Sorghum ear biomass real-time simulation method, device and equipment and storage medium

Abstract

The invention relates to a sorghum spike biomass real-time simulation method, device and equipment and a storage medium. According to the main technical scheme, the method comprises the steps that meteorological information, field management information, a multispectral image and sorghum biomass information in the whole growth cycle of sorghum growth are obtained, the total biological analog quantity of the overground part of sorghum is determined based on a pre-trained WOFOST crop model according to the meteorological information and the field management information, and the total biological analog quantity of the overground part of sorghum is calculated according to the multispectral image and the sorghum biomass information. According to the method, a multi-variety sorghum biomass ratio and red-edge chlorophyll response index estimation model is constructed, the spike biomass of sorghum is simulated in real time according to the overground part total biological analog quantity and the multi-variety sorghum biomass ratio and red-edge chlorophyll response index estimation model, and the spike biomass estimation precision and timeliness can be improved; meanwhile, spectral information of a large-area sorghum planting area can be rapidly obtained by obtaining the multi-spectral image of sorghum, and accurate management of a large-scale breeding field and a sorghum production area is facilitated.

Description

Technical Field

[0001] This application relates to the field of agricultural remote sensing technology, and in particular to a method, apparatus, equipment and storage medium for real-time simulation of sorghum ear biomass. Background Technology

[0002] Sorghum, especially high-quality glutinous sorghum, is a core raw material in the brewing industry due to its high starch content and moderate tannin ratio, making it an irreplaceable ingredient for brewing high-end baijiu (Chinese liquor). Modern research shows that during the reproductive growth stage of sorghum, from heading to maturity, the rate of spike biomass accumulation is significantly positively correlated with grain yield, making it a crucial factor in determining yield. Therefore, accurately obtaining sorghum spike biomass is of great significance for predicting the final sorghum yield.

[0003] Traditional methods for obtaining sorghum ear biomass rely on field measurements. While highly accurate, these methods are time-consuming, labor-intensive, and resource-intensive, making it difficult to observe large areas of fields. Furthermore, traditional methods require destructive observation, and the results are often delayed, resulting in high costs and long timeframes. Summary of the Invention

[0004] Based on this, this application provides a method, apparatus, equipment and storage medium for real-time simulation of sorghum ear biomass, in order to solve the problem that simulating sorghum ear biomass requires a lot of time, manpower and material resources, and the results obtained are lagging and difficult to predict large-scale fields.

[0005] Firstly, a method for real-time simulation of sorghum ear biomass is provided, the method comprising:

[0006] Acquire meteorological information, field management information, multispectral images, and sorghum biomass information throughout the entire growth cycle of sorghum;

[0007] Based on meteorological and field management information, the total biosimulation of the aboveground parts of sorghum was determined using a pre-trained WOFOST crop model.

[0008] Based on multispectral images and sorghum biomass information, an estimation model for the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index was constructed.

[0009] Based on the estimation model of total aboveground biomass, the ratio of biomass of various sorghum varieties, and the red-edged chlorophyll response index, the panicle biomass of sorghum is simulated in real time.

[0010] According to one achievable method in the embodiments of this application, sorghum biomass information includes measured panicle biomass and measured total aboveground biomass; based on multispectral images and sorghum biomass information, an estimation model for the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index is constructed, including:

[0011] Based on multispectral images, obtain the canopy multispectral reflectance of sorghum.

[0012] The red-edge chlorophyll response index of sorghum was determined based on the canopy multispectral reflectance.

[0013] Based on the red-edge chlorophyll response index of sorghum, measured panicle biomass, and measured total aboveground biomass, a model for estimating the biomass ratio and red-edge chlorophyll response index of multiple sorghum varieties was constructed.

[0014] According to one feasible method in the embodiments of this application, a model for estimating the biomass ratio and red-edge chlorophyll response index of multiple sorghum varieties is constructed based on the red-edge chlorophyll response index of sorghum, measured panicle biomass, and measured total aboveground biomass, including:

[0015] Determine the biomass ratio of the measured ear biomass to the measured total aboveground biomass;

[0016] Based on the red-edge chlorophyll response index and biomass ratio of sorghum, an estimation model for the biomass ratio and red-edge chlorophyll response index of multiple sorghum varieties was constructed.

[0017] According to one achievable method in an embodiment of this application, the canopy multispectral reflectance of sorghum includes red-edge canopy multispectral reflectance and near-infrared canopy spectral reflectance; based on the canopy multispectral reflectance of sorghum, the red-edge chlorophyll response index of sorghum is determined, including:

[0018] Based on the multispectral reflectance of the red-edge canopy and the near-infrared canopy spectral reflectance, the red-edge chlorophyll response index of sorghum is determined using the following formula:

[0019] CI red edge =ρ NIR / ρ red edge -1

[0020] Among them, CI red edge ρ represents the red-edged chlorophyll response index of sorghum. red edge ρ represents the multispectral reflectance of the red-edged canopy. NIR This indicates the spectral reflectance of the canopy in the near-infrared band.

[0021] According to one feasible method in the embodiments of this application, the estimation model of the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index is expressed as follows:

[0022]

[0023] in, This represents the ratio of measured spike biomass to measured total aboveground biomass.

[0024] a represents the slope of the estimation model of the biomass ratio of multiple sorghum varieties versus the red-edge chlorophyll response index, b represents the intercept of the estimation model of the biomass ratio of multiple sorghum varieties versus the red-edge chlorophyll response index, CI red edge This indicates the red-edged chlorophyll response index of sorghum.

[0025] According to one feasible method in the embodiments of this application, based on the estimation model of total aboveground biomass, the ratio of biomass of multiple sorghum varieties, and the red-edge chlorophyll response index, the panicle biomass of sorghum is simulated in real time, including:

[0026] Substitute the total aboveground biomass into the estimation model of biomass ratio of multiple sorghum varieties and red-edge chlorophyll response index to construct a sorghum ear biomass simulation model.

[0027] Based on the real-time determined total biomass of the aboveground parts and the red-edge chlorophyll response index, the sorghum spike biomass was simulated using a sorghum spike biosimulation model.

[0028] According to one feasible method in the embodiments of this application, the sorghum ear biomass simulation model is expressed as follows:

[0029] PB Modified =(a×CI) red edge +b)×AGB WOFOST

[0030] Among them, PB Mdified denoted as sorghum panicle biomass after correction, 'a' represents the slope of the multi-variety sorghum biomass ratio estimation model relative to the red-edge chlorophyll response index, and 'b' represents the intercept of the multi-variety sorghum biomass ratio estimation model relative to the red-edge chlorophyll response index. CI red edge The red-edge chlorophyll response index of sorghum, AGB WOFOST This represents the total biomass of the aboveground portion.

[0031] Secondly, a real-time sorghum ear biomass simulation device is provided, the device comprising:

[0032] The acquisition module is used to acquire meteorological information, field management information, multispectral images, and sorghum biomass information throughout the entire growth cycle of sorghum.

[0033] The determination module is used to determine the total biosimulation of the aboveground parts of sorghum based on meteorological and field management information and a pre-trained WOFOST crop model.

[0034] The module is used to construct estimation models for the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index based on multispectral images and sorghum biomass information.

[0035] The simulation module is used to estimate the panicle biomass of sorghum based on the total biomass of the aboveground parts, the biomass ratio of various sorghum varieties, and the red-edged chlorophyll response index, and to simulate the panicle biomass in real time.

[0036] Thirdly, a computer device is provided, comprising:

[0037] At least one processor; and

[0038] A memory communicatively connected to the at least one processor; wherein,

[0039] The memory stores computer instructions that can be executed by the at least one processor to enable the at least one processor to perform the method involved in the first aspect above.

[0040] Fourthly, a computer-readable storage medium is provided, having stored thereon computer instructions, wherein the computer instructions are used to cause a computer to perform the methods involved in the first aspect above.

[0041] According to the technical content provided in the embodiments of this application, meteorological information, field management information, multispectral images, and sorghum biomass information are obtained throughout the entire growth cycle of sorghum. Based on the meteorological and field management information, the total biomass of the aboveground parts of sorghum is determined using a pre-trained WOFOST crop model. Based on the multispectral images and sorghum biomass information, an estimation model of the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index is constructed. Based on the total biomass of the aboveground parts and the estimation model of the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index, the panicle biomass of sorghum is simulated in real time. By utilizing the significant correlation between the red-edge chlorophyll response index, the total biomass of the aboveground parts, and the panicle biomass, the accuracy and timeliness of panicle biomass estimation are improved. At the same time, by acquiring multispectral images of sorghum, spectral information of large-area sorghum planting areas can be quickly obtained, avoiding the high cost and lag of traditional manual sampling methods. This facilitates the precise management of large-scale breeding fields and sorghum production areas. Combining the WOFOST crop model and multispectral images enables rapid and efficient simulation of sorghum panicle biomass. Attached Figure Description

[0042] Figure 1 This is a diagram illustrating the application environment of a real-time sorghum ear biomass simulation method in one embodiment.

[0043] Figure 2 This is a flowchart illustrating a real-time sorghum ear biomass simulation method in one embodiment;

[0044] Figure 3 This is a structural block diagram of a real-time sorghum ear biomass simulation device in one embodiment;

[0045] Figure 4 This is a schematic structural diagram of a computer device in one embodiment. Detailed Implementation

[0046] The present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the scope of the present application.

[0047] For ease of understanding, the system to which this application applies will first be described. The real-time sorghum ear biomass simulation method provided in this application can be applied to, for example... Figure 1 In the system architecture shown, terminal 110 communicates with server 120 via a network. This method can be applied to either terminal 110 or server 120. Taking server 120 as an example, server 120 acquires meteorological information, field management information, multispectral images, and sorghum biomass information throughout the entire growth cycle of sorghum. Based on the meteorological and field management information, it determines the total biomass of the aboveground parts of sorghum using a pre-trained WOFOST crop model. Based on the multispectral images and sorghum biomass information, it constructs an estimation model for the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index. Based on the total biomass of the aboveground parts and the estimation model for the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index, it simulates the panicle biomass of sorghum in real time. Terminal 110 can be, but is not limited to, various personal computers, laptops, smartphones, and tablets. Server 120 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0048] The real-time sorghum panicle biomass simulation method is applicable to sorghum fields of various sizes. For example, a sorghum germplasm resource experimental field can include six sorghum varieties: Hongyingzi 1619, Hongzhenzhu 989, Qingnuo 319, Heinuo 229, Hongyingzi 519, and Tainuo 9. Sorghum was sown on April 5, 2024, and harvested on August 18, 2024. Sowing was done with three replicates, totaling 18 plots. The planting method was manual hill sowing, with 10 rows per plot, a row spacing of 0.6m, a row length of 7.2m, and a hill spacing of 0.3m. The plot area was 7.2m × 5.4m. To ensure no gaps in seedlings after sowing, the seed rate was 1.5 to 2 kg per mu (approximately 0.067 hectares). When the seedlings reached 3 to 4 leaves, thinning, fixing, and replanting were performed, leaving two uniformly sized, robust sorghum seedlings per hill. At each stage of sorghum development, a drone was deployed to acquire multispectral images of all sorghum fields, resulting in a total of 12 drone image data sets.

[0049] Figure 2 A flowchart illustrating a real-time sorghum ear biomass simulation method provided in this application embodiment is shown. This method can be implemented by, for example... Figure 1The system shown executes this command on terminal 110 or server 120. For example... Figure 2 As shown, the method may include the following steps:

[0050] S210 acquires meteorological information, field management information, multispectral images, and sorghum biomass information throughout the entire growth cycle of sorghum.

[0051] The meteorological information includes daily minimum and maximum temperatures, morning vapor pressure, cumulative radiation, wind speed, and precipitation. Field management information includes the timing of activities such as sowing and harvesting sorghum, including the sowing and harvesting dates.

[0052] Meteorological information is collected by setting up weather stations within the sorghum planting area. The weather stations can be HQZDZ-2 type automatic weather stations, erected at the center of the planting area, 1.5m above the ground, with no obstructions. The weather stations are equipped with a multi-sensor integrated system to detect daily minimum and maximum temperatures, morning vapor pressure, cumulative radiation, wind speed, and precipitation. Field management information for sorghum is recorded manually on the ground. Both meteorological and field management information are used as input to the WOFOST crop model.

[0053] Meteorological information is converted according to the standard input format of the WOFOST crop model and stored as a weather file. The weather file contains information on station names, year and day sequence, and meteorological variables, and is stored in the folder

\METEO\CABOWE

[0054] The sorghum sowing date and harvest date from the field management information are set as fixed sowing date and fixed end date parameters in the WOFOST crop model, respectively, as inputs to the time module of the WOFOST crop model. The sowing date is used to initialize the model run and determine the starting point of the sorghum growth period; the harvest date is used to determine the end point of the simulation cycle, thus ensuring the synchronization of the WOFOST crop model output with the actual sorghum growth period.

[0055] Multispectral sensors, mounted on a drone platform, acquire multispectral images of sorghum throughout its entire growth cycle at an altitude of 50–120 meters above the sorghum planting area. Data acquisition is conducted under clear, partly cloudy weather conditions, with flight missions carried out daily between 10:00 and 14:00. A total of 12 flight operations are performed throughout the entire sorghum growth cycle, covering key growth stages such as seedling, jointing, heading, grain-filling, and maturity, obtaining multispectral images of the sorghum throughout its entire growth period.

[0056] Sorghum biomass information includes measured panicle biomass and measured total aboveground biomass. In the sorghum breeding field, n1 sorghum plants were sampled per square meter. Destructive sampling was used to obtain the measured values ​​of total aboveground biomass and panicle biomass, where n1 is a set threshold. Specifically, while acquiring multispectral images from UAVs, representative sample plants of various sorghum varieties were simultaneously and destructively sampled from the ground to determine the true values ​​of total aboveground biomass and panicle biomass. Three sorghum sample plants were selected from each plot. The stems, leaves, and panicles were manually separated, bagged, labeled, and placed in an oven at 105℃ for 30 minutes to kill the enzymes, then dried at 80℃ to constant weight. The dry weight of the three sorghum plants was recorded using an electronic balance, and the average dry weight was calculated and converted to mass per unit area (g / m²) based on the sorghum planting density. 2 This represents the biomass of the plot. A total of 8 measurements of the aboveground biomass and 5 measurements of the panicle biomass were taken throughout the sorghum's growth process.

[0057] S220, based on meteorological and field management information, determines the total biosimulation of the aboveground parts of sorghum using a pre-trained WOFOST crop model.

[0058] The World Food Studies (WOFOST) model is a dynamic crop growth simulation model based on physiological and ecological processes. It can take meteorological data (temperature, radiation, precipitation), soil parameters (texture, water holding capacity), and management practices (sowing date, irrigation amount) as input to simulate crop biomass accumulation and final yield. Meteorological and field management information for sorghum is used as input to the WOFOST crop model, while the total aboveground biomass and panicle biomass of sorghum are used as output to train the model.

[0059] The pre-trained WOFOST crop model was localized by selecting the sorghum crop type built into the WOFOST crop model's crop module and loading its default physiological parameters. The soil module used default parameter settings, including soil texture and moisture characteristics, suitable for general farmland conditions. The WOFOST crop model was then invoked in a Python or Fortran environment to perform simulations. The model operates on a daily time step, dynamically simulating the entire growth process of sorghum from sowing to harvest, and obtaining dynamic simulation results of daily aboveground biomass (AGB) and panicle biomass (PB) throughout the entire growth period.

[0060] S230, based on multispectral images and sorghum biomass information, constructs an estimation model for the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index.

[0061] Based on multispectral images, image preprocessing operations such as band registration, geometric processing, and radiometric calibration were performed sequentially to obtain the canopy multispectral reflectance of sorghum throughout its entire growth period. The red-edge chlorophyll response index of sorghum was calculated based on the canopy multispectral reflectance. Based on the red-edge chlorophyll response index, measured panicle biomass, and measured total aboveground biomass, an estimation model of the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index was constructed.

[0062] During the growth and development of sorghum, the color of its panicle changes, significantly affecting the color and reflectivity of the sorghum canopy. Specifically, in the early stages of sorghum development, before the panicle appears, the sorghum canopy is green, with the pigment primarily being chlorophyll from the leaves. In the early heading stage, the panicle is typically green, and the canopy is also green, with the pigment mainly consisting of the abundant chlorophyll found in the leaves and panicle. As the sorghum matures, the panicle color changes, gradually shifting from green to yellow, red, or brown. This color change is primarily due to the gradual decomposition of chlorophyll, allowing other pigments (such as carotenoids and anthocyanins) to become more visible. At this stage, as the panicle structure gradually enlarges, the canopy primarily displays the color of the panicle, with anthocyanins or other carotenoids as the main pigments, and a significant reduction in chlorophyll content.

[0063] The Red Edge Chlorophyll Response Index (CIrededge) for sorghum is an index used in vegetation remote sensing analysis, particularly suitable for assessing chlorophyll content and photosynthetic activity. Utilizing spectral information in the red edge band, it can more sensitively reflect the health status and chlorophyll content of vegetation. Using the CIrededge index can effectively reflect chlorophyll changes during sorghum growth, thus reflecting the growth status of sorghum.

[0064] Given the unique characteristics of sorghum's pigment metabolism and canopy structure, the red-edge band (680–750 nm) is more sensitive to the physiological state of sorghum than traditional greenness and senescence indices, reflecting biomass dynamics earlier and more accurately. Furthermore, monitoring sorghum panicle biomass requires a greater emphasis on panicle-specific anisotropy compared to monitoring total aboveground biomass. During the grain-filling stage, the red-edge characteristic shift in the panicle due to anthocyanin and starch accumulation directly correlates with panicle dry matter accumulation. The red-edge chlorophyll response index is a specific response to physiological changes in the panicle. Multidimensional modeling combining the red-edge chlorophyll response index with biomass ratios not only effectively distinguishes between panicle and non-panicle signals but also fully utilizes the physiological mechanisms of panicle biomass to comprehensively reflect panicle development.

[0065] S240 uses a model that estimates the total biomass of the aboveground parts, the ratio of biomass of various sorghum varieties, and the red-edged chlorophyll response index to simulate the panicle biomass of sorghum in real time.

[0066] An improved sorghum panicle biomass simulation model was constructed by substituting the total aboveground biomass into the estimation model of the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index. Based on the real-time acquired total aboveground biomass and the red-edge chlorophyll response index determined in real time based on multispectral imagery, the panicle biomass of sorghum was simulated in real time using the sorghum panicle biomass simulation model.

[0067] As can be seen, the embodiments of this application acquire meteorological information, field management information, multispectral images, and sorghum biomass information throughout the entire growth cycle of sorghum. Based on the meteorological and field management information, the total biomass of the aboveground parts of sorghum is determined using a pre-trained WOFOST crop model. Based on the multispectral images and sorghum biomass information, an estimation model of the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index is constructed. Based on the total biomass of the aboveground parts and the estimation model of the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index, the panicle biomass of sorghum is simulated in real time. By utilizing the significant correlation between the red-edge chlorophyll response index, the total aboveground biomass, and the panicle biomass, the accuracy and timeliness of panicle biomass estimation are improved. At the same time, by acquiring multispectral images of sorghum, spectral information of large-area sorghum planting areas can be quickly obtained, avoiding the high cost and lag of traditional manual sampling methods. This facilitates the precise management of large-scale breeding fields and sorghum production areas. Combining the WOFOST crop model and multispectral images, the panicle biomass of sorghum can be simulated quickly and efficiently.

[0068] As a feasible approach, based on multispectral imagery and sorghum biomass information, a model for estimating the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index is constructed, including:

[0069] Based on multispectral images, obtain the canopy multispectral reflectance of sorghum.

[0070] The red-edge chlorophyll response index of sorghum was determined based on the canopy multispectral reflectance.

[0071] Based on the red-edge chlorophyll response index of sorghum, measured panicle biomass, and measured total aboveground biomass, a model for estimating the biomass ratio and red-edge chlorophyll response index of multiple sorghum varieties was constructed.

[0072] Multispectral images require preprocessing, including band registration, geometric correction, and radiometric calibration. Before aerial photography, the multispectral camera should be calibrated in a laboratory to obtain its objective lens distortion correction coefficient, and this parameter should be used for inter-band optical registration. After registration, the multiple band images will be in a unified coordinate system, ensuring that the same pixel appears in images of the same ground feature captured by different lenses. After objective lens distortion correction, the camera system's built-in PixelWrech2 software is used to perform registration based on similarity relationships, placing all band images in the same coordinate system.

[0073] Multispectral imagery yields image pixel DN (Digital Number) values, which need to be converted into surface reflectance or radiance. There is a linear relationship between DN values ​​and surface reflectance, which can be converted using the following formula:

[0074] R i =Gain i ×DN i +offset i (1)

[0075] Where i represents the band number of the multispectral camera, R i DN represents the radiance or reflectance at the entrance pupil of the i-th band. i Gain represents the DN value of the pixel corresponding to the i-th band. i With offset i These represent the gain coefficient and bias coefficient of the i-th band, respectively.

[0076] For example, six standard reflective mats with approximate Lambertian reflectances were placed at the edge of the test field, with reflectivities of 0.03, 0.12, 0.24, 0.36, 0.56, and 0.80, respectively. The ROI (Region of Interest) tool in Envi 5.3 was used to delineate the central region of the calibration mats, and the average DN values ​​of each calibration mat in 12 different bands were obtained, establishing the following formula:

[0077]

[0078] Among them, DN 0.03 DN 0.12 DN 0.24 DN 0.36 DN 0.56 and DN 0.80 These represent the average DN values ​​of the six calibration blankets on the multispectral image.

[0079] The gain coefficient Gain for each band is calculated using the least squares method. i and offset coefficienti Then, all DN values ​​in the multispectral image are converted into reflectance according to formula (1). After obtaining the radiometrically calibrated image, ROI is delineated in the central area of ​​each sorghum variety using Envi 5.3, and the average reflectance is calculated as the representative value of the canopy multispectral reflectance of that sorghum variety.

[0080] Based on the canopy multispectral reflectance of sorghum, the red-edge chlorophyll response index of sorghum was determined according to formula (3):

[0081] CI red edge =ρ NIR / ρ red edge -1 (3)

[0082] Among them, CI red edge ρ represents the chlorophyll response index of red-edged sorghum leaves. red edge and ρ NIR These are the canopy spectral reflectances in the red-edge and near-infrared bands, respectively, i.e., the canopy spectral reflectances in the 720nm and 800nm ​​bands of the multispectral sensor.

[0083] Based on the red-edge chlorophyll response index of sorghum, measured panicle biomass and measured total aboveground biomass, a model for estimating the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index was constructed. Specifically, the biomass ratio of measured panicle biomass to measured total aboveground biomass was determined.

[0084] Based on the red-edge chlorophyll response index and biomass ratio of sorghum, an estimation model for the biomass ratio and red-edge chlorophyll response index of multiple sorghum varieties was constructed.

[0085] The formula for calculating the biomass ratio of measured ear biomass to measured total aboveground biomass can be expressed as:

[0086]

[0087] in, The ratio of measured panicle biomass to measured aboveground total biomass is represented by PinacleBiomass, which represents measured panicle biomass, and Above Ground Biomass, which represents measured aboveground total biomass.

[0088] Based on the red-edge chlorophyll response index and biomass ratio of sorghum, and using the linear regression method in conjunction with formulas (3) and (4), the estimation model of biomass ratio and red-edge chlorophyll response index of multiple sorghum varieties is constructed as shown in formula (5):

[0089]

[0090] Where a represents the slope of the estimation model of the ratio of biomass of multiple sorghum varieties and the red-edged chlorophyll response index, and b represents the intercept.

[0091] As a feasible approach, based on the estimation model of total aboveground biomass and the ratio of biomass of various sorghum varieties with the red-edge chlorophyll response index, the panicle biomass of sorghum can be simulated in real time, including:

[0092] Substitute the total aboveground biomass into the estimation model of biomass ratio of multiple sorghum varieties and red-edge chlorophyll response index to construct a sorghum ear biomass simulation model.

[0093] Based on the real-time determined total biomass of the aboveground parts and the red-edge chlorophyll response index, the sorghum spike biomass was simulated using a sorghum spike biosimulation model.

[0094] Substituting the total biomass of the aboveground parts into formula (5), the sorghum ear biomass simulation model is obtained, expressed as the following formula:

[0095] PB Modified =(a×CI) red edge +b)×AGB WOFOST (6)

[0096] Among them, PB Modified AGB represents the corrected ear biomass. WOFOST This represents the total biosimulation of the aboveground parts obtained from the WOFOST crop model.

[0097] To address the issue of poor panicle biomass simulation results in the WOFOST crop model, this study integrates remote sensing data. Multispectral images of the sorghum canopy are acquired using a UAV equipped with a multispectral sensor. Canopy reflectance is extracted, and the red-edge chlorophyll response index of sorghum is calculated. These images are then compared with measured panicle biomass and measured total aboveground biomass to construct a multi-variety sorghum biomass ratio and red-edge chlorophyll response index estimation model. This model corrects the WOFOST-simulated sorghum panicle biomass, resulting in a significantly improved accuracy compared to the WOFOST crop model. This enables rapid, accurate, and real-time simulation of sorghum panicle biomass.

[0098] After obtaining the sorghum panicle biomass simulation model, it was applied to sorghum breeding fields. Real-time meteorological information, field management information, multispectral imagery, and sorghum biomass information were acquired throughout the entire sorghum growth cycle. The meteorological and field management information were input into a pre-trained WOFOST crop model to obtain the total aboveground biomass of sorghum. The canopy multispectral reflectance of sorghum was obtained from the multispectral imagery, and the red-edge chlorophyll response index of sorghum was calculated based on the canopy multispectral reflectance. The real-time determined total aboveground biomass and red-edge chlorophyll response index were then substituted into the sorghum panicle biomass simulation model to simulate the panicle biomass of sorghum.

[0099] There is a strong linear correlation between the red-edged chlorophyll response index and the ratio of sorghum spike biomass to total aboveground biomass, with a correlation coefficient R0. 2 The value reached above 0.86. Taking the aforementioned sorghum germplasm resource experimental field as an example, 2 / 3 of the 18 fields (i.e., 12 fields) were used as the training set, and 1 / 3 of the fields were used as the validation set. The biomass ratio of the artificially obtained measured panicle biomass to the measured total aboveground biomass and the red-edge chlorophyll response index of these fields were linearly fitted, and the estimation model of the biomass ratio and red-edge chlorophyll response index of multiple sorghum varieties was obtained as follows:

[0100]

[0101] Substituting the total aboveground biomass simulated by the WOFOST crop model into formula (7), we obtain the improved sorghum spike biomass simulation model, expressed as the following formula:

[0102] PB Modified =(-0.3525×CI) red edge +0.9038)×AGB WOFOST (8)

[0103] The improved sorghum panicle biosimulation model was applied to an experimental field of sorghum germplasm resources, with one-third of the field used as the validation set to verify the error between the results of the improved sorghum panicle biosimulation model and the measured panicle biomass. The correlation coefficient R between the estimated panicle biomass value and the measured panicle biomass value was obtained. 2 The value is 0.90, and the root mean square error (RMSE) is 198.23 g / m². 2 The relative root mean square error (rRMSE) is 0.35, indicating good model accuracy and quality.

[0104] It should be understood that, although Figure 2 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated in this application, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Furthermore, Figure 2 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0105] Figure 3 This is a schematic diagram of a real-time sorghum ear biomass simulation device provided in an embodiment of this application. The device can be installed in... Figure 1 The terminal 110 or server 120 in the system shown is used to perform, for example... Figure 2 The method flow is shown below. Figure 3 As shown, the device may include: an acquisition module 310, a determination module 320, a construction module 330, and a simulation module 340. The main functions of each component module are as follows:

[0106] The acquisition module 310 is used to acquire meteorological information, field management information, multispectral images and sorghum biomass information throughout the entire growth cycle of sorghum.

[0107] The determination module 320 is used to determine the total biosimulation of the aboveground parts of sorghum based on meteorological information and field management information and a pre-trained WOFOST crop model.

[0108] Module 330 is used to construct an estimation model of the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index based on multispectral images and sorghum biomass information.

[0109] The simulation module 340 is used to simulate the panicle biomass of sorghum in real time based on the total biomass of the aboveground parts, the ratio of biomass of multiple sorghum varieties, and the red-edged chlorophyll response index estimation model.

[0110] As an achievable method, sorghum biomass information includes measured panicle biomass and measured total aboveground biomass; module 330 is specifically used to: obtain the canopy multispectral reflectance of sorghum based on multispectral images; determine the red-edge chlorophyll response index of sorghum based on the canopy multispectral reflectance; and construct an estimation model for the biomass ratio and red-edge chlorophyll response index of multiple sorghum varieties based on the red-edge chlorophyll response index, measured panicle biomass, and measured total aboveground biomass.

[0111] As one feasible approach, module 330 is constructed specifically for: determining the biomass ratio of the measured panicle biomass to the measured total aboveground biomass; and constructing an estimation model for the biomass ratio and red-edge chlorophyll response index of multiple sorghum varieties based on the red-edge chlorophyll response index and biomass ratio of sorghum.

[0112] As one feasible approach, the canopy multispectral reflectance of sorghum includes red-edge canopy multispectral reflectance and near-infrared canopy spectral reflectance; the determination module 320 is specifically used to: determine the red-edge chlorophyll response index of sorghum based on the red-edge canopy multispectral reflectance and near-infrared canopy spectral reflectance using the following formula:

[0113] CI red edge =ρ NIR / ρ red edge -1

[0114] Among them, CI red edge ρ represents the red-edged chlorophyll response index of sorghum. red edge ρ represents the multispectral reflectance of the red-edged canopy. NIR This indicates the spectral reflectance of the canopy in the near-infrared band.

[0115] As an feasible approach, the estimation model for the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index is expressed as follows:

[0116]

[0117] in, This represents the ratio of measured spike biomass to measured total aboveground biomass.

[0118] a represents the slope of the estimation model of the biomass ratio of multiple sorghum varieties versus the red-edge chlorophyll response index, b represents the intercept of the estimation model of the biomass ratio of multiple sorghum varieties versus the red-edge chlorophyll response index, CI red edge This indicates the red-edged chlorophyll response index of sorghum.

[0119] As one feasible approach, the simulation module 340 is specifically used to: substitute the total biomass of the aboveground parts into the estimation model of the biomass ratio of multiple sorghum varieties and the red-edged chlorophyll response index to construct a sorghum ear biomass simulation model; and simulate the ear biomass of sorghum based on the real-time determined total biomass of the aboveground parts and the red-edged chlorophyll response index.

[0120] As one feasible approach, the sorghum ear biomass simulation model is represented as follows:

[0121] PB Modified =(a×CI) red edge +b)×AGB WOFOST

[0122] Among them, PB Modified denoted as sorghum panicle biomass after correction, 'a' represents the slope of the multi-variety sorghum biomass ratio estimation model relative to the red-edge chlorophyll response index, and 'b' represents the intercept of the multi-variety sorghum biomass ratio estimation model relative to the red-edge chlorophyll response index. CI red edge The red-edge chlorophyll response index of sorghum, AGB WOFOST This represents the total biomass of the aboveground portion.

[0123] The same or similar parts among the above embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments.

[0124] It should be noted that the embodiments of this application may involve the use of user data. In practical applications, user-specific personal data may be used in the scheme described herein within the scope permitted by applicable laws and regulations, provided that it complies with the applicable laws and regulations of the country (e.g., explicit consent from the user, actual notification to the user, explicit authorization from the user, etc.).

[0125] According to embodiments of this application, this application also provides a computer device and a computer-readable storage medium.

[0126] like Figure 4 The diagram shown is a block diagram of a computer device according to an embodiment of this application. The term "computer device" is intended to represent various forms of digital computers or mobile devices. The digital computer may include a desktop computer, a portable computer, a workbench, a personal digital assistant, a server, a mainframe computer, and other suitable computers. The mobile device may include a tablet computer, a smartphone, a wearable device, etc.

[0127] like Figure 4 As shown, the computer device 400 includes a computing unit 401, a ROM 402, a RAM 403, a bus 404, and an input / output (I / O) interface 405. The computing unit 401, ROM 402, and RAM 403 are interconnected via the bus 404. The input / output (I / O) interface 405 is also connected to the bus 404.

[0128] The computing unit 401 can execute various processes in the method embodiments of this application according to computer instructions stored in the read-only memory (ROM) 402 or computer instructions loaded from the storage unit 408 into the random access memory (RAM) 403. The computing unit 401 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. The computing unit 401 can include, but is not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. In some embodiments, the methods provided in the embodiments of this application can be implemented as computer software programs, which are tangibly contained in a computer-readable storage medium, such as the storage unit 408.

[0129] RAM 403 can also store various programs and data required for the operation of computer device 400. Part or all of the computer program can be loaded and / or installed on computer device 400 via ROM 402 and / or communication unit 409.

[0130] The input unit 406, output unit 407, storage unit 408, and communication unit 409 in the computer device 400 can be connected to the I / O interface 405. The input unit 406 can be, for example, a keyboard, mouse, touchscreen, or microphone; the output unit 407 can be, for example, a monitor, speaker, or indicator light. The computer device 400 can exchange information and data with other devices through the communication unit 409.

[0131] It should be noted that the device may also include other components necessary for normal operation. It may also include only the components necessary for implementing the solution of this application, without necessarily including all the components shown in the figures.

[0132] Various implementations of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SOCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof.

[0133] The computer instructions used to implement the methods of this application may be written in any combination of one or more programming languages. These computer instructions may be provided to the computing unit 401 such that, when executed by the computing unit 401, such as a processor, the computer instructions cause the execution of the steps involved in the embodiments of the methods of this application.

[0134] The computer-readable storage medium provided in this application can be a tangible medium that can contain or store computer instructions for performing the steps involved in the method embodiments of this application. The computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, and other forms of storage media.

[0135] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for real-time simulation of sorghum ear biomass, characterized in that, The method includes: Acquire meteorological information, field management information, multispectral images, and sorghum biomass information throughout the entire growth cycle of sorghum; Based on the meteorological information and the field management information, the total biosimulation of the aboveground parts of the sorghum is determined using a pre-trained WOFOST crop model; Based on the multispectral images and the sorghum biomass information, an estimation model for the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index was constructed. Based on the total biomass of the aboveground parts and the biomass ratio of the various sorghum varieties, the spike biomass of the sorghum is simulated in real time using the red-edged chlorophyll response index estimation model.

2. The method according to claim 1, characterized in that, The sorghum biomass information includes measured panicle biomass and measured total aboveground biomass; the step of constructing an estimation model for the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index based on the multispectral imagery and the sorghum biomass information includes: Based on the multispectral image, the canopy multispectral reflectance of the sorghum is obtained; The red-edge chlorophyll response index of the sorghum was determined based on the canopy multispectral reflectance. Based on the red-edge chlorophyll response index of the sorghum, the measured panicle biomass, and the measured total aboveground biomass, a model for estimating the biomass ratio and red-edge chlorophyll response index of multiple sorghum varieties was constructed.

3. The method according to claim 2, characterized in that, The method for constructing a multi-variety sorghum biomass ratio and red-edge chlorophyll response index estimation model based on the red-edge chlorophyll response index of the sorghum, the measured panicle biomass, and the measured total aboveground biomass includes: Determine the biomass ratio of the measured ear biomass to the measured total aboveground biomass; Based on the red-edge chlorophyll response index of the sorghum and the biomass ratio, an estimation model for the biomass ratio and red-edge chlorophyll response index of multiple sorghum varieties was constructed.

4. The method according to claim 2, characterized in that, The canopy multispectral reflectance of the sorghum includes red-edge canopy multispectral reflectance and near-infrared canopy spectral reflectance; the determination of the red-edge chlorophyll response index of the sorghum based on the canopy multispectral reflectance includes: Based on the red-edge canopy multispectral reflectance and the near-infrared canopy spectral reflectance, the red-edge chlorophyll response index of the sorghum is determined using the following formula: CI red edge =ρ NIR / r red edge -1 Among them, CI red edge The red-edge chlorophyll response index of the sorghum is represented by ρ. red edge ρ represents the multispectral reflectance of the red-edged canopy. NIR This indicates the spectral reflectance of the canopy in the near-infrared band.

5. The method according to claim 2, characterized in that, The estimation model for the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index is expressed as follows: in, This represents the biomass ratio of the measured ear biomass to the measured total aboveground biomass; a represents the slope of the estimation model of the biomass ratio of multiple sorghum varieties versus the red-edge chlorophyll response index, b represents the intercept of the estimation model of the biomass ratio of multiple sorghum varieties versus the red-edge chlorophyll response index, CI red edge The red-edged chlorophyll response index of the sorghum is indicated.

6. The method according to claim 1, characterized in that, The method for estimating the panicle biomass of sorghum based on the total biomass of the aboveground parts, the ratio of biomass of various sorghum varieties, and the red-edge chlorophyll response index includes: Substitute the total biomass of the aboveground parts into the estimation model of the biomass ratio of the multi-variety sorghum and the red-edge chlorophyll response index to construct a sorghum ear biomass simulation model. Based on the real-time determined total biomass of the aboveground parts and the red-edged chlorophyll response index, the sorghum panicle biomass is simulated using the sorghum panicle biosimulation model.

7. The method according to claim 6, characterized in that, The sorghum ear biomass simulation model is expressed as follows: PB Modified =(a×CI red edge +b)×AGB WOFOST Among them, PB Modified CI represents the corrected panicle biomass of the sorghum, a represents the slope of the multi-variety sorghum biomass ratio estimation model with respect to the red-edge chlorophyll response index, b represents the intercept of the multi-variety sorghum biomass ratio estimation model with respect to the red-edge chlorophyll response index, and CI red edge The red-edge chlorophyll response index (AGB) of the sorghum is represented. WOFOST This represents the total biomass of the aboveground portion.

8. A real-time sorghum ear biomass simulation device, characterized in that, The device includes: The acquisition module is used to acquire meteorological information, field management information, multispectral images, and sorghum biomass information throughout the entire growth cycle of sorghum. The determination module is used to determine the total biosimulation of the aboveground parts of the sorghum based on the meteorological information and the field management information, using a pre-trained WOFOST crop model. The construction module is used to construct an estimation model of the biomass ratio of multiple sorghum varieties and the red-edge chlorophyll response index based on the multispectral images and the sorghum biomass information. The simulation module is used to simulate the panicle biomass of sorghum in real time based on the total biomass of the aboveground parts, the ratio of biomass of the various sorghum varieties, and the red-edged chlorophyll response index estimation model.

9. A computer device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores computer instructions executable by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method of any one of claims 1-7.

10. A computer-readable storage medium storing computer instructions thereon, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1 to 7.

Images