Soybean growth monitoring methods and systems based on monitoring sensors

Abstract

This invention discloses a soybean growth monitoring method and system based on monitoring sensors, relating to the technical field of plant growth monitoring. The method includes: collecting multi-source growth-driven data using distributed sensors deployed in soybean planting areas; constructing a growth-driven pathway characterizing energy and material transfer from the rhizosphere to the stem-leaf-canopy by temporal correlation and path reconstruction according to the transfer directions of water, heat, and light energy; performing a three-dimensional continuity evaluation of pathway response delay, energy attenuation, and stability; and fusing the growth-driven continuity evaluation result with a second determination result based on monitoring images to output a fused monitoring result. This invention solves the technical problems of insufficient information correlation, lack of evaluation continuity, and low data fusion in existing technologies, which make it difficult to comprehensively and accurately reveal the soybean growth-driven state. It achieves precise determination of the soybean growth-driven state, thus improving the accuracy and comprehensiveness of soybean growth status monitoring.

Description

Technical Field

[0001] This invention relates to the field of plant growth monitoring technology, specifically to a soybean growth monitoring method and system based on monitoring sensors. Background Technology

[0002] Traditional soybean growth monitoring mainly relies on regular manual inspections and single-point environmental data collection, such as visually observing the morphological characteristics of plants or using single soil temperature and humidity sensors or air temperature and humidity sensors to assess the growth environment. However, soybean growth involves a close coupling of multiple links, including the soil rhizosphere environment, stem transport, and canopy photosynthesis. Water, nutrients, and energy are transferred and transformed in specific directions within the rhizosphere-stem-leaf-canopy, jointly determining the final growth state of soybeans. Current growth monitoring lacks a comprehensive consideration of the complete energy and material transport chain from rhizosphere supply to stem-leaf transport to canopy release, making it difficult to reveal changes in the intrinsic driving forces during growth. It also lacks temporal correlation and physical path reconstruction of multi-source data, making it difficult to establish effective causal relationships between monitoring data. For example, it is impossible to accurately determine whether delayed water supply in the rhizosphere or reduced light energy utilization efficiency in the canopy leads to the overall growth hindrance of the plant. Furthermore, it is difficult to quantify the dynamic performance of growth pathways, such as the response speed of material and energy transfer, the degree of energy decay during the process, and the long-term stability of the pathways. As a result, it is impossible to continuously evaluate the growth driving forces of soybeans, and the monitoring results are biased, making it difficult to comprehensively and accurately determine the true growth status of soybeans.

[0003] Therefore, current technologies suffer from insufficient information correlation, lack of evaluation continuity, and low data fusion, making it difficult to comprehensively and accurately reveal the driving state of soybean growth. Summary of the Invention

[0004] This application provides a soybean growth monitoring method and system based on monitoring sensors, which solves the technical problems in the prior art, such as insufficient information correlation, lack of evaluation continuity, and low data fusion, which make it difficult to comprehensively and accurately reveal the driving state of soybean growth. It achieves accurate determination of the driving state of soybean growth and improves the accuracy and comprehensiveness of soybean growth status monitoring.

[0005] This application provides a soybean growth monitoring method based on monitoring sensors. The method includes: deploying distributed monitoring sensors within a soybean planting area to collect multi-source growth-driven data characterizing rhizosphere supply status, canopy energy release status, and near-surface transport conditions; performing temporal correlation and path reconstruction on the multi-source growth-driven data according to the direction of water, heat, and light energy transfer during soybean growth to construct a growth-driven pathway characterizing the energy and material transfer relationship between the rhizosphere, stems and leaves, and the canopy; using the growth-driven pathway, performing an evaluation analysis of a three-dimensional growth-driven continuity evaluation channel to establish a growth-driven continuity evaluation result, wherein the three-dimensional growth-driven continuity evaluation channel includes a pathway response delay sub-channel, an energy attenuation degree sub-channel, and a pathway stability sub-channel; determining the current growth status of soybeans based on the growth-driven continuity evaluation result to establish a first determination result; acquiring monitoring images of soybeans and using the monitoring images to establish a second determination result; and performing monitoring fusion on the first determination result and the second determination result to output a fused monitoring result.

[0006] In a possible implementation, a growth-driven pathway characterizing the energy and mass transfer relationship between the rhizosphere, stem-leaf, and canopy is constructed, including: based on the multi-source growth-driven data, constructing sets of driving nodes to characterize the rhizosphere supply state, stem-leaf transport state, and canopy energy release state, wherein each driving node is characterized by the driving intensity features and changes within a corresponding time window; performing cross-layer temporal alignment processing on the rhizosphere driving nodes, stem-leaf driving nodes, and canopy driving nodes according to the direction of water, heat, and light energy transfer during soybean growth to determine the effective response time period between adjacent driving nodes; within the effective response time period, establishing directed driving association edges between the rhizosphere driving nodes and stem-leaf driving nodes, and between stem-leaf driving nodes and canopy driving nodes, based on the correlation of changes in driving intensity and the consistency of response time; and generating a growth-driven pathway characterizing the energy and mass transfer relationship between the rhizosphere, stem-leaf, and canopy based on the directed driving association edges.

[0007] In a possible implementation, the growth-driven path is used to perform an evaluation analysis of a three-dimensional growth-driven continuity evaluation channel, including: activating a path response delay sub-channel based on the drive response time difference between adjacent drive nodes on the growth-driven path; the path response delay sub-channel is used to analyze the temporal distribution characteristics of the drive response time difference and generate a response delay evaluation result characterizing the temporal consistency of drive transmission; activating an energy attenuation degree sub-channel based on the drive intensity transmission ratio between adjacent drive nodes on the growth-driven path; the activated energy attenuation degree sub-channel is used to analyze the drive intensity attenuation trend and cumulative attenuation characteristics based on the drive intensity transmission ratio and generate an energy attenuation evaluation result characterizing the integrity of drive transmission; activating a path stability sub-channel based on the degree of maintenance of drive correlation within a continuous time window on the growth-driven path; the path stability sub-channel analyzes the fluctuation amplitude and duration of drive correlation to generate a path stability evaluation result characterizing structural stability; and establishing a growth-driven continuity evaluation result based on the response delay evaluation result, energy attenuation evaluation result, and path stability evaluation result.

[0008] In a possible implementation, the current growth state of soybean is determined based on the growth-driven continuity evaluation result, and a first determination result is established. This includes: performing evaluation pattern recognition on the growth-driven continuity evaluation result, and extracting a combination of continuity features reflecting the temporal consistency, transmission integrity, and structural stability of the pathway; determining the continuity state type of the growth-driven pathway based on the combination of continuity features, wherein the continuity state type includes a continuous stable state, a continuous fluctuating state, and a continuous interrupted state; mapping the continuity state type to the corresponding soybean growth state category, and establishing a first discrimination result, wherein the soybean growth state category includes an effective growth state, a growth-restricted state, and a physiological stress state.

[0009] In a possible implementation, acquiring monitoring images of soybeans and establishing a second determination result using the monitoring images includes: performing crop region separation processing on the monitoring images to extract soybean plant image regions; using the soybean plant image regions to extract an image feature set characterizing the spatial morphological features and canopy structure features of the plants, the image feature set including plant coverage features, canopy density features, and morphological consistency features; establishing a description of the soybean population growth morphology based on the image feature set, and establishing the second determination result.

[0010] In a possible implementation, monitoring and fusing the first determination result and the second determination result, and outputting the fused monitoring result, includes: constructing a growth state consistency determination relationship based on the first determination result and the second determination result, wherein the growth state consistency determination relationship characterizes the degree of consistency between the two determination results in the growth state category; performing monitoring fusion using the growth state consistency determination relationship, and outputting the fused monitoring result.

[0011] In possible implementations, outputting the fused monitoring results also includes: performing anomaly analysis on soybean growth based on the fused monitoring results and configuring anomaly warning signals; and performing anomaly reporting management for growth monitoring based on the anomaly warning signals.

[0012] This application also provides a soybean growth monitoring system based on monitoring sensors. The system includes: a data acquisition module for deploying distributed monitoring sensors within a soybean planting area, the distributed sensors collecting multi-source growth-driven data characterizing rhizosphere supply status, canopy energy release status, and near-surface transport conditions; a growth-driven pathway construction module for performing temporal correlation and path reconstruction of the multi-source growth-driven data according to the direction of water, heat, and light energy transfer during soybean growth, constructing growth-driven pathways characterizing the relationship between rhizosphere-stem-leaf-canopy energy and mass transfer; and an evaluation result establishment module for utilizing the growth-driven... The system includes a dynamic pathway, which performs evaluation and analysis of a three-dimensional growth-driven continuity evaluation channel to establish growth-driven continuity evaluation results. The three-dimensional growth-driven continuity evaluation channel includes a pathway response delay sub-channel, an energy attenuation degree sub-channel, and a pathway stability sub-channel. A growth state determination module is used to determine the current growth state of soybeans based on the growth-driven continuity evaluation results and establish a first determination result. A second determination result establishment module is used to acquire monitoring images of soybeans and establish a second determination result using the monitoring images. A fusion monitoring result output module is used to perform monitoring fusion of the first determination result and the second determination result and output a fused monitoring result.

[0013] This application proposes a soybean growth monitoring method and system based on monitoring sensors. This method collects multi-source growth-driven data using distributed sensors deployed in soybean planting areas. It then performs temporal correlation and path reconstruction according to the transfer directions of water, heat, and light energy to construct growth-driven pathways characterizing energy and material transfer from the rhizosphere to the stem-leaf-canopy. The method performs a three-dimensional continuity evaluation of pathway response delay, energy decay, and stability. Finally, it fuses this growth-driven continuity evaluation result with a second determination result based on monitoring images to output a fused monitoring result. This addresses the technical problems in existing technologies, such as insufficient information correlation, lack of evaluation continuity, and low data fusion, which make it difficult to comprehensively and accurately reveal the soybean growth-driven state. The method achieves precise determination of the soybean growth-driven state, thus improving the accuracy and comprehensiveness of soybean growth status monitoring. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0015] Figure 1 This is a schematic diagram of the soybean growth monitoring method based on monitoring sensors provided in an embodiment of this application.

[0016] Figure 2 This is a schematic diagram of the structure of a soybean growth monitoring system based on monitoring sensors provided in an embodiment of this application.

[0017] Figure labeling: Data acquisition module 10, growth-driven pathway construction module 20, evaluation result establishment module 30, growth status determination module 40, second determination result establishment module 50, fusion monitoring result output module 60. Detailed Implementation

[0018] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description is provided in conjunction with the accompanying drawings and preferred embodiments, based on the specific implementation methods, structure, features, and effects of the present invention.

[0019] This application provides a soybean growth monitoring method based on monitoring sensors, such as... Figure 1 As shown, the method includes: Step S100: In the soybean planting area, distributed monitoring sensors are deployed to collect multi-source growth-driven data characterizing the rhizosphere supply status, canopy energy release status, and near-surface transport conditions.

[0020] Preferably, within the soybean planting area, multiple sensors are physically deployed at different locations and heights in space, forming a distributed monitoring sensor network, based on the different monitoring targets such as the rhizosphere, canopy, and near-soil layer. This network includes at least the following: probe-type sensors inserted into the soil, such as soil temperature and humidity sensors, soil conductivity sensors, soil pH sensors, and soil oxygen sensors; sensors attached to or non-contact to measure the plant's own condition, such as leaf surface temperature and humidity sensors, stem flow meters for monitoring the rate of water transport within the stem, and photosynthetically active radiation sensors for monitoring the light energy received by the canopy; and small meteorological sensors deployed above or inside the canopy, such as wind... The system includes a velocity sensor, a canopy thermometer, an air temperature and humidity sensor, a carbon dioxide concentration sensor, and a photosynthetically active radiation sensor. Distributed monitoring sensors are used to collect multi-source growth-driven data characterizing rhizosphere supply status, canopy energy release status, and near-surface transport conditions. Among these, the growth-driven data characterizing rhizosphere supply status reflects the ability of the soybean root microenvironment to provide water and mineral nutrients to the aboveground parts. Parameters may include soil volumetric water content, soil water potential, soil temperature, and soil electrical conductivity, which respectively reflect rhizosphere water supply capacity, water availability and ease of supply to plants, rhizosphere heat status, and rhizosphere soluble salt ion concentration.

[0021] Preferably, the growth-driven data characterizing the canopy energy release state reflects the process by which the soybean canopy converts absorbed light and heat energy into biochemical energy and releases energy and substances in the form of water vapor during photosynthesis and transpiration. Parameters may include canopy temperature, light and effective radiation, transpiration rate, and carbon dioxide flux, which respectively reflect the intensity of leaf transpiration heat dissipation, the intensity of solar light in a specific wavelength used for photosynthesis, the amount of water vapor lost per unit leaf area per unit time through transpiration, and the intensity of photosynthesis. The growth-driven data characterizing near-ground transport conditions reflects the aerodynamic environment above and inside the canopy, determining the efficiency of energy and substance exchange between the crop and the atmosphere. Parameters may include wind speed, air temperature and humidity, and turbulence intensity. Wind speed affects the thickness of the canopy boundary layer, which in turn affects transpiration and carbon dioxide exchange. Air temperature and humidity determine the saturated vapor pressure difference of the air, and turbulence intensity refers to the degree of air turbulence, affecting the vertical transport efficiency of heat and water vapor.

[0022] Step S200: According to the direction of water, heat and light energy transfer during soybean growth, the multi-source growth driving data is temporally correlated and reconstructed to build a growth driving pathway that characterizes the relationship between energy and material transfer in the rhizosphere-stem-leaf-canopy.

[0023] Step S200 further includes: based on the multi-source growth-driven data, constructing sets of driving nodes to characterize the rhizosphere supply state, stem-leaf transport state, and canopy energy release state, wherein each driving node is characterized by the driving intensity features and changes within the corresponding time window; performing cross-layer temporal alignment processing on the rhizosphere driving nodes, stem-leaf driving nodes, and canopy driving nodes according to the direction of water, heat, and light energy transfer during soybean growth to determine the effective response time period between adjacent driving nodes; within the effective response time period, establishing directed driving association edges between rhizosphere driving nodes and stem-leaf driving nodes, and between stem-leaf driving nodes and canopy driving nodes, based on the correlation of changes in driving intensity and the consistency of response time; and generating growth-driven pathways characterizing the energy and material transfer relationship between the rhizosphere, stem-leaf, and canopy based on the directed driving association edges.

[0024] Preferably, the continuous multi-source growth-driven data is discretized to construct sets of driving nodes to characterize the rhizosphere supply state, stem-leaf transport state, and canopy energy release state. Driving nodes refer to representative data slices or data state points on the time axis, i.e., data abstractions of soybean rhizosphere, stem-leaf, and canopy parts within a specific time period. Specifically, rhizosphere driving nodes are composed of rhizosphere sensor data, reflecting the ability of the roots to provide water and nutrients to the aboveground parts during that period; stem-leaf driving nodes are composed of data from stem flow meters, leaf temperature and humidity sensors, etc., reflecting the stem's... The rate and efficiency at which vascular bundles transport water and nutrients absorbed from the rhizosphere upwards to the leaves; the canopy driving nodes are composed of data such as canopy temperature, photosynthetically active radiation, and transpiration rate, reflecting the intensity at which leaves convert light energy into heat and chemical energy; each driving node is characterized by the driving intensity characteristics and changing driving forces within the corresponding time window. The driving intensity characteristics refer to the statistical quantities of data within the time window, such as average soil moisture content, peak transpiration rate, and cumulative photosynthetically active radiation, while the changing driving forces refer to the dynamic trends of data within the time window, such as the rate of decrease in soil water potential and the slope of increase in canopy temperature.

[0025] Preferably, according to the direction of water, heat and light energy transfer during soybean growth, cross-layer temporal alignment processing is performed on the rhizosphere driving nodes, stem-leaf driving nodes and canopy driving nodes. That is, a causal matching mechanism is established to solve the time response delay of sensor data in different parts. Specifically, the rhizosphere nodes, stem-leaf nodes and canopy nodes are matched on the time axis according to the order of physical transfer. For example, the rhizosphere water change occurs first, the stem flow change occurs later and the canopy transpiration change occurs last. At the same time, the effective response period between adjacent driving nodes is determined, that is, the reasonable time window from the start of data change in the previous node to the production of detectable response data change in the next node. Within a defined effective response period, based on the correlation of driving intensity changes and the consistency of response timing, the causal relationship between upper and lower level nodes is verified and calculated. The correlation of driving intensity changes refers to calculating the correlation between data changes at two nodes. For example, the correlation coefficient or regression coefficient between the decrease in rhizosphere soil moisture content and the decrease in stem flow rate is calculated. A higher coefficient indicates a stronger driving force of rhizosphere supply on stem-leaf transport. Response timing consistency refers to checking whether the actual time delay from rhizosphere change to stem-leaf change is stable within the preset effective response period. Excessive delay or large fluctuations indicate an unstable correlation. Then, based on the verification results, directional logical connections are established between rhizosphere driving nodes and stem-leaf driving nodes, and between stem-leaf driving nodes and canopy driving nodes. Directed driving correlation edges are determined, with a fixed direction: rhizosphere → stem-leaf → canopy. Based on these directed driving correlation edges, growth driving pathways representing the energy and material transfer relationship between rhizosphere, stem-leaf, and canopy are generated. This represents the data-level mapping of the complete process of water, heat, and light energy entering the root system from the soil, being transported through the stem, and finally released in the canopy leaves.

[0026] Step S300: Using the growth-driven pathway, perform evaluation analysis of the three-dimensional growth-driven continuity evaluation channel, and establish growth-driven continuity evaluation results. The three-dimensional growth-driven continuity evaluation channel includes a pathway response delay sub-channel, an energy decay degree sub-channel, and a pathway stability sub-channel.

[0027] Step S300 further includes: on the growth driving path, activating a path response delay sub-channel based on the drive response time difference between adjacent drive nodes, wherein the path response delay sub-channel is used to analyze the temporal distribution characteristics of the drive response time difference and generate a response delay evaluation result characterizing the temporal consistency of drive transmission; on the growth driving path, activating an energy attenuation degree sub-channel based on the drive intensity transmission ratio between adjacent drive nodes, wherein the activated energy attenuation degree sub-channel is used to analyze the drive intensity attenuation trend and cumulative attenuation characteristics based on the drive intensity transmission ratio and generate an energy attenuation evaluation result characterizing the integrity of drive transmission; on the growth driving path, activating a path stability sub-channel based on the degree of maintenance of drive correlation within a continuous time window, wherein the path stability sub-channel analyzes the fluctuation amplitude and duration of drive correlation to generate a path stability evaluation result characterizing structural stability; and establishing a growth driving continuity evaluation result based on the response delay evaluation result, energy attenuation evaluation result, and path stability evaluation result.

[0028] Preferably, the evaluation and analysis are conducted through a three-dimensional growth-driven continuity evaluation channel. This channel includes a pathway response delay sub-channel, an energy attenuation degree sub-channel, and a pathway stability sub-channel. Specifically, the pathway response delay sub-channel is a quantitative evaluation unit for the transmission speed of energy and matter between the rhizosphere, stem-leaf, and canopy. It is activated based on the driving response time difference between adjacent driving nodes in the growth-driven pathway. The driving response time difference refers to the time interval between the moment when the data at the rhizosphere node changes and the moment when the data at the stem-leaf node changes accordingly. It also applies to the interval between the stem-leaf node and the canopy node. The pathway response delay sub-channel is used to analyze the temporal distribution characteristics of the driving response time difference. That is, it performs statistical analysis on multiple driving response time difference values ​​generated within multiple consecutive time windows, calculates their mean, variance, range, and other statistical quantities, and judges whether the response speed is stable and whether there is a trend of slowing down. The response delay evaluation result is output to characterize the temporal consistency of driving transmission. For example, if the driving response time difference value is stable within a small range, it is evaluated as "low delay and good consistency". If the driving response time difference value fluctuates greatly or continues to increase, it is evaluated as "unstable delay or increased delay".

[0029] Preferably, the energy attenuation subchannel is a quantitative evaluation unit for the intensity loss of energy and matter during transport. It is activated in the growth-driven pathway based on the ratio of driving intensity transfer between adjacent driving nodes. The driving intensity transfer ratio refers to the ratio of the driving intensity of the later node to the driving intensity of the earlier node. For example, the rhizosphere to stem-leaf transfer ratio is the ratio of the stem-leaf node driving intensity stem flow rate to the reciprocal of the rhizosphere node driving intensity soil water potential; the stem-leaf to canopy transfer ratio is the ratio of the canopy node driving intensity transpiration rate to the stem-leaf node driving intensity stem flow rate. Activating the energy attenuation subchannel is used for driving intensity transfer based on the ratio of the driving intensity of the stem-leaf node to the stem flow rate. The dynamic intensity transmission ratio is analyzed to determine the driving intensity decay trend and cumulative decay characteristics. The driving intensity decay trend refers to the change of the transmission ratio over time, whereby it gradually increases (decreases decay) or gradually decreases (intensifies decay). The cumulative decay characteristics refer to the total decay degree from the rhizosphere to the canopy, which is the ratio of the canopy output intensity to the original input intensity of the rhizosphere. The resulting energy decay evaluation results are used to characterize the integrity of the driving transmission. For example, if the transmission ratio is high and stable, it is evaluated as "low decay degree and complete transmission"; if the transmission ratio is low or continuously decreasing, it is evaluated as "severe energy loss and obstructed transmission".

[0030] Preferably, the pathway stability subchannel is a quantitative evaluation unit for the robustness of the causal relationship between the rhizosphere, stem and leaf, and canopy. It is activated based on the degree of maintenance of the driving association within a continuous time window in the growth-driven pathway. The degree of maintenance of the driving association refers to whether the established directed driving association edge remains significantly valid over multiple consecutive time windows. The pathway stability subchannel analyzes the fluctuation amplitude and duration of the driving association, i.e., the degree of variation in the correlation coefficient or regression coefficient. Large fluctuations indicate that the causal relationship is sometimes strong and sometimes weak. Simultaneously, it analyzes the duration of continuous validity of the association; short durations indicate that the pathway structure is fragile and easily disturbed. This outputs a pathway stability evaluation result to characterize the robustness of the driving structure. For example, if the association maintains a high coefficient with small fluctuations for a long period, it is evaluated as "structurally stable"; if the association frequently breaks or the coefficient fluctuates, it is evaluated as "structurally unstable". Finally, the response delay evaluation result, energy decay evaluation result, and pathway stability evaluation result are combined to generate a growth-driven continuity evaluation result, which describes whether the overall operation of the soybean growth-driven pathway is smooth, efficient, and reliable.

[0031] Step S400: Based on the growth-driven continuity evaluation results, determine the current growth status of soybeans and establish a first determination result.

[0032] Step S400 further includes: performing evaluation pattern recognition on the growth-driven continuity evaluation results, extracting a combination of continuity features reflecting the temporal consistency, transmission integrity, and structural stability of the pathway; determining the continuity state type of the growth-driven pathway based on the combination of continuity features, wherein the continuity state type includes a continuous stable state, a continuous fluctuating state, and a continuous interrupted state; mapping the continuity state type to the corresponding soybean growth state category, and establishing a first discrimination result, wherein the soybean growth state category includes an effective growth state, a growth-restricted state, and a physiological stress state.

[0033] Preferably, the evaluation results of growth-driven continuity are comprehensively analyzed to identify corresponding evaluation patterns, such as the "three highs" pattern or the "one low and two highs" pattern. Key continuity feature combinations representing the overall operational status of the pathway, including the consistency of transmission timing, transmission integrity, and structural stability, are extracted from the evaluation pattern identification results. These include the mean and coefficient of variation of the response time difference, the total transmission efficiency from the rhizosphere to the canopy and its decay rate, and the correlation coefficient strength of the driving association edge and its duration on the time axis. Then, based on the extracted continuity feature combinations, the current growth state of soybean is determined, i.e., the growth-driven pathway is qualitatively classified to determine the continuity state type of the growth-driven pathway, including continuous stable state, continuous fluctuating state, and continuous interrupted state. The continuous stable state refers to a state where all three dimensions of characteristics are excellent. A good state is characterized by a short and stable response time difference, a low energy decay ratio, and a strong and long-lasting driving correlation. This state indicates that the material and energy transport channels between the rhizosphere, stems and leaves, and the canopy are completely unobstructed, efficient, and reliable. A continuous fluctuating state indicates that the characteristics are intermittent or oscillating, with the response time difference fluctuating, the energy decay ratio fluctuating, and the driving correlation fluctuating. This state indicates that although the transport channels are not completely interrupted, they are unstable and may be affected by environmental fluctuations or the plant's own rhythms. A continuous interrupted state indicates that the characteristics are severely blocked or broken, with the response time difference being extremely large or no effective response can be detected, the energy decay ratio approaching zero, and the driving correlation being insignificant or disappearing. This state indicates that the causal chain between the rhizosphere, stems and leaves, and the canopy is broken, and the material and energy transport channels are blocked or closed.

[0034] Preferably, continuous state types are mapped to corresponding soybean growth state categories, which include effective growth state, growth-restricted state, and physiological stress state. Specifically, a continuous stable state is mapped to an effective growth state, where the soybean root system's absorption function, stem transport function, and leaf transpiration and photosynthesis function operate efficiently and in synergy, and the plant is in a normal, uninhibited growth and development stage. A continuous fluctuating state is mapped to a growth-restricted state, where transport channels are unstable, leading to inconsistent water or nutrient supply or fluctuations in energy conversion efficiency. Although the plant is not dead or severely damaged, its growth rate and biomass accumulation are inhibited, possibly caused by mild drought, temporary waterlogging, or nutrient imbalance. A continuous interrupted state is mapped to a physiological stress state, where transport channels are severely blocked or broken, the rhizosphere cannot supply nutrients, and the canopy cannot release nutrients, disrupting the plant's normal physiological metabolic functions, corresponding to severe drought stress, waterlogging, disease, or scorching caused by extreme high temperatures. Finally, a first discrimination result is generated, including labels for growth state categories such as effective growth, restricted growth, and stress.

[0035] Step S500: Obtain a monitoring image of soybeans and use the monitoring image to establish a second determination result.

[0036] Step S500 further includes performing crop region separation processing on the monitoring image to extract soybean plant image regions; using the soybean plant image regions, extracting an image feature set characterizing the spatial morphological features and canopy structure features of the plants, the image feature set including plant coverage features, canopy density features and morphological consistency features; establishing a description of the soybean population growth morphology based on the image feature set, and establishing a second judgment result.

[0037] Preferably, monitoring images of soybeans are acquired using imaging equipment mounted on a visible light camera, multispectral camera, or drone. These images may include top-down or side views of the soybean canopy. Image segmentation algorithms are used to separate crop regions based on color features (such as the green vegetation index), texture features, or depth information. This includes pixel-level separation of soybean plants from non-plant backgrounds such as soil, weeds, shadows, and stubble, which is presented as a binary mask. In this binary mask, white represents plants and black represents the background, thereby extracting the soybean plant image region.

[0038] Preferably, the soybean plant image region is quantitatively calculated to extract a set of image features characterizing the spatial morphological and canopy structural features of the plant, including plant coverage features, canopy density features, and morphological consistency features. Specifically, plant coverage features refer to the percentage of the vertical projection area of ​​the soybean plant canopy on a unit land area. This is calculated by dividing the number of pixels in the soybean plant image region by the total number of pixels in the entire monitoring area, reflecting the degree of land occupation by the soybean population and the potential area for photosynthesis. Canopy density features refer to the density of canopy leaves in three-dimensional space, reflecting... The porosity of the canopy is indirectly calculated by analyzing the roughness of the image texture, the proportion of light-transmitting pores, or by calculating the leaf area index based on the canopy height model. High canopy density indicates that there are many overlapping leaves and less light transmission; low canopy density indicates that the canopy is sparse and there is more light leakage. Morphological uniformity refers to the degree of uniformity in size, shape, and color among different plants in the monitoring area. This is achieved by segmenting individual plants and calculating the coefficient of variation of morphological parameters such as projected area, plant height, and roundness of all plants, i.e., the ratio of standard deviation to mean. The smaller the coefficient of variation, the higher the morphological uniformity and the more uniform the growth of the group.

[0039] Preferably, based on the image feature set, a description of the soybean population growth morphology is established from a visual phenotypic perspective. This includes converting the extracted features into linguistic descriptions. For example, high coverage, high density, and good uniformity are converted into "lush population, good canopy closure, and uniform growth," while low coverage, low density, and poor uniformity are converted into "sparse population, missing seedlings and gaps in the canopy, and uneven growth." High coverage, abnormally high density, and dark green color are converted into "potentially excessive growth or nitrogen excess." Finally, the description of the soybean growth morphology is quantified and encoded to generate a second judgment result, which may include growth status category labels such as lush, sparse, uniform, and uneven.

[0040] Step S600: Monitor and fuse the first determination result and the second determination result, and output the fused monitoring result.

[0041] Step S600 further includes: constructing a growth state consistency determination relationship based on the first determination result and the second determination result, wherein the growth state consistency determination relationship characterizes the degree of consistency between the two determination results in the growth state category; performing monitoring fusion using the growth state consistency determination relationship, and outputting the fused monitoring result.

[0042] Preferably, the first and second judgment results are jointly analyzed to establish a logical correspondence between them. This involves comparing the specific growth state category labels in the two judgment results to determine whether the conclusions of the two judgment results when describing the same plant or the same area are mutually corroborative, contradictory, or partially consistent. For example, if the first judgment result is "effective growth state" and the second judgment result is "lush population and uniform growth," it indicates that the two judgment results are highly consistent. If the first judgment result is "physiological stress state" and the sensor shows that root and stem transmission is interrupted, but the second judgment result is "lush population" and the image shows that the canopy is still green and has high coverage, it indicates that the plant has died but the leaves have not withered or that it has recovered after temporary wilting caused by disease. This leads to the determination of the consistency judgment relationship of growth state, which characterizes the degree of consistency between the two judgment results in terms of growth state category, such as "completely consistent," "partially consistent," or "inconsistent."

[0043] Preferably, monitoring fusion refers to comprehensively reasoning about the first and second judgment results based on the consistency judgment relationship, generating and outputting a fused monitoring result. Specifically, when the two judgment results are highly consistent, the fused monitoring result directly adopts this common conclusion and has the highest credibility; when the two are partially consistent, the fused result may take the union or weighted average. For example, if the sensor shows "growth is restricted" and the image shows "uniform population", the fused monitoring result may be "uniform population but restricted growth"; when the two are inconsistent, the final fused monitoring result is determined according to the preset priority. The output fused monitoring result is used to comprehensively analyze the intrinsic physiological evidence and extrinsic phenotypic evidence of soybean growth, achieving accurate determination of the soybean growth driving state. This can avoid misjudgments caused by sensor drift, image occlusion, or abnormal single indicators, thereby improving the accuracy and comprehensiveness of soybean growth status monitoring.

[0044] In the above text, refer to Figure 1 A soybean growth monitoring method based on monitoring sensors according to embodiments of the present invention has been described in detail. Next, reference will be made to... Figure 2 A soybean growth monitoring system based on monitoring sensors according to an embodiment of the present invention is described.

[0045] The soybean growth monitoring system based on monitoring sensors according to embodiments of the present invention addresses the technical problems in the prior art, such as insufficient information correlation, lack of evaluation continuity, and low data fusion, which make it difficult to comprehensively and accurately reveal the driving state of soybean growth. It achieves precise determination of the driving state of soybean growth, thereby improving the accuracy and comprehensiveness of soybean growth status monitoring. Figure 2As shown, the soybean growth monitoring system based on monitoring sensors includes: a data acquisition module 10, a growth-driven pathway construction module 20, an evaluation result establishment module 30, a growth status determination module 40, a second determination result establishment module 50, and a fusion monitoring result output module 60.

[0046] The data acquisition module 10 is used to deploy distributed monitoring sensors within the soybean planting area. These sensors collect multi-source growth-driven data characterizing rhizosphere supply status, canopy energy release status, and near-surface transport conditions. The growth-driven pathway construction module 20 is used to perform temporal correlation and path reconstruction on the multi-source growth-driven data according to the direction of water, heat, and light energy transfer during soybean growth, constructing a growth-driven pathway characterizing the energy and material transfer relationship between the rhizosphere, stems / leaf, and canopy. The evaluation result establishment module 30 is used to perform evaluation analysis of a three-dimensional growth-driven continuity evaluation channel using the growth-driven pathway, establishing a growth-driven continuity evaluation result. This three-dimensional growth-driven continuity evaluation channel includes a pathway response delay sub-channel, an energy attenuation degree sub-channel, and a pathway stability sub-channel. The growth state determination module 40 is used to determine the current growth state of the soybean based on the growth-driven continuity evaluation result, establishing a first determination result. The second determination result establishment module 50 is used to acquire monitoring images of the soybean and establish a second determination result using these images. The fusion monitoring result output module 60 is used to fuse the first and second determination results and output a fused monitoring result.

[0047] The specific configuration of the growth-driven pathway construction module 20 will be described in detail below. The growth-driven pathway construction module 20 further includes: constructing sets of driving nodes based on the multi-source growth-driven data to characterize the rhizosphere supply state, stem-leaf transport state, and canopy energy release state, wherein each driving node is characterized by the driving intensity characteristics and changes within a corresponding time window; performing cross-layer temporal alignment processing on the rhizosphere driving nodes, stem-leaf driving nodes, and canopy driving nodes according to the direction of water, heat, and light energy transfer during soybean growth to determine the effective response time period between adjacent driving nodes; establishing directed driving association edges between rhizosphere driving nodes and stem-leaf driving nodes, and between stem-leaf driving nodes and canopy driving nodes, based on the correlation of changes in driving intensity and the consistency of response time within the effective response time period; and generating growth-driven pathways characterizing the energy and material transfer relationship between the rhizosphere, stem-leaf, and canopy based on the directed driving association edges.

[0048] The specific configuration of the evaluation result establishment module 30 will be described in detail below. The evaluation result establishment module 30 further includes: On the growth driving path, activating a path response delay sub-channel based on the driving response time difference between adjacent driving nodes; the path response delay sub-channel is used to analyze the temporal distribution characteristics of the driving response time difference and generate a response delay evaluation result characterizing the temporal consistency of driving transmission; On the growth driving path, activating an energy attenuation degree sub-channel based on the driving intensity transmission ratio between adjacent driving nodes; the activated energy attenuation degree sub-channel is used to analyze the driving intensity attenuation trend and cumulative attenuation characteristics based on the driving intensity transmission ratio and generate an energy attenuation evaluation result characterizing the integrity of driving transmission; On the growth driving path, activating a path stability sub-channel based on the degree of maintenance of the driving correlation within a continuous time window; the path stability sub-channel analyzes the fluctuation amplitude and duration of the driving correlation to generate a path stability evaluation result characterizing structural stability; and establishing a growth driving continuity evaluation result based on the response delay evaluation result, energy attenuation evaluation result, and path stability evaluation result.

[0049] The specific configuration of the growth state determination module 40 will be described in detail below. The growth state determination module 40 further includes: performing evaluation pattern recognition on the growth-driven continuity evaluation results, extracting a combination of continuity features reflecting the temporal consistency, transmission integrity, and structural stability of the pathway; determining the continuity state type of the growth-driven pathway based on the combination of continuity features, wherein the continuity state type includes a continuous stable state, a continuous fluctuating state, and a continuous interrupted state; mapping the continuity state type to the corresponding soybean growth state category, and establishing a first discrimination result, wherein the soybean growth state category includes an effective growth state, a growth-restricted state, and a physiological stress state.

[0050] The specific configuration of the second determination result establishment module 50 will be described in detail below. The second determination result establishment module 50 further includes: performing crop region separation processing on the monitoring image to extract soybean plant image regions; using the soybean plant image regions, extracting an image feature set characterizing the spatial morphological features and canopy structure features of the plants, the image feature set including plant coverage features, canopy density features, and morphological consistency features; establishing a description of the soybean population growth morphology based on the image feature set, and establishing the second determination result.

[0051] The specific configuration of the fusion monitoring result output module 60 will be described in detail below. The fusion monitoring result output module 60 further includes: constructing a growth state consistency determination relationship based on the first determination result and the second determination result, wherein the growth state consistency determination relationship characterizes the degree of consistency between the two types of determination results in the growth state category; performing monitoring fusion using the growth state consistency determination relationship, and outputting the fusion monitoring result.

[0052] The specific configuration of the fusion monitoring result output module 60 will be described in detail below. The fusion monitoring result output module 60 further includes: performing anomaly analysis of soybean growth based on the fusion monitoring results, configuring anomaly warning signals, and performing anomaly reporting management of growth monitoring according to the anomaly warning signals.

[0053] The soybean growth monitoring system based on monitoring sensors provided in this embodiment of the invention can execute the soybean growth monitoring method based on monitoring sensors provided in this embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0054] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A soybean growth monitoring method based on monitoring sensors, characterized in that, The method includes: In soybean planting areas, distributed monitoring sensors are deployed to collect multi-source growth-driving data characterizing rhizosphere supply status, canopy energy release status, and near-surface transport conditions. Based on the direction of water, heat and light energy transfer during soybean growth, the multi-source growth-driven data are temporally correlated and reconstructed to build a growth-driven pathway that characterizes the relationship between energy and material transfer in the rhizosphere-stem-leaf-canopy. Using the growth-driven pathway, an evaluation analysis of the three-dimensional growth-driven continuity evaluation channel is performed to establish the growth-driven continuity evaluation results. The three-dimensional growth-driven continuity evaluation channel includes a pathway response delay sub-channel, an energy decay degree sub-channel, and a pathway stability sub-channel. Based on the growth-driven continuity evaluation results, the current growth status of soybeans is determined, and a first determination result is established; Acquire monitoring images of soybeans, and use the monitoring images to establish a second determination result; The first and second determination results are monitored and fused to output the fused monitoring result; Constructing growth-driven pathways that characterize the energy and mass transfer relationships between the rhizosphere, stems and leaves, and the canopy, including: Based on the multi-source growth driving data, sets of driving nodes are constructed to characterize the rhizosphere supply state, stem and leaf transport state, and canopy energy release state, respectively. Each driving node is characterized by the driving intensity features and changes within the corresponding time window. Based on the direction of water, heat and light energy transfer during soybean growth, cross-layer temporal alignment was performed on rhizosphere driving nodes, stem and leaf driving nodes and canopy driving nodes to determine the effective response time period between adjacent driving nodes. During the effective response period, based on the correlation between changes in driving intensity and the consistency of response timing, directed driving association edges are established between rhizosphere driving nodes and stem-leaf driving nodes, as well as between stem-leaf driving nodes and canopy driving nodes. Based on the directed driving association edges, a growth-driven pathway representing the energy and material transfer relationship between the rhizosphere, stem and leaf, and canopy is generated. Based on the growth-driven continuity evaluation results, the current growth status of soybeans is determined, and a first determination result is established, including: Evaluation pattern recognition is performed on the growth-driven continuity evaluation results to extract a combination of continuity features that reflect the temporal consistency, transmission integrity, and structural stability of the pathway. Based on the combination of the continuity features, the continuity state type of the growth driving pathway is determined, including continuous stable state, continuous fluctuating state and continuous interrupted state. The continuous state type is mapped to the corresponding soybean growth state category to establish a first discrimination result, wherein the soybean growth state category includes effective growth state, growth-restricted state and physiological stress state; Acquire monitoring images of soybeans, and establish a second determination result using the monitoring images, including: The monitored images were processed to separate crop regions, and soybean plant image regions were extracted. Using the soybean plant image region, an image feature set representing the spatial morphological features and canopy structure features of the plant is extracted. The image feature set includes plant coverage features, canopy density features, and morphological consistency features. A description of the growth morphology of soybean populations is established based on the image feature set, and a second determination result is established.

2. The soybean growth monitoring method based on monitoring sensors as described in claim 1, characterized in that, Using the aforementioned growth-driven pathway, an evaluation analysis of the three-dimensional growth-driven continuity evaluation channel is performed, including: On the growth driving path, a path response delay sub-channel is activated based on the driving response time difference between adjacent driving nodes. The path response delay sub-channel is used to analyze the temporal distribution characteristics of the driving response time difference and generate a response delay evaluation result to characterize the consistency of driving transmission timing. On the growth driving path, an energy attenuation sub-channel is activated based on the driving intensity transfer ratio between adjacent driving nodes. The activated energy attenuation sub-channel is used to analyze the driving intensity attenuation trend and cumulative attenuation characteristics based on the driving intensity transfer ratio, and to generate an energy attenuation evaluation result to characterize the integrity of driving transfer. In the growth-driven pathway, a pathway stability sub-channel is activated based on the degree of maintenance of the driving correlation within a continuous time window. The pathway stability sub-channel generates a pathway stability evaluation result characterizing the structural stability by analyzing the fluctuation amplitude and duration of the driving correlation. The growth-driven continuity evaluation results were established based on the response delay evaluation results, energy decay evaluation results, and pathway stability evaluation results.

3. The soybean growth monitoring method based on monitoring sensors as described in claim 1, characterized in that, The first and second determination results are monitored and fused to output the fused monitoring results, including: Based on the first and second determination results, a growth state consistency determination relationship is constructed, which represents the degree of consistency between the two determination results in the growth state category. The monitoring fusion is performed using the aforementioned growth state consistency determination relationship, and the fusion monitoring results are output.

4. The soybean growth monitoring method based on monitoring sensors as described in claim 1, characterized in that, The output of fusion monitoring results also includes: Based on the fusion monitoring results, perform anomaly analysis on soybean growth and configure anomaly early warning signals; Based on the aforementioned abnormal warning signal, perform abnormal reporting management for growth monitoring.

5. A soybean growth monitoring system based on monitoring sensors, characterized in that, The system is used to implement the soybean growth monitoring method based on monitoring sensors according to any one of claims 1 to 4, and the system comprises: The data acquisition module is used to deploy distributed monitoring sensors in the soybean planting area. The distributed monitoring sensors are used to collect multi-source growth driving data that characterize the rhizosphere supply status, canopy energy release status, and near-surface transport conditions. The growth-driven pathway construction module is used to perform temporal correlation and path reconstruction on the multi-source growth-driven data according to the direction of water, heat and light energy transfer during soybean growth, and to construct a growth-driven pathway that characterizes the relationship between energy and material transfer in the rhizosphere-stem-leaf-canopy. The evaluation result establishment module is used to perform evaluation analysis of the three-dimensional growth-driven continuity evaluation channel using the growth-driven pathway, and establish the growth-driven continuity evaluation result. The three-dimensional growth-driven continuity evaluation channel includes a pathway response delay sub-channel, an energy decay degree sub-channel, and a pathway stability sub-channel. The growth status determination module is used to determine the current growth status of soybeans based on the growth-driven continuity evaluation results and establish a first determination result. The second determination result establishment module is used to acquire monitoring images of soybeans and establish a second determination result using the monitoring images; The fusion monitoring result output module is used to monitor and fuse the first judgment result and the second judgment result, and output the fusion monitoring result.

Images