation chlorophyll fluorescence imaging technology. The method uses FluorTron® multifunctional hyperspectral imaging analysis equipment to collect fluorescence signals. The system has a system for collecting plant leaf fluorescence spectrum images; FluorVision hyperspectral fluorescence imaging analysis software for processing and analyzing fluorescence spectrum images; and ultraviolet and red light sources for exciting chlorophyll fluorescence. Therefore, it can simultaneously have (reflected light) hyperspectral imaging analysis, chlorophyll fluorescence hyperspectral imaging analysis, and UV-excited bioluminescence (UV-MCF) hyperspectral imaging analysis functions to fully decode bioluminescence. This patent is based on multi-excitation chlorophyll fluorescence hyperspectral imaging analysis, which can not only perform chlorophyll fluorescence imaging analysis, but also obtain its hyperspectral data cube and then analyze its spectral characteristics (spectral-resolved), so that the two-dimensional imaging analysis of chlorophyll fluorescence is upgraded to the level of hyperspectral imaging analysis (up to hundreds of spectral latitudes), filling the gap that the chlorophyll fluorescence imaging instrument products on the market can only analyze its fluorescence dynamics but cannot analyze chlorophyll fluorescence in the spectral dimension (spectral-resolved).

[0018] A method for imaging and detecting flavonols in living plants based on multi-excitation chlorophyll fluorescence imaging technology, comprising the following steps: 1) Use FluorTron® multifunctional hyperspectral imaging system to collect fluorescence spectral images of Ginkgo biloba leaves in a non-destructive and high-throughput manner; 2) Use FluorVision hyperspectral fluorescence imaging analysis software to perform denoising and fluorescence spectrum analysis on the collected spectral images; 3) FluorVision hyperspectral fluorescence imaging analysis software was used to analyze the flavonol model Flav index and generate images, thereby calculating the ginkgo leaf flavonol index and the distribution of flavonol content on the leaf surface.

[0019] That is, the chlorophyll fluorescence intensity excited by ultraviolet light is inversely proportional to the flavonol concentration in the epidermis of ginkgo leaves, while the chlorophyll fluorescence intensity excited by red light is not affected by the flavonol concentration. By comparing the chlorophyll fluorescence signals under two different light excitations, according to the Beer-Lambert law, the flavonol index FLAV can be obtained. Its characteristic step 1) is that the specific process is to collect ginkgo leaves with good sunlight irradiation conditions (light group) and ginkgo leaves in the shade all year round (shade group), adapt to indoor light for 15 minutes, and then place the samples on the FluorTron® system sample tray, close the FluorTron® system mainframe box to make it in a completely dark environment, and then turn on the ultraviolet light and 620nm red light excitation light sources respectively, control the FluorTron® system sample tray to automatically move to the multifunctional hyperspectral imaging workstation, and collect the chlorophyll fluorescence hyperspectral data cubes excited by ultraviolet light and red light in turn; In addition, when acquiring images, in order to obtain more accurate sample fluorescence signal data for easy analysis and comparison, a dark reference image is acquired at the same time as the sample spectral image, that is, an image acquired in total darkness to represent the dark current noise of the camera.

[0020] In step 2), the collected spectral image is denoised using FluorVision hyperspectral fluorescence imaging analysis software, which mainly includes dark current removal, threshold segmentation, noise filtering, background masking and other steps. The dark current removal correction formula is as follows: Ic = Iraw - Idark (1) Where Ic represents the corrected fluorescence spectrum signal data, Iraw represents the collected original fluorescence spectrum signal data, and Idark represents the fluorescence spectrum signal data collected in total darkness with the shutter closed.

[0021] Fluorescence spectrum processing includes: using FluorVision hyperspectral fluorescence imaging analysis software to obtain the region of interest ROI, the method is: automatically identify and accurately extract the region of interest (ROI) of the ginkgo leaf through the threshold segmentation method, and distinguish the leaf from the background. Extract the complete leaf ROI from the image after threshold segmentation, calculate the mean fluorescence value of all pixels in the ROI of the same group of samples, and obtain the average fluorescence spectrum curve within the same group. The calculation formula is as follows: (2) Where Flsi represents the average fluorescence spectrum of the i-th channel, ∑Si represents the fluorescence spectrum of the pixel in the i-th channel, and Ni represents the number of pixels in the region of interest of the i-th channel.

[0022] In step 3), the principle of using FluorVision hyperspectral fluorescence imaging analysis software to analyze the Flav index of the flavonol model is: solar ultraviolet light irradiation can induce the production of flavonols, thereby leading to the accumulation of flavonols on the leaf surface. Both ultraviolet light and visible light (such as red light) can excite chlorophyll fluorescence, but flavonols can absorb ultraviolet light, blocking it from penetrating into the chloroplasts of mesophyll cells to excite fluorescence, while red light is not absorbed by flavonols and can be used as a reference excitation light. Based on this, the ultraviolet and red light dual excitation chlorophyll fluorescence technology (CHlFES) can be used to detect the flavonol content (Agati et al., 2011; Cerovic et al., 2012; Goulas et al., 2004; Julkunen-Tiitto et al., 2014), and its algorithm model is: FLAV=log(FRF_R / FRF_UV)(3) Among them, FRF_R is the far-red fluorescence (740nm) excited by red light, and FRF_UV is the far-red fluorescence excited by ultraviolet light; this index can reflect the flavonol content in the leaf epidermis.

[0023] In summary, the in vivo flavonol imaging detection method of plants based on multi-excitation chlorophyll fluorescence imaging technology has the following specific procedures: collecting ginkgo leaves with good sunlight exposure conditions (light group) and ginkgo leaves in the shade all year round (shade group), adapting to indoor light for 15 minutes, and then placing the samples on the FluorTron® system sample tray. The main body of the FluorTron® system is closed to make it in a completely dark environment. Then, the flavonol imaging protocol is run, and the ultraviolet light and 620nm red light excitation light sources are turned on respectively. The FluorTron® system sample tray is controlled to automatically move to the multi-functional hyperspectral imaging workstation, and the chlorophyll fluorescence hyperspectral data cubes excited by ultraviolet light and red light are collected in turn. The FluorVision hyperspectral fluorescence imaging analysis software is used for denoising and fluorescence spectrum analysis, the flavonol model Flav index is analyzed and plotted, and the ginkgo leaf flavonol index and the distribution of flavonol content on the leaf surface are calculated. This method uses multi-excitation chlorophyll fluorescence hyperspectral imaging analysis technology to establish a new in vivo imaging detection method for plant flavonol content distribution, which has broad application potential in plant phenotypic analysis, secondary metabolite imaging analysis and other fields. In addition, this method can also be extended to other plant physiological and ecological research and analysis, resistance screening, high-light-efficiency and excellent variety screening, etc.

[0024] The detection method is further described below in conjunction with specific implementation methods.

[0025] 1. Materials and Methods 1.1 FluorTron® Multifunctional Hyperspectral Fluorescence Imaging Analysis System In this embodiment, the data acquisition device is a FluorTron® multifunctional hyperspectral imaging analysis system developed based on multi-excitation chlorophyll fluorescence measurement and imaging technology and hyperspectral fluorescence imaging technology.

[0026] The system integrates hyperspectral imaging, chlorophyll fluorescence imaging, UV-MCF bioluminescence imaging and spectral analysis. It adopts a fully enclosed box design, which helps to create a stable and uniform light field, a completely dark measurement environment, eliminates the interference of ambient light and stray light, and ensures the quality and accuracy of hyperspectral fluorescence imaging. The system uses a visible near-infrared band hyperspectral imaging camera with linear array push-scanning imaging, which can continuously collect images at different wavelengths to obtain a 400-1000nm band hyperspectral data cube. The system is equipped with a variety of excitation light sources such as UV, red light, and blue light, with symmetrical incidence and 0-100% linear stepless regulation of light intensity. This embodiment uses two excitation light sources, UV and red light. It is powered by an uninterruptible power supply to ensure the stability of the power supply during long-term measurements. The system is equipped with a computer for platform control, data acquisition, processing and storage of hyperspectral fluorescence data.

[0027] The leaf is placed on the sample tray of the FluorTron® multifunctional hyperspectral imaging analysis system. The sample tray is made of black anodized aluminum and is sandblasted to produce a tiny concave-convex texture on its surface to minimize background noise and ensure that the collected fluorescence signal comes only from the leaf itself. When the sample passes through the camera line scan area, the synchronous control software performs data acquisition to obtain a hyperspectral fluorescence image of the object being measured in the range of 400-1000nm.

[0028] In addition, the processing and analysis of the hyperspectral fluorescence data collected in this embodiment are completed using professional FluorVision hyperspectral fluorescence imaging analysis software, which can analyze multiple fluorescence parameters with one click, output result charts, and intuitively obtain visual results and data.

[0029] 1.2 Materials In this example, fresh ginkgo leaves were picked in the Entrepreneurship and Research Park of Xi'an High-tech Zone. Ginkgo leaves with good sunlight exposure conditions (light group) and ginkgo leaves in the shade all year round (shade group) were collected. The samples were stored in an incubator at a temperature of 4 degrees Celsius to maintain their freshness and biological activity. They were transported to a non-destructive testing laboratory to avoid physical damage to the leaves. After returning to temperature, the ginkgo leaves were placed on a non-fluorescent sample tray of the FluorTron® multifunctional hyperspectral imaging analysis system for image acquisition.

[0030] 1.3 Hyperspectral fluorescence data acquisition and processing 1.3.1 Hyperspectral fluorescence image acquisition Ginkgo leaves were adapted to indoor light for 15 minutes, then placed in the sample tray of the FluorTron® multifunctional hyperspectral imaging system, and the flavonol imaging protocol was run in total darkness: the sample ginkgo leaves were spectrally imaged, with the distance between the imaging sensor and the sample being 20 cm, the exposure time being 10 ms, and the sample moving speed being 7 mm / s. Under UV and red light irradiation, 400-1000 nm band hyperspectral fluorescence imaging data were collected, respectively, and two 350×350×224 hyperspectral fluorescence data cubes were finally obtained. The hyperspectral fluorescence imaging data contains both image information at each wavelength and fluorescence spectrum information corresponding to each pixel point, and the spectral data of the ginkgo leaf sample set was obtained. This laid the foundation for analyzing the content of flavonols in ginkgo leaves and their distribution on the leaf surface.

[0031] 1.3.2 Hyperspectral fluorescence spectrum curve extraction By using the difference in fluorescence signals between leaves and background areas, the threshold segmentation method was used in the FluorVision hyperspectral fluorescence imaging analysis software to automatically identify and accurately extract the region of interest (ROI) of the ginkgo leaves, and distinguish the leaves from the background. The complete leaf ROI was extracted from the image after threshold segmentation, and the fluorescence values ​​of all pixels in the ROI of the same group of samples were averaged to obtain the average fluorescence spectrum curve within the same group. The calculation formula is as follows: ((1) Where Flsi represents the average fluorescence spectrum of the i-th channel, ∑Si represents the fluorescence spectrum of the pixel in the i-th channel, and Ni represents the number of pixels in the region of interest of the i-th channel.

[0032] 1.3.3 Fluorescence signal correction of fluorescence spectral image In order to obtain more accurate sample fluorescence signal data for analysis and comparison, a dark reference image is collected while collecting the sample spectral image, that is, an image collected in total darkness to represent the dark current noise of the camera. The correction formula is as follows: Ic=Iraw-Idark(2) Where Ic represents the corrected fluorescence spectrum image, Iraw represents the collected original fluorescence spectrum image, and Idark represents the fluorescence spectrum image collected when the shutter is closed in complete darkness.

[0033] 1.3.4 Flavonol Model Flav Index Solar ultraviolet light can induce the production of flavonols, resulting in the accumulation of flavonols on the leaf surface. Both ultraviolet light and visible light (such as red light) can excite chlorophyll fluorescence, but flavonols can absorb ultraviolet light, blocking it from penetrating into the chloroplasts of mesophyll cells to excite fluorescence, while red light is not absorbed by flavonols and can be used as a reference excitation light. Based on this, the dual excitation chlorophyll fluorescence technique (CHlFES) can be used to detect the flavonol content (Agati et al., 2011; Cerovic et al., 2012; Goulas et al., 2004; Julkunen-Tiitto et al., 2014). The algorithm model is: FLAV=log(FRF_R / FRF_UV)(3) Among them, FRF_R is the far-red fluorescence (740nm) excited by red light, and FRF_UV is the far-red fluorescence excited by ultraviolet light; this index can reflect the flavonol content in the leaf epidermis.

[0034] 2 Results and analysis 2.1 Ginkgo leaf fluorescence spectrum information extraction and preprocessing The sample materials in this embodiment were collected three times, and the ginkgo leaves of the illumination group and the shade group were collected in August, September and October respectively. All pixels in the leaves of the same group and the same time were taken as the region of interest (ROI), and the average fluorescence spectral signal value of all pixels in the ROI in each band was calculated one by one. The fluorescence signal values ​​of the ginkgo leaves collected in the same group and the same month were averaged and the fluorescence spectral curve was drawn.

[0035] The fluorescence spectrum showed that the mean value of far-red fluorescence (740nm) excited by ultraviolet light was significantly higher in the shaded group than in the light group, while the far-red fluorescence excited by red light had no significant difference among different groups.

[0036] 2.2 Evaluation and prediction of flavonol content using hyperspectral fluorescence imaging technology The fluorescence spectrum data were substituted into the formula FLAV = log(ChlF_R / ChlF_UV) (1) to calculate the Ginkgo leaf flavonol index and the distribution map of flavonol content on the leaf surface.

[0037] The results showed that the average Flav value of Ginkgo leaves under light conditions was 1.397 (standard deviation 0.084), which was significantly greater than the average value of 0.861 (standard deviation 0.044) under shade conditions.

[0038] 3. Conclusion The following conclusions can be drawn from the plant living flavonol imaging detection method based on multi-excitation light chlorophyll fluorescence imaging technology proposed in the present invention: 1) It was verified that the flavonol content of ginkgo leaves under light conditions (exposed to more ultraviolet rays) (light group) was higher than that of ginkgo leaves under shade conditions (shade group).

[0039] 2) Using multi-excitation chlorophyll fluorescence hyperspectral imaging analysis technology, the distribution of plant flavonol content can be imaged and analyzed in vivo, and anthocyanin index Anth, chlorophyll index and nitrogen status NBI can also be imaged and analyzed.

[0040] The theoretical basis of the plant living flavonol imaging detection method based on multi-excitation light chlorophyll fluorescence imaging technology of the present invention is: solar ultraviolet light irradiation of plant leaves can induce the production of flavonols, thereby leading to the accumulation of flavonols on the surface of the leaves. Both ultraviolet light and visible light (such as red light) can excite chlorophyll fluorescence, but flavonols can absorb ultraviolet light and block it from penetrating into the chloroplasts of mesophyll cells to excite fluorescence, while red light is not absorbed by flavonols and can be used as a reference excitation light. Based on this, dual-excitation chlorophyll fluorescence technology (CHlFES) can be used to detect flavonol content (Agati et al., 2011; Cerovic et al., 2012; Goulas et al., 2004; Julkunen-Tiitto et al., 2014), and its algorithm model is FLAV=log (FRF_R / FRF_UV), wherein FRF_R is far-red fluorescence (740nm) excited by red light, and FRF_UV is far-red fluorescence excited by ultraviolet light; this index can reflect the flavonol content in the leaf epidermis. Force-A of France has produced products such as DUALEX and MultiPLEX based on this principle. DUALEX is a point measurement instrument that measures by clamping the leaves. The measurement area is relatively small, only a circular area with a diameter of 5 mm. MultiPLEX is also a point measurement instrument that can provide a variety of measurement area options. The measurement area includes 50 cm² (8 cm diameter), 28 cm² (6 cm diameter) or 12.5 cm² (4 cm diameter) to measure the flavonol index Flav and anthocyanin index Anth of living plant leaves or canopies. Although DUALEX and MultiPLEX provide fast and non-destructive measurement methods, they are only suitable for measuring smaller leaves, such as living leaves with an area of ​​less than 50 cm2. They are powerless for measuring larger complete leaves, canopies, etc., and cannot perform detailed spatial distribution analysis of the flavonol content at each point on the leaves, and cannot intuitively and visually characterize the content and distribution of flavonols. At present, there is no instrument technology on the market that can perform imaging analysis of the flavonol content in plant leaves or canopies.

[0041] In order to solve the above problems existing in the prior art, the present invention studies a method for imaging and detecting flavonols in living plants based on multi-excitation chlorophyll fluorescence imaging technology. This method uses a specific light source (red light, ultraviolet light) equipped with the FluorTron multifunctional hyperspectral imaging analysis system to excite chlorophyll fluorescence, and captures 400-1000nm fluorescence signals through a highly sensitive hyperspectral imaging sensor. This method can not only perform chlorophyll fluorescence imaging analysis, but also obtain its hyperspectral data cube and then analyze its spectral characteristics (spectral-resolved), so that the two-dimensional imaging analysis of chlorophyll fluorescence is upgraded to the level of hyperspectral imaging analysis (up to hundreds of spectral latitudes). Combining the advantages of hyperspectral imaging technology and chlorophyll fluorescence imaging technology such as non-destructive, high-throughput, digital, and visualized in vivo detection, the distribution and imaging detection of flavonol content in leaves and canopies of plants under natural growth conditions are realized. The specific steps include: 1) Sample preparation: Ginkgo leaves from the sunlight group (light group) and the ginkgo leaves from the shade group (shade group) were collected and adapted to indoor light for 15 minutes.

[0042] 2) Fluorescence signal acquisition: Place the prepared ginkgo leaves on the sample stage of the FluorTron® multifunctional hyperspectral imaging system, run the flavonol imaging protocol, turn on the ultraviolet light and 620nm red light excitation light sources respectively, and collect the chlorophyll fluorescence hyperspectral data cubes excited by ultraviolet light and red light in turn.

[0043] 3) Image processing and analysis: FluorVision hyperspectral fluorescence imaging analysis software was used to take all pixels in the leaves as the region of interest (ROI), and the average fluorescence spectral signal value of all pixels in the ROI in each band was calculated one by one. The fluorescence signal values ​​of leaves under the same illumination conditions in the same group were averaged and the fluorescence spectral curve was drawn ( Figure 1 ). And analyze the Flav index and make a map to achieve a visual analysis of the distribution of flavonol content It should be noted that the term "comprises" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article, or device that includes a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such process, method, article, or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article, or device that includes the element.

[0044] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for detecting flavonols in living plants based on multi-excitation chlorophyll fluorescence spectrum imaging technology, characterized in that: This method uses FluorTron® multifunctional hyperspectral imaging analysis technology to collect multi-excitation chlorophyll fluorescence signals. The system has the function of collecting fluorescence spectrum images of plant leaves, fruits and vegetables, and also has (reflected light) hyperspectral imaging function; it has FluorVision hyperspectral fluorescence imaging analysis software for processing and analyzing fluorescence spectrum images; it has ultraviolet light, red light source and green light source (multi-excitation light) for stimulating chlorophyll fluorescence. Therefore, it can simultaneously have (reflected light) hyperspectral imaging analysis, chlorophyll fluorescence hyperspectral imaging analysis, UV-excited biofluorescence (UV-MCF) hyperspectral imaging analysis functions, and fully decode biofluorescence. This patent is based on multi-excitation chlorophyll fluorescence hyperspectral imaging analysis. It can not only perform chlorophyll fluorescence imaging analysis, but also obtain its hyperspectral data cube and then analyze its spectral characteristics (spectral-resolved), so that the two-dimensional imaging analysis of chlorophyll fluorescence is upgraded to the level of hyperspectral imaging analysis (up to hundreds of spectral latitudes), filling the gap that the chlorophyll fluorescence imaging instrument products on the market can only analyze its fluorescence dynamics but cannot analyze chlorophyll fluorescence in the spectral dimension (spectral-resolved).

2. A method for detecting flavonols in living plants based on multi-excitation chlorophyll fluorescence imaging technology according to claim 1, wherein the characteristic steps are as follows: include: 1) Use FluorTron® multifunctional hyperspectral imaging system to collect fluorescence spectral images of Ginkgo biloba leaves in a non-destructive and high-throughput manner; 2) Use FluorVision hyperspectral fluorescence imaging analysis software to perform denoising and fluorescence spectrum analysis on the collected spectral images; 3) FluorVision hyperspectral fluorescence imaging analysis software was used to analyze the flavonol model Flav index and generate images, thereby calculating the ginkgo leaf flavonol index and the distribution of flavonol content on the leaf surface.

3. A method for detecting flavonols in living plants based on multi-excitation chlorophyll fluorescence imaging technology according to claim 1, that is, the chlorophyll fluorescence intensity excited by ultraviolet light is inversely proportional to the flavonol concentration in the epidermis of ginkgo leaves, and the chlorophyll fluorescence intensity excited by red light is not affected by the flavonol concentration. By comparing the chlorophyll fluorescence signals under two different light excitations, according to the Beer-Lambert law, the flavonol index FLAV can be obtained, wherein the step 1) (the application case of flavonols in living ginkgo leaves) is characterized in that the specific process is to collect ginkgo leaves with good sunlight irradiation conditions (light group) and ginkgo leaves in the shade all year round (shade group), adapt to indoor light for 15 minutes, then place the sample on the FluorTron® system sample tray, put it into the FluorTron® system mainframe box to make it in a completely dark environment, then turn on the ultraviolet light and 620nm red light excitation light sources respectively, control the FluorTron® system sample tray to automatically move to the multifunctional hyperspectral imaging workstation, and collect the chlorophyll fluorescence hyperspectral data cubes excited by ultraviolet light and red light in turn; In addition, when acquiring images, in order to obtain more accurate sample fluorescence signal data for easy analysis and comparison, a dark reference image is acquired at the same time as the sample spectral image, that is, an image acquired in total darkness to represent the dark current noise of the camera.

4. The method for detecting flavonols in living plants based on multi-excitation light chlorophyll fluorescence imaging technology according to claim 1, wherein the In step 2), the collected spectral image is denoised using FluorVision hyperspectral fluorescence imaging analysis software, which mainly includes dark current removal, threshold segmentation, noise filtering, background masking and other steps. The dark current removal correction formula is as follows: Ic = Iraw - Idark (1) Where Ic represents the corrected fluorescence spectrum signal data, Iraw represents the collected original fluorescence spectrum signal data, and Idark represents the fluorescence spectrum signal data collected in total darkness with the shutter closed; Fluorescence spectrum processing includes: using FluorVision hyperspectral fluorescence imaging analysis software to perform image segmentation and obtain the region of interest ROI. The method is: automatically identify and accurately extract the region of interest (ROI) of the ginkgo leaf by the threshold segmentation method to distinguish the leaf from the background. Extract the complete leaf ROI from the image after threshold segmentation, calculate the mean fluorescence value of all pixels in the ROI of the same group of samples, and obtain the average fluorescence spectrum curve within the same group. The calculation formula is as follows: （2） Where Flsi represents the average fluorescence spectrum of the i-th channel, ∑Si represents the fluorescence spectrum of the pixel in the i-th channel, and Ni represents the number of pixels in the region of interest of the i-th channel.

5. The method for detecting flavonols in living plants based on multi-excitation light chlorophyll fluorescence imaging technology according to claim 1, wherein the In step 3), the principle of using FluorVision hyperspectral fluorescence imaging analysis software to analyze the Flav index of the flavonol model is: solar ultraviolet light irradiation can induce the production of flavonols, thereby leading to the accumulation of flavonols on the leaf surface. Both ultraviolet light and visible light (such as red light) can excite chlorophyll fluorescence, but flavonols can absorb ultraviolet light, blocking it from penetrating into the chloroplasts of mesophyll cells to excite fluorescence, while red light is not absorbed by flavonols and can be used as a reference excitation light. Based on this, the ultraviolet and red light dual excitation chlorophyll fluorescence technology (CHlFES) can be used to detect the flavonol content (Agati et al., 2011; Cerovic et al., 2012; Goulas et al., 2004; Julkunen-Tiitto et al., 2014), and its algorithm model is: FLAV=log(FRF_R / FRF_UV)(3) Among them, FRF_R is the far-red fluorescence (740nm) excited by red light, and FRF_UV is the far-red fluorescence excited by ultraviolet light; this index can reflect the flavonol content in the leaf epidermis. It should be noted that chlorophyll fluorescence has two peaks, namely the 685nm red peak and the 740nm far-red peak. In view of the phenomenon that the 685nm chlorophyll fluorescence is reabsorbed by chlorophyll, the model selects the 740nm far-red peak.

Images