Intelligent plant vitality monitoring system and method based on multi-dimensional perception compound eye imaging

Abstract

The present invention relates to a monitoring system and a monitoring method for plant germination vitality, and specifically to an intelligent monitoring system and method for plant vitality based on multi-dimensional perception compound eye imaging, in order to solve the shortcomings of low accuracy and reliability of plant vitality monitoring in the prior art, as well as poor integration, real-time and intelligence. The intelligent monitoring system for plant vitality based on multi-dimensional perception compound eye imaging of the present invention includes an information acquisition module, a light source module and an intelligent monitoring visualization platform. The information acquisition module is configured as a multi-dimensional perception compound eye imaging device, and its lens includes a plurality of sub-eyes arranged on a hemispherical shell for acquiring multi-viewing images. The intelligent monitoring visualization platform connects the light source module and the information acquisition module, controls them, and solves and obtains growth status information. At the same time, a monitoring method based on the above-mentioned monitoring system is also provided, which obtains growth status information losslessly, accurately and reliably by using multi-dimensional information in multiple ways.

Description

Technical Field

[0001] The present invention relates to a monitoring system and a monitoring method for plant germination vitality, and in particular to an intelligent monitoring system and a method for plant vitality based on multi-dimensional perception compound eye imaging. Background Art

[0002] Traditional plant growth monitoring methods are time-consuming and labor-intensive, and have subjective uncertainties, poor repeatability, and high experimental pollution. Therefore, in recent years, with the continuous development of optoelectronic devices, botanists have begun to explore non-destructive monitoring technologies for plant vitality, including non-destructive monitoring based on optical properties and non-destructive monitoring based on appearance characteristics. Non-destructive monitoring based on optical properties includes near-infrared spectroscopy technology, hyperspectral imaging technology, and laser speckle technology; non-destructive monitoring based on appearance characteristics includes machine vision technology, which obtains plant growth conditions by processing the acquired images. Near-infrared spectroscopy technology conducts qualitative and quantitative analysis of plant vitality by studying the absorption characteristics of hydrogen-containing groups in organic matter to near-infrared spectra, but it is easily interfered by other factors, especially water. Hyperspectral imaging technology non-destructively monitors plant vitality through intuitive images and spectral data, and its monitoring results are not accurate enough; laser speckle technology has large differences in monitoring results for different varieties of plants; machine vision technology has high requirements for the measurement environment, and it is difficult to judge the strength of plant vitality for plants with unclear characteristics. In addition, the above technologies have a single monitoring function and cannot take into account monitoring other than their own functions. They cannot monitor the growth rate of plants at different growth stages. For plants cultured in water, the mirror reflection caused by the water surface will reduce the accuracy and reliability of monitoring. At the same time, in order to obtain multi-perspective information, the monitoring equipment needs to be rotated or moved. Therefore, the existing monitoring technology has poor integration, real-time and intelligence. Summary of the invention

[0003] The purpose of the present invention is to solve the shortcomings of the prior art in plant vitality monitoring, such as low accuracy and reliability, and poor integration, real-time performance and intelligence, and to provide a plant vitality intelligent monitoring system and method based on multi-dimensional perception compound eye imaging.

[0004] To achieve the above purpose, the technical solution provided by the present invention is as follows:

[0005] A plant vitality intelligent monitoring system based on multi-dimensional perception compound eye imaging includes an information acquisition module, wherein the information acquisition module is configured as a multi-dimensional perception compound eye imaging device for acquiring multi-view images of a monitoring target; the lens of the multi-dimensional perception compound eye imaging device includes a plurality of sub-eyes arranged on a hemispherical shell; the system is special in that:

[0006] It also includes light source modules and intelligent monitoring visualization platform;

[0007] The light source module and the multi-dimensional sensing compound eye imaging device are arranged in an incubator for culturing monitoring targets, and the light source module is used to evenly illuminate the monitoring targets in the incubator;

[0008] Definition: Each sub-eye and its surrounding adjacent sub-eyes are a cluster eye. The cluster eye is used to obtain multi-view images of the same monitoring target. Filters or polarizers are respectively set on the sub-eyes in the same cluster eye.

[0009] The intelligent monitoring visualization platform includes a data display and processing module and a monitoring data information management module; the data display and processing module is connected to the light source module and the information acquisition module, controls the operation of the light source module and the information acquisition module, obtains corresponding multi-dimensional information according to the multi-view image, solves the multi-dimensional information to obtain the growth status information of the corresponding monitoring target, and displays it on a visualization interface, wherein the multi-dimensional information includes spectral information, polarization information, image information and multi-view information; the monitoring data information management module is connected to the data display and processing module, and is used for recording, saving, statistically analyzing and querying multi-view images, multi-dimensional information and growth status information.

[0010] Further, the plurality of sub-eyes are arranged on the hemispherical shell in a hexagonal honeycomb structure, and each sub-eye is respectively arranged at the six vertices and the center of the hexagonal honeycomb structure; it is defined that the seven sub-eyes in each hexagonal honeycomb structure are a cluster eye, and the cluster eye is used to obtain multi-view images of the same monitoring target, and filters or polarizers are respectively arranged on the seven sub-eyes in the same cluster eye;

[0011] The intelligent monitoring visualization platform also includes a vitality prediction module, which trains the collected multidimensional information and the growth status information of the corresponding monitoring target to obtain a vitality prediction model and a prediction curve of the monitoring target, and in subsequent cultivation, obtains a growth status information prediction value based on the multidimensional information obtained in the early cultivation of the monitoring target.

[0012] Furthermore, it also includes an early warning module, which is connected to the data display and processing module and reminds the monitoring time or the cultivation operation time according to the growth status information obtained by the data display and processing module.

[0013] Furthermore, the working wavelength of the multi-dimensional perception compound eye imaging device is in the visible light band and the near infrared band;

[0014] The light source module includes a plurality of light sources, and the light sources are configured as annular light sources or coaxial light sources, and are used to provide light in the visible light band and the near infrared band.

[0015] Furthermore, filters with wavelengths at R, G, B and NIR, and linear polarizers with angles of 0°, 60° and 120° are respectively arranged on the seven sub-eyes in the same cluster eye.

[0016] Furthermore, the vitality prediction module trains the collected multi-dimensional information and the corresponding growth status information of the monitoring target through an AI algorithm.

[0017] At the same time, the present invention also provides a method for intelligent monitoring of plant vitality based on multi-dimensional perception compound eye imaging, which adopts the above-mentioned intelligent monitoring system for plant vitality based on multi-dimensional perception compound eye imaging, and its special feature is that it includes the following steps:

[0018] Step 1: at different moments of the growth of the monitored target, controlling the multi-dimensional perception compound eye imaging device to shoot the monitored target to obtain corresponding multi-view images;

[0019] Step 2, the data display and processing module obtains corresponding multi-dimensional information according to the multi-view images at different times, and stores the multi-dimensional information in the monitoring data information management module;

[0020] Step 3: The data display and processing module obtains the growth status information of the monitored target based on the multi-dimensional information at different times.

[0021] Furthermore, step 3 is specifically as follows:

[0022] 3.1. The data display and processing module identifies the edge of the monitored target based on the image information at different times and calculates the corresponding size;

[0023] 3.2. The data display and processing module processes the spectral information at different times, or processes the spectral information and polarization information at different times, draws the corresponding multi-spectral radiation curve, obtains the composition and / or chlorophyll content of the monitored target at different times according to the multi-spectral radiation curve, and draws the growth curve in combination with the image information to obtain the optimal growth time; the composition and / or chlorophyll content of all monitored targets within the field of view are counted, and the seed germination rate of the monitored target is obtained in combination with the image information;

[0024] 3.3. The data display and processing module obtains the real-time height and hypocotyl shade avoidance of the monitored target based on the multi-view information, and then obtains its growth rate based on the real-time height; the data display and processing module obtains the root growth of the monitored target in water-based culture based on the multi-view information, spectral information, image information and polarization information;

[0025] Among them, step 3.1 to step 3.3 are performed simultaneously or sequentially.

[0026] Furthermore, step 3 also includes:

[0027] 3.4. The vitality prediction module obtains the growth status information of the monitored target based on the multi-dimensional information through the vitality prediction model and prediction curve therein.

[0028] Furthermore, in step 1, the working wavelength selection method of the multi-dimensional perception compound eye imaging device is specifically as follows:

[0029] The data display and processing module processes the spectral information of different categories of monitoring targets, obtains the corresponding characteristic spectra, establishes a characteristic spectrum database, stores it in the monitoring data information management module, and selects the working wavelength of the multi-dimensional perception compound eye imaging device according to the characteristic spectra of different categories of monitoring targets.

[0030] Beneficial effects of the present invention:

[0031] 1. The plant vitality intelligent monitoring system based on multi-dimensional perception compound eye imaging of the present invention realizes the high integration and intelligence of the monitoring system through the cooperation of information acquisition module and intelligent monitoring visualization platform.

[0032] 2. In the plant vitality intelligent monitoring system based on multi-dimensional perception compound eye imaging of the present invention, the information acquisition module is configured as a multi-dimensional perception compound eye imaging device, whose lens includes a plurality of sub-eyes arranged in a hexagonal manner on a hemispherical shell, so as to realize information acquisition with a large field of view, and a multi-perspective image of all plants in the incubator can be obtained through a single exposure. The cluster eyes in different areas based on the same multi-dimensional perception compound eye imaging device ensure the monitoring accuracy and consistency of all plants, avoid system errors, and improve the monitoring efficiency at the same time.

[0033] 3. The plant vitality intelligent monitoring method based on multi-dimensional perception compound eye imaging of the present invention can obtain the growth status information of the monitored target non-destructively, accurately and reliably through the multi-channel use of multi-dimensional information, thereby realizing the monitoring and evaluation of plant vitality.

[0034] 4. The plant vitality intelligent monitoring method based on multi-dimensional perception compound eye imaging of the present invention utilizes a multi-dimensional perception compound eye imaging device to shoot at different times of the growth of the monitored target to obtain corresponding multi-view images, and then solves the corresponding growth information through an intelligent monitoring visualization platform, which can adapt to the monitoring of various plants cultivated in soil or water. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a structural schematic diagram of Embodiment 1 of the plant vitality intelligent monitoring system based on multi-dimensional perception compound eye imaging of the present invention;

[0036] Figure 2 It is a schematic diagram of the working principle of the information collection module in the first embodiment of the plant vitality intelligent monitoring system based on multi-dimensional perception compound eye imaging of the present invention;

[0037] Figure 3 This is a flow chart of Example 1 of the method for intelligent monitoring of plant vitality based on multi-dimensional perception compound eye imaging of the present invention.

[0038] Description of reference numerals:

[0039] 1-light source module, 2-information acquisition module, 3-monitoring target, 4-incubator, 5-culture medium, 6-intelligent monitoring visualization platform, 7-early warning module. DETAILED DESCRIPTION

[0040] Embodiment 1

[0041] The present invention proposes a practical plant vitality intelligent monitoring system and method based on multi-dimensional perception compound eye imaging, aiming to make the plant vitality intelligent monitoring system highly integrated and intelligent, while improving the accuracy, real-time and reliability of plant vitality intelligent monitoring data. The present invention can monitor at different stages during the growth of seeds or sprouts into seedlings.

[0042] The plant vitality intelligent monitoring system based on multi-dimensional perception compound eye imaging of the present invention comprises a light source module 1, an information acquisition module 2, an intelligent monitoring visualization platform 6 and an early warning module 7. Figure 1 The light source module 1 and the information acquisition module 2 are arranged in the incubator 4, the culture medium 5 is arranged at the lower part of the incubator 4, the plants are located in the culture medium 5, the culture medium 5 can be water-based or soil-based, and the incubator 5 can also monitor environmental parameters such as temperature, humidity, and light.

[0043] The light source module 1 includes a plurality of light sources, which are respectively located on the side wall and the top of the incubator 4. The plurality of light sources on the side wall are evenly arranged along the circumference, and the plurality of light sources on the top are evenly distributed along the circumference of the incubator and arranged toward the monitoring target 3, so that the light beam irradiated to the monitoring target 3 is uniform and without shadow. The light source is connected to a voltage-stabilized power supply to improve the quality of information obtained by the information acquisition module 2. The light source can provide light in the visible light band and the near-infrared band. The monitoring target 3 of the present invention can be a plant at different growth stages, including seeds, sprouts or seedlings. In other embodiments of the present invention, the light source can be an annular light source composed of multiple layers of annular LED lamp beads. The light emitted by the annular light source is irradiated on the surface of the monitoring target 3 at a certain angle, and the monitoring target 3 then emits the light into the information acquisition module 2 in the form of diffuse reflection. It can also be a coaxial light source, which is composed of an LED and a speckle splitter. A laser speckle meter can also be used.

[0044] The information acquisition module 2 is used to collect information of the monitoring target 3 at different growth moments in a large field of view, including a multi-dimensional perception compound eye imaging device arranged on the top of the incubator 4. The lens of the multi-dimensional perception compound eye imaging device includes multiple sub-eyes arranged in a hexagonal honeycomb structure on a hemispherical shell. The honeycomb structure arrangement and the hemispherical shell expand the detection range on the one hand, and a multi-view image of all monitoring targets 3 in the incubator 4 can be obtained through a single exposure. The cluster eyes in different areas of the same multi-dimensional perception compound eye imaging device ensure the monitoring accuracy and consistency of all monitoring targets 3 and avoid system errors. On the other hand, at least seven adjacent sub-eyes can capture the same monitoring target 3 from different perspectives at the same time. Figure 2 As shown, the multi-dimensional perception compound eye imaging device is used to collect multi-view images of the monitoring target 3 in the incubator 4 and transmit them to the intelligent monitoring visualization platform 6. In the present invention, except for the sub-eyes located at the edge, the remaining sub-eyes and the six adjacent sub-eyes form a cluster eye respectively, and each cluster eye is used to observe different monitoring targets 3 respectively. Filters or polarizers are respectively set on the seven sub-eyes in the same cluster eye. The present invention can be applied to monitoring the growth of the target 3 under strong light and dark light conditions by changing the optical materials and the matching photosensitive chips in the multi-dimensional perception compound eye imaging device. In this embodiment, the multi-dimensional information perception system based on the curved bionic compound eye disclosed in the Chinese patent with publication number CN116485676A is preferably used.

[0045] The intelligent monitoring visualization platform 6 includes a data display and processing module, a monitoring data information management module, and a vitality prediction module. The data display and processing module is connected to the light source module 1, the information collection module 2, and the early warning module 7, and is used to control its working state. At the same time, multi-dimensional information including spectral information, polarization information, multi-view information, and image information is obtained according to the multi-view image of the monitoring target 3, and the multi-dimensional information is solved to obtain the growth status information of the monitoring target 3, and it is displayed on the visualization interface. The monitoring data information management module is connected to the data display and processing module, and is used to record, save, statistically analyze and query the multi-view image, multi-dimensional information, and growth status information, and obtain the current growth curve of the monitoring target 3 through statistical analysis. The vitality prediction module trains the collected multi-dimensional information and its corresponding growth status information of the monitoring target 3 through the AI ​​algorithm to obtain the vitality prediction model and prediction curve of the monitoring target 3. In the subsequent cultivation monitoring process, only the multi-dimensional information of the monitoring target 3 in the pre-cultivation stage can be obtained, and it can be input into the vitality prediction module to obtain the growth status information prediction value, thereby shortening the cultivation cycle. The growth status information of monitoring target 3 includes growth curve, optimal growth time, seed germination rate, real-time height of sprouts and seedlings, shade avoidance of hypocotyls, growth rate and root growth.

[0046] During the working process, the data display and processing module processes the spectral information of different categories of monitoring targets 3, obtains the corresponding characteristic spectra, establishes a characteristic spectrum database, and stores it in the monitoring data information management module, and then selects the working wavelength of the multi-dimensional perception compound eye imaging device according to the characteristic spectra of different categories of monitoring targets 3, thereby controlling the operation of the multi-dimensional perception compound eye imaging device. Since the multi-dimensional perception compound eye imaging device has good interchangeability, the spectral information corresponding to different categories of monitoring targets 3 can be obtained by replacing the filter in front of its lens.

[0047] like Figure 2 As shown, multi-dimensional information is used to study the optical properties, material composition and vitality-related metabolism of the monitoring target 3, so as to quantitatively evaluate the growth status of the monitoring target 3, wherein both spectral information and polarization information are used to draw a growth curve, and the optimal growth moment of the monitoring target 3 is obtained according to the growth curve, and the seed germination rate is obtained by statistics; image information is used to combine with spectral information and polarization information to assist in obtaining the growth curve and seed germination rate, thereby improving the accuracy of the results; multi-view information is used to obtain the real-time height of sprouts and seedlings, the shade avoidance of hypocotyls and the growth of roots, and then the growth rate is obtained according to the real-time height statistics.

[0048] The image information is also used to obtain the size of the monitoring target 3, wherein the circumferential size of the seed can be measured, specifically: using an edge detection algorithm to identify the outer contour of the seed. The edge detection algorithm used in this embodiment is the Sobel operator, and the Canny algorithm or other detection algorithms can also be used. The actual circumferential size of the seed can be obtained by calculating the number of pixels occupied by the circumferential maximum value of the outer contour of the seed, combined with the focal length and working distance of the multi-dimensional perception compound eye imaging device, and the corresponding growth status information is obtained based on the auxiliary spectral information and polarization information of the circumferential size of the seed, and the data is saved in the monitoring data information management module. For young sprouts or seedlings, their edges can be identified according to the edge detection algorithm, thereby identifying their outer contours and obtaining the size of the young sprouts or seedlings in the horizontal direction.

[0049] The spectral information is used to evaluate the monitoring target 3 of soil-based culture, in which the working spectrum in the multi-dimensional sensing compound eye imaging device is easy to replace, and the working wavelength of the spectrum can be selected according to the different categories of the monitoring target 3. The spectral radiation curves of soil and water in the R, G, B and NIR bands are relatively flat, without obvious reflection peaks and absorption valleys. After the seeds break through the soil, the young shoots will have obvious changes in the above bands due to the effect of chlorophyll. In this embodiment, filters with four spectrum bands of 450 nm, 550 nm, 650 nm and 830 nm are selected for soil-based culture monitoring target 3 to achieve multispectral imaging. In other embodiments, filters with other wavelengths in the R, G, B and NIR bands can also be used to achieve this. The multi-dimensional sensing compound eye imaging device needs to be spectrally calibrated before use. The real radiation intensity received by the device in the actual working scene is corrected by spectral calibration, so as to draw a multi-spectral radiation curve. The components of the monitoring target 3 or the chlorophyll content of the young shoots are obtained according to the multi-spectral radiation curve, and then the state of the monitoring target 3 is judged, and the seed germination rate is obtained by statistics. The components of the monitoring target 3 or the chlorophyll content of the young shoots are used, combined with the temperature, humidity, light or fertilizer of the incubator 4, and other factors to draw a growth curve of the monitoring target 3, and confirm the optimal growth time of the monitoring target 3. The growth curve can be a curve of chlorophyll content changing with factors such as temperature, humidity, light or fertilizer, or a curve of each component changing with factors such as temperature, humidity, light or fertilizer. In the process of counting the seed germination rate, drawing the growth curve and determining the optimal growth time, this embodiment also combines image information for assistance to improve the accuracy of the results. As for the monitoring target 3 of water-based culture, since the water surface has extremely strong specular reflection during the imaging process, in addition to the spectral information, the polarization information must be combined to obtain the above results. The polarization information is obtained by setting linear polarizers with angles of 0°, 60°, and 120°, and then the polarization information is used to calculate the Mueller matrix. The polarization information and the spectral information of the above four spectral bands are combined to draw the multi-spectral radiation curve of water-based culture. At the same time, by combining spectral information, polarization information, image information and multi-viewing angle information, the root growth of the monitoring target 3 of water-based culture can also be monitored. Specifically, the change of spectral information at different positions in the root tissue can obtain the active degree of root growth of the monitoring target 3, providing data support for the root growth of the monitoring target 3. The multi-viewing angle information is used to measure the total length of the roots, the total length of the main roots and the lateral roots of the monitoring target 3, understand the expansion ability and depth of the roots, and then judge the healthy degree of growth of the monitoring target 3.

[0050] Based on the requirements for the above-mentioned spectral information and polarization information, in this embodiment, the front ends of the seven sub-eyes in the same cluster of eyes are respectively provided with filters of the above-mentioned four spectral bands and linear polarizers with angles of 0°, 60°, and 120°, so that the corresponding spectral information and polarization information can be obtained simultaneously through multi-view images.

[0051] In other embodiments of the present invention, the multiple sub-eyes in the multi-dimensional perception compound eye imaging device can also be arranged in the form of other regular polygons, each sub-eye and its surrounding adjacent sub-eyes are a cluster eye, and different wavelength filters or polarizers are respectively arranged on the sub-eyes in the same cluster eye to obtain corresponding spectral information and polarization information.

[0052] Multi-perspective information can also be used to calculate the real-time height of sprouts or seedlings, evaluate the longitudinal growth of sprouts or seedlings to monitor their growth rate, and at the same time combine the incubator 4 environment to obtain the hypocotyl shade avoidance situation. The present invention does not need to move or rotate the multi-dimensional perception compound eye imaging device, and can directly observe the complete root growth and hypocotyl shade avoidance situation. The use of multi-perspective images can solve the problem of inaccurate image processing caused by seedling shadows in a lighting environment, thereby improving monitoring accuracy.

[0053] The early warning module 7 performs early warning reminders according to the early warning strategy. In this embodiment, the early warning strategy includes reminders of the vitality monitoring time of the monitoring target 3, reminders of the breeding and updating time, etc., and reminders for monitoring at key nodes of the growth of the monitoring target 3. The early warning module 7 is connected to the data display and processing module, and reminds according to the growth status information obtained by the data display and processing module, so as to provide guarantee for the safe cultivation of the monitoring target 3.

[0054] The present invention provides a plant vitality intelligent monitoring method based on multi-dimensional perception compound eye imaging, such as Figure 3 As shown, the following steps are included:

[0055] S1. At different moments of the growth of the monitoring target 3, the multi-dimensional perception compound eye imaging device is controlled to shoot the monitoring target 3 to obtain multi-view images at different moments; wherein, the early warning module 7 can be used to remind shooting.

[0056] S2. The data display and processing module obtains corresponding multi-dimensional information based on the multi-view images at different times and stores it in the monitoring data information management module.

[0057] S3. The data display and processing module obtains the growth status information of the monitoring target 3 according to the multi-dimensional information at different times, specifically:

[0058] S3.1. The data display and processing module identifies the edge of the monitoring target 3 according to the image information at different times and calculates the corresponding size.

[0059] S3.2. The data display and processing module processes the spectral information at different times (corresponding to the monitoring target 3 of soil-based culture), or processes the spectral information and polarization information at different times (corresponding to the monitoring target 3 of water-based culture), draws the corresponding multi-spectral radiation curve, obtains the composition and / or chlorophyll content of the monitoring target 3 at different times according to the multi-spectral radiation curve, draws the growth curve in combination with the image information, and then obtains the optimal growth time; counts the composition and / or chlorophyll content of all monitoring targets 3 within the field of view, and obtains the seed germination rate in combination with the image information.

[0060] S3.3. The data display and processing module obtains the real-time height and hypocotyl shade avoidance of the young shoots or seedlings in soil-based and water-based cultures based on the multi-perspective information, and then obtains their growth rate based on the real-time height; the data display and processing module obtains the root growth status of the monitoring target 3 in water-based culture based on the multi-perspective information, spectral information, image information and polarization information.

[0061] Embodiment 2

[0062] The difference between this embodiment and the first embodiment is only in step S3 of the plant vitality intelligent monitoring method based on multi-dimensional perception compound eye imaging. Step S3 of this embodiment further includes:

[0063] S3.4. Input the multi-dimensional information into the vitality prediction module, and obtain the growth status information of the monitoring target 3 through the vitality prediction model and prediction curve therein.

Claims

1. A plant vitality intelligent monitoring method based on multi-dimensional perception compound eye imaging, characterized in that: A plant vitality intelligent monitoring system is used, the plant vitality intelligent monitoring system comprising an information acquisition module (2), a light source module (1) and an intelligent monitoring visualization platform (6), the information acquisition module (2) being configured as a multi-dimensional perception compound eye imaging device for acquiring multi-viewing angle images of a monitoring target (3); The lens of the multi-dimensional perception compound eye imaging device includes a plurality of sub-eyes arranged on a hemispherical shell; It also includes a light source module (1) and an intelligent monitoring visualization platform (6); The light source module (1) and the multi-dimensional sensing compound eye imaging device are arranged in a culture box (4) for culturing the monitoring target (3), and the light source module (1) is used to evenly illuminate the monitoring target (3) in the culture box (4); Definition: Each sub-eye and its surrounding adjacent sub-eyes are a cluster eye, and the cluster eye is used to obtain multi-view images of the same monitoring target (3), and filters or polarizers are respectively set on the sub-eyes in the same cluster eye; The intelligent monitoring visualization platform (6) comprises a data display and processing module and a monitoring data information management module; the data display and processing module is connected to the light source module (1) and the information acquisition module (2), controls the light source module (1) and the information acquisition module (2) to work, obtains corresponding multi-dimensional information according to the multi-view image, solves the multi-dimensional information to obtain growth status information of the corresponding monitoring target (3), and displays it on a visualization interface, wherein the multi-dimensional information comprises spectral information, polarization information, image information and multi-view information; the monitoring data information management module is connected to the data display and processing module, and is used for recording, saving, statistically analyzing and querying the multi-view image, the multi-dimensional information and the growth status information; The following steps are involved: Step 1, at different moments of growth of the monitoring target (3), controlling the multi-dimensional perception compound eye imaging device to photograph the monitoring target (3) to obtain corresponding multi-view images; Step 2, the data display and processing module obtains corresponding multi-dimensional information according to the multi-view images at different times, and stores the multi-dimensional information in the monitoring data information management module; Step 3, the data display and processing module obtains the growth status information of the monitoring target (3) according to the multi-dimensional information at different times, specifically: 3.

1. The data display and processing module identifies the edge of the monitoring target (3) according to the image information at different times and calculates the corresponding size; 3.

2. The data display and processing module processes the spectral information at different times, or processes the spectral information and polarization information at different times, draws the corresponding multi-spectral radiation curve, obtains the component content of the monitoring target (3) at different times according to the multi-spectral radiation curve, and draws the growth curve in combination with the image information to obtain the optimal growth time; the component content of all monitoring targets (3) within the field of view is counted, and the seed germination rate of the monitoring target (3) is obtained in combination with the image information; 3.

3. The data display and processing module obtains the real-time height and hypocotyl shade avoidance of the monitoring target (3) based on the multi-view information, and then obtains its growth rate based on the real-time height; the data display and processing module obtains the root growth of the monitoring target (3) in water-based culture based on the multi-view information, spectral information, image information and polarization information; in, Step 3.1 to step 3.3 are performed simultaneously or sequentially.

2. According to claim 1, the method for intelligent monitoring of plant vitality based on multi-dimensional perception compound eye imaging is characterized in that: a plurality of sub-eyes are arranged on the hemispherical shell in a hexagonal honeycomb structure, and each sub-eye is respectively arranged at six vertices and a center position of the hexagonal honeycomb structure; it is defined that the seven sub-eyes in each hexagonal honeycomb structure are a cluster eye, and the cluster eye is used to obtain multi-view images of the same monitoring target (3), and filters or polarizers are respectively arranged on the seven sub-eyes in the same cluster eye; The intelligent monitoring visualization platform (6) also includes a vitality prediction module, which trains the collected multi-dimensional information and the growth status information of the corresponding monitoring target (3) to obtain a vitality prediction model and a prediction curve of the monitoring target (3), and in subsequent cultivation, obtains a growth status information prediction value based on the multi-dimensional information obtained in the early cultivation of the monitoring target (3).

3. According to claim 2, the method for intelligent monitoring of plant vitality based on multi-dimensional perception compound eye imaging is characterized in that step 3 also includes: 3.

4. The vitality prediction module obtains the growth status information of the monitoring target (3) based on the multi-dimensional information through the vitality prediction model and prediction curve therein.

4. The method for intelligent monitoring of plant vitality based on multi-dimensional perception compound eye imaging according to claim 3 is characterized in that: In step 1, the working wavelength selection method of the multi-dimensional perception compound eye imaging device is specifically as follows: The data display and processing module processes the spectral information of different types of monitoring targets (3) to obtain corresponding characteristic spectra, establishes a characteristic spectrum database, stores it in the monitoring data information management module, and selects the working wavelength of the multi-dimensional perception compound eye imaging device according to the characteristic spectra of different types of monitoring targets (3).

5. The method for intelligent monitoring of plant vitality based on multi-dimensional perception compound eye imaging according to any one of claims 1 to 4, characterized in that: It also comprises an early warning module (7), which is connected to the data display and processing module and reminds the monitoring time and / or the cultivation operation time according to the growth status information obtained by the data display and processing module.

6. The method for intelligent monitoring of plant vitality based on multi-dimensional perception compound eye imaging according to claim 5 is characterized in that: The working wavelength of the multi-dimensional perception compound eye imaging device is in the visible light band and the near infrared band; The light source module (1) comprises a plurality of light sources, which are arranged as annular light sources or coaxial light sources and are used to provide light in the visible light band and the near infrared band.

7. The method for intelligent monitoring of plant vitality based on multi-dimensional perception compound eye imaging according to claim 6 is characterized in that: The seven sub-eyes in the same cluster eye are respectively provided with filters with wavelengths at R, G, B and NIR, and linear polarizers with angles of 0°, 60° and 120°.

8. The method for intelligent monitoring of plant vitality based on multi-dimensional perception compound eye imaging according to claim 2 is characterized in that: The vitality prediction module is trained by an AI algorithm on the collected multi-dimensional information and the corresponding growth status information of the monitoring target (3).

Images