Gastrodia elata growth monitoring method and system based on image recognition

Abstract

The invention provides a gastrodia elata growth monitoring method and system based on image recognition, and relates to the technical field of gastrodia elata growth monitoring, and the method comprises the core steps: dividing a gastrodia elata planting region into a plurality of grids, collecting related environment data through an environment sensor, constructing a growth monitoring model through a convolutional neural network, and carrying out the calculation of the growth monitoring model. The growth data of the gastrodia elata are obtained, and model output is further analyzed to obtain health indexes. According to the method, the growth condition and potential diseases of the gastrodia elata can be quickly identified through accurate growth data analysis. According to the method, a more comprehensive growth monitoring means is provided, and the comprehensive growth index is calculated by comprehensively analyzing the health index, the environmental data and the planting density, so that the growth state of the gastrodia elata is comprehensively and scientifically evaluated. By means of the method, a planter can make scientific decisions in time, management measures are optimized, and finally the yield and quality of gastrodia elata are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Gastrodia elata growth monitoring, and specifically provides a Gastrodia elata growth monitoring method and system based on image recognition. Background Art

[0002] Gastrodia elata is a precious traditional Chinese medicine and has received extensive attention due to its unique medicinal value and economic value. With the increasing emphasis on the concept of health preservation among people, the market demand for Gastrodia elata continues to grow, driving the expansion of its planting area. However, the growth cycle of Gastrodia elata is relatively long, and it has strict requirements for environmental conditions. Changes in factors such as soil moisture, light intensity, and temperature will directly affect its growth quality. Therefore, it is particularly important to monitor the growth status and health condition of Gastrodia elata in order to take management measures in a timely manner to ensure yield and quality.

[0003] The growth of Gastrodia elata is affected by various factors such as environmental conditions, soil quality, and diseases. Changes in these factors are directly related to the growth status and yield of Gastrodia elata. However, traditional monitoring methods mainly rely on manual observation and empirical judgment, which are not only inefficient but also prone to subjective biases and difficult to provide scientific and accurate growth data. This results in the inability of growers to identify growth problems in a timely manner during actual planting, affecting the quality and yield of Gastrodia elata. In addition, the lack of systematic data support also limits the effectiveness of scientific research and management decision-making.

[0004] The deficiencies of the prior art are as follows:

[0005] Relying on manual inspections and traditional growth assessment methods, in the face of complex growth environments and potential diseases, it is difficult for traditional methods to quickly identify problems, thus affecting the timely response of growers to the growth status of Gastrodia elata and increasing the risk and loss of disease spread. In addition, existing technologies often lack systematic monitoring of environmental factors and are unable to deeply analyze variables closely related to the growth of Gastrodia elata such as soil quality and climate change. This one-sided monitoring method makes growers lack a comprehensive understanding of the growth environment of Gastrodia elata, further limiting the scientific assessment of the growth status and the effectiveness of management decision-making. Therefore, the deficiencies of the prior art lie in the singularity and lag of its monitoring means, which cannot meet the requirements of modern agriculture for efficient, precise, and intelligent management, nor can it effectively promote the sustainable development of Gastrodia elata planting.

[0006] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0007] The purpose of the present invention is to provide a Gastrodia elata growth monitoring method and system based on image recognition to solve the problems raised in the above background art.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A Gastrodia elata growth monitoring method based on image recognition, the specific steps include:

[0010] Step 1: Divide the soil area where Gastrodia elata grows into several grids, and collect the environmental data of each grid at the current moment. The environmental data includes soil humidity, temperature, and light intensity;

[0011] Step 2: Obtain multiple Gastrodia elata growth images with known growth data. Using the Gastrodia elata growth images as input and the corresponding growth data as output, construct and train a growth monitoring model based on a convolutional neural network;

[0012] Step 3: Divide all grids into multiple community sets according to the environmental data of each grid. Select a current sampling area within each community set, and determine the planting density of the current sampling area based on the number of tubers and the area of the sampling area of Gastrodia elata within the current sampling area;

[0013] Step 4: Obtain the growth image of Gastrodia elata within the current sampling area, and input it into the growth monitoring model to obtain the growth data of Gastrodia elata within the current sampling area, and analyze the health status of Gastrodia elata within the current sampling area based on the growth data;

[0014] Step 5: Generate a comprehensive growth index according to the health status, environmental data, and planting density of Gastrodia elata in the current sampling area within each community set to analyze the growth status of Gastrodia elata in each grid, and screen out the grids with abnormal growth status based on the comprehensive growth index.

[0015] Further, collecting the environmental data of each grid at the current moment specifically includes:

[0016] Evenly divide the planting soil area of Gastrodia elata into multiple grids, and arrange soil humidity sensors, temperature sensors, and light sensors at the grid centers to collect the environmental data of the Gastrodia elata growth area.

[0017] Further, constructing and training the growth monitoring model specifically includes:

[0018] Obtain multiple Gastrodia elata growth images with known growth data. Each group of Gastrodia elata growth images includes images of Gastrodia elata taken from various angles. The growth data includes tuber surface area, tuber volume, and the proportion of the area with pests and diseases on the tuber. Use the average value of the tuber surface area of each group of Gastrodia elata growth images as the tuber surface area of the corresponding Gastrodia elata. The proportion refers to the ratio of the surface area with pests and diseases on the tuber to the tuber surface area. The formula for calculating the proportion of the area with pests and diseases on the tuber is:

[0019]

[0020] Among them, P is the proportion of the area with pests and diseases on the tuber, and N d is the surface area of the area with pests and diseases in the tuber, and N t is the surface area of the tuber;

[0021] Label the growth data for each group of Gastrodia elata growth images. Preprocess each group of labeled Gastrodia elata growth images, including removing blurred and duplicate images, randomly rotating, flipping, cropping, and adjusting the brightness of the images, and finally unifying the image size;

[0022] Using the labeled and preprocessed Gastrodia elata growth images as inputs, select a convolutional neural network as the basic model, and the corresponding growth data as labels to train the convolutional neural network model and construct a growth monitoring model.

[0023] Furthermore, determining the planting density of the grid specifically includes:

[0024] Take each grid as a node, and use the environmental data corresponding to each node as the attributes of the node to construct an undirected graph. If two nodes are adjacent, add an edge in the undirected graph to form an undirected graph network;

[0025] Place each node in an independent community. For each node in the undirected graph network, move it successively to the communities where all its neighbor nodes are located, and calculate the increased value of the environmental change corresponding to each move, specifically including:

[0026] First, calculate the mean value of the environmental data between a node and any of its neighbor nodes, and then analyze the change between the node and the mean value to obtain the increased value of the environmental change. The calculation formula is:

[0027]

[0028] Among them, △E(ij) is the increased value of the environmental change between the i-th node and the j-th node, is the mean soil humidity between the i-th node and the j-th node, M(i) is the soil humidity of the i-th node, is the mean temperature between the i-th node and the j-th node, T(i) is the temperature of the i-th node, is the mean light intensity between the i-th node and the j-th node, L(i) is the light intensity of the i-th node, and i and j are both nodes in the undirected graph, and i and j are neighbor nodes;

[0029] Move the node to the neighbor community with the smallest increased value of the environmental change, repeat moving the nodes until the community membership of all nodes no longer changes. At this time, randomly select a node in each community as the current sampling area corresponding to all nodes in the community set, obtain the planting density of the current sampling area, and the calculation formula for the planting density is:

[0030]

[0031] Among them, D is the planting density of the current sampling area, n is the number of tubers in the current sampling area, and A is the area of the current sampling area;

[0032] Regarding the planting density of the current sampling area as the planting density of the community set where the current sampling area is located, the growth area of Gastrodia elata can be classified by the planting density: if D = 0, it is judged as an area without Gastrodia elata; if 0 < D < T1, it is judged as a sparse planting area; if T1 ≤ D < T2, it is judged as a lush planting area; if D ≥ T2, it is judged as an overcrowded planting area. When the growth area of Gastrodia elata is in the area without Gastrodia elata and the sparse planting area, a signal to increase the planting of Gastrodia elata in this grid is sent to the planter. When the growth area of Gastrodia elata is in the overcrowded planting area, a signal to reduce the planting of Gastrodia elata in this grid is sent to the planter. T1 is the sparse threshold, and T is the dense threshold.

[0033] Furthermore, analyzing the health status of Gastrodia elata in the current sampling area specifically includes:

[0034] Obtain the growth image of Gastrodia elata in the current sampling area, input it into the trained growth monitoring model, obtain the growth data of Gastrodia elata in the current sampling area output by the model, and generate a health index based on the growth data of Gastrodia elata to evaluate the health status of Gastrodia elata in the current sampling area. The calculation formula is:

[0035] H = α1·ln(1 + S) + α2·ln(1 + V) - α3·(e P -1)

[0036] Among them, H is the health index, S is the tuber size, V is the tuber volume, P is the proportion of the area with pests and diseases on the tuber, and α1, α2, and α3 are the weight coefficients of each item, 0 < α1 < α2 < α3 < 1, and α1 + α2 + α3 = 1; regarding the health index of Gastrodia elata in the current sampling area as the health index of the grid where the current sampling area is located.

[0037] Furthermore, analyzing the environmental data specifically includes:

[0038] According to the environmental data of each current sampling area, calculate the environmental comprehensive index to evaluate the environment in the soil area where Gastrodia elata grows. The calculation formula is:

[0039]

[0040] Among them, E is the environmental comprehensive index, M is the soil humidity of the current sampling area, T is the temperature of the current sampling area, L is the light intensity of the current sampling area, M max 、T max 、Lmax are the maximum suitable values of soil humidity, temperature, and light intensity, d, e, and f are the weight coefficients of soil humidity, temperature, and light intensity, and 0 < f < e < d < 1.

[0041] Further, screening the current sampling areas with abnormal growth conditions specifically includes:

[0042] The formula for calculating the comprehensive growth index of each current sampling area is:

[0043]

[0044] where I is the comprehensive growth index, H is the health index, D is the planting density of the current sampling area, λ1, λ2, and λ3 are the weight coefficients of the health index, environmental comprehensive index, and planting density of Gastrodia elata, 0 < λ3 ≤ λ2 < λ1 < 1, and λ1 + λ2 + λ3 = 1;

[0045] Compare the comprehensive growth index of each current sampling area with the preset growth threshold. If the comprehensive growth index of the current sampling area at the current moment is higher than the preset growth threshold, it indicates that the growth condition of Gastrodia elata in this current sampling area is good, that is, it indicates that all grids in the community set corresponding to this current sampling area are in good growth state of Gastrodia elata, then normal monitoring is maintained. If the comprehensive growth index of the current sampling area at the current moment is lower than the preset growth threshold, it indicates that the growth state of Gastrodia elata in this current sampling area is poor, that is, it indicates that all grids in the community set corresponding to this current sampling area are in poor growth state of Gastrodia elata, then all grids in the community set of this current sampling area are marked as grids with abnormal growth, and a warning of poor growth condition is sent. And for the grids marked with abnormal growth, for their adjacent grids, the environmental comprehensive index is also judged. If the environmental comprehensive index of an adjacent grid is lower than the preset environmental threshold, then this adjacent grid is also marked as abnormal and a warning of poor growth condition is sent.

[0046] The present invention also provides a Gastrodia elata growth monitoring system based on image recognition. The Gastrodia elata growth monitoring system based on image recognition is used to implement the above-mentioned Gastrodia elata growth monitoring method based on image recognition, including:

[0047] A data acquisition module, used to divide the soil area where Gastrodia elata grows into several grids and collect the environmental data of each grid at the current moment. The environmental data includes soil humidity, temperature, and light intensity;

[0048] A model construction module, used to obtain multiple Gastrodia elata growth images with known growth data, use the Gastrodia elata growth images as input and the corresponding growth data as output, and construct and train a growth monitoring model based on a convolutional neural network;

[0049] The planting density analysis module is used to divide all grids into multiple community sets according to the environmental data of each grid, select a current sampling area within each community set, and determine the planting density of the current sampling area based on the number of gastrodia tuber and the sampling area in the current sampling area;

[0050] The real-time monitoring module is used to obtain the growth images of gastrodia in the current sampling area and input them into the growth monitoring model to obtain the growth data of gastrodia in the current sampling area, and analyze the health status of gastrodia in the current sampling area based on the growth data;

[0051] The comprehensive analysis module is used to generate a comprehensive growth index according to the health status, environmental data and planting density of gastrodia in the current sampling area within each community set to analyze the growth status of gastrodia in each grid, and screen out the grids with abnormal growth status according to the comprehensive growth index.

[0052] In the above technical solution, the technical effects and advantages provided by the present invention are as follows:

[0053] The gastrodia growth monitoring method based on image recognition of the present invention can monitor the growth of gastrodia in real time and accurately by introducing modern image processing technology and deep learning algorithms, and can identify diseases and abnormal growth in time. The beneficial effect of this method is that it significantly improves the efficiency and accuracy of monitoring, enables growers to make quick responses based on scientific data, thereby optimizing management measures to ensure the healthy growth of gastrodia. In addition, by comprehensively analyzing the growth data and environmental parameters of gastrodia, this method provides comprehensive decision-making support for growers, promotes the scientific and precise cultivation of gastrodia, and also provides a guarantee for improving the overall yield and quality. Description of the Drawings

[0054] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0055] Figure 2 It is a schematic diagram of the system structure of the present invention. Detailed Embodiments

[0056] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to specific embodiments.

[0057] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meanings understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar terms used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Words such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "linked" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right", etc. are only used to represent relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0058] Embodiment:

[0059] Please refer to Figure 1 , the present invention provides a technical solution:

[0060] A Gastrodia elata growth monitoring method based on image recognition, the specific steps include:

[0061] Step 1: Divide the soil area where Gastrodia elata grows into several grids, and collect the environmental data of each grid at the current moment. The environmental data includes soil humidity, temperature and light intensity;

[0062] In this embodiment, collecting the environmental data of each grid at the current moment specifically includes:

[0063] Evenly divide the planting soil area of Gastrodia elata into multiple grids, and arrange soil humidity sensors, temperature sensors and light sensors at the grid centers to collect the environmental data of the Gastrodia elata growth area. The environmental data includes soil humidity, temperature (here the temperature refers to the ambient temperature) and light intensity.

[0064] The grid-based division method effectively realizes the refined management of the Gastrodia elata planting area, enabling each grid to collect data independently and providing more accurate and localized information for subsequent analysis. This method can help growers promptly identify growth anomalies and environmental changes in specific areas and make targeted adjustments quickly.

[0065] Step 2: Obtain multiple Gastrodia elata growth images with known growth data, use the Gastrodia elata growth images as inputs and the corresponding growth data as outputs, and construct and train a growth monitoring model based on a convolutional neural network;

[0066] In this embodiment, constructing and training the growth monitoring model specifically includes:

[0067] Obtain Gastrodia elata growth images with multiple known growth data. Each group of Gastrodia elata growth images includes images of Gastrodia elata taken from various angles, and the various angles can be the angles taken from four azimuths in the front and two azimuths above and below. The growth data includes tuber surface area, tuber volume, and the proportion of the area with pests and diseases on the tuber. Take the average value of the tuber surface areas of each group of Gastrodia elata growth images as the tuber surface area of the corresponding Gastrodia elata. The proportion refers to the ratio of the surface area of the tuber with pests and diseases to the tuber surface area. The formula for calculating the proportion of the area with pests and diseases on the tuber is:

[0068]

[0069] Where P is the proportion of the area with pests and diseases on the tuber, N d is the surface area of the area with pests and diseases in the tuber, N t is the tuber surface area;

[0070] Label the growth data for each group of Gastrodia elata growth images, and preprocess each group of labeled Gastrodia elata growth images. The preprocessing includes removing blurred and duplicate images, and randomly rotating, flipping, cropping, and adjusting the brightness of the images, and finally unifying the image size;

[0071] Use the Gastrodia elata growth images after labeling and preprocessing as the input, select a convolutional neural network as the basic model, and the corresponding growth data as the label to train the convolutional neural network model and construct a growth monitoring model.

[0072] By manually measuring the growth data of Gastrodia elata tubers, this provides reliable basic data for the subsequent training of the model. This high-quality labeling can effectively improve the learning effect of the convolutional neural network, making the model more accurate and reliable in practical applications. Secondly, the image preprocessing step ensures the consistency and effectiveness of the input data. By removing blurred and duplicate images, and operations such as randomly rotating, flipping, cropping, and adjusting the brightness, the adaptability of the model to images under different conditions can be enhanced. This data augmentation technology not only improves the generalization ability of the model and reduces the risk of overfitting, but also ensures that the model can accurately identify the growth characteristics of Gastrodia elata when facing the variable lighting and angles in the real environment.

[0073] Step 3: Divide all grids into multiple community sets according to the environmental data of each grid. Select a current sampling area within each community set, and determine the planting density of the current sampling area based on the number of Gastrodia elata tubers and the sampling area in the current sampling area;

[0074] In this embodiment, determining the planting density of the grid specifically includes:

[0075] Take each grid as a node and use the environmental data corresponding to each node as the attributes of the node to construct an undirected graph. If two nodes are adjacent, add an edge in the undirected graph to form an undirected graph network;

[0076] Place each node in an independent community. For each node in the undirected graph network, move it successively to the communities where all its neighbor nodes are located, and calculate the incremental value of the environmental change corresponding to each move, specifically including:

[0077] First, calculate the mean value of the environmental data between a node and any of its neighbor nodes, and then analyze the change between the node and the mean value to obtain the incremental value of the environmental change. The calculation formula is:

[0078]

[0079] Among them, △E(ij) is the incremental value of the environmental change between the i-th node and the j-th node, is the mean soil moisture between the i-th node and the j-th node, M(i) is the soil moisture of the i-th node, is the mean temperature between the i-th node and the j-th node, T(i) is the temperature of the i-th node, is the mean light intensity between the i-th node and the j-th node, L(i) is the light intensity of the i-th node, and i and j are both nodes in the undirected graph, and i and j are neighbor nodes;

[0080] Move the node to the neighbor community with the smallest incremental value of the environmental change. Repeat moving the nodes until the community membership of all nodes no longer changes. At this time, randomly select a node in each community as the current sampling area corresponding to all nodes in the community set, obtain the planting density of the current sampling area, and the calculation formula for the planting density is:

[0081]

[0082] Among them, D is the planting density of the current sampling area, n is the number of tubers in the current sampling area, and A is the area of the current sampling area;

[0083] Regard the planting density of the current sampling area as the planting density of the community set where the current sampling area is located. The growth area of Gastrodia elata can be classified by the planting density: if D = 0, it is judged as an area without Gastrodia elata; if 0 < D < T1, it is judged as a sparse planting area; if T1 ≤ D < T2, it is judged as a lush planting area; if D ≥ T2, it is judged as an overcrowded planting area. When the growth area of Gastrodia elata is in the area without Gastrodia elata and the sparse planting area, send a signal to the planter to increase the planting of Gastrodia elata in this grid. When the growth area of Gastrodia elata is in the overcrowded planting area, send a signal to the planter to reduce the planting of Gastrodia elata in this grid. T1 is the sparse threshold, and T is the dense threshold.

[0084] The neighbor nodes refer to the situation where the grid areas corresponding to two nodes are adjacent. The environmental change increment specifically reflects the degree of change in environmental conditions between node i and node j. By calculating the differences between the mean values of soil humidity, temperature, and light intensity and the node's own attributes, the potential impact of environmental changes on plant growth can be quantified. The larger the value of △E(ij), the more obvious the environmental difference between node i and its neighbor node j, which may mean that node i is not suitable as a sampling area under the current environmental conditions, thus suggesting the need to consider moving to a more suitable neighbor community. Here, the purpose is to determine the current sampling area. Therefore, the smaller the calculated environmental change increment between adjacent grids, the better, indicating that the environmental data difference between the two grids is small. Analyzing areas with small environmental differences as sampling areas can more accurately represent the situations of other grids, minimizing the differences between different grids. When the differences in soil humidity, temperature, and light intensity between different nodes (grids) are smaller, the environmental change increment is smaller. On the contrary, when the differences in soil humidity, temperature, and light intensity between different nodes (grids) are larger, the environmental change increment is larger. The differences in environmental data between each grid and the environmental change increment show a positive correlation.

[0085] According to different planting density classifications (areas without gastrodia elata, sparse planting areas, lush planting areas, and overcrowded planting areas), growers can adjust planting strategies in a timely manner. For example, sending signals to increase planting in areas without gastrodia elata and sparse planting areas can improve land utilization; while sending signals to reduce planting in overcrowded planting areas can effectively prevent resource competition and the occurrence of plant diseases. The thresholds T1 and T2 should be set based on historical data, experimental results, and the actual growth situation of the region. The corresponding distribution characteristics can be found through statistical analysis of planting density data over a period of time, so as to reasonably determine these two thresholds; or research data on the planting density of similar crops can be referred to in agricultural research literature, and the planting density standard suitable for gastrodia elata can be determined through comparative analysis.

[0086] Step 4: Obtain the growth images of gastrodia elata in the current sampling area and input them into the growth monitoring model to obtain the growth data of gastrodia elata in the current sampling area, and analyze the health status of gastrodia elata in the current sampling area based on the growth data;

[0087] In this embodiment, analyzing the health status of gastrodia elata in the current sampling area specifically includes:

[0088] Obtain the growth images of gastrodia elata in the current sampling area, input them into the trained growth monitoring model, obtain the growth data of gastrodia elata in the current sampling area output by the model, generate a health index based on the growth data of gastrodia elata to evaluate the health status of gastrodia elata in the current sampling area, and the calculation formula is:

[0089] H = α1·ln(1 + S)+α2·ln(1 + V)-α3·(e P - 1)

[0090] Where H is the health index, S is the tuber size, V is the tuber volume, P is the proportion of the area with pests and diseases on the tuber, and α1, α2, and α3 are the weight coefficients of each item, 0 < α1 < α2 < α3 < 1, and α1 + α2 + α3 = 1.

[0091] The natural logarithm function is used to represent the influence of tuber size, ensuring that small tuber areas can be handled. For example, when the tuber area is 0, a constant is added to avoid the logarithm being undefined. The logarithmic function can effectively reflect the diminishing marginal utility of tuber size on health. As the tuber size increases, the improvement effect on health per unit area gradually decreases. For example, for a small tuber, assuming its area is 2 cm 2 , and a medium-sized tuber, assuming its area is 10 cm 2 , in this case, the increase from 2 cm 2 to 10 cm 2 may significantly improve the health index because small tubers tend to be nutritionally deficient and have poor growth conditions. However, if we increase from 10 cm 2 to 20 cm 2 , although the tuber is still increasing, since its growth environment is already relatively good, the increased health benefits may not be as obvious as the improvement from a small tuber to a medium-sized tuber. The larger the tuber size, the higher the health index H, showing a proportional relationship. When the tuber area is small, the growth rate of the health index is relatively small. The same is true for the tuber volume. Using the logarithmic function can capture the non-linear relationship between the tuber volume and the health index. Especially when the data shows a gradually increasing effect, it can more realistically reflect the improvement of health. In addition, the logarithmic transformation can reduce the influence of large values on the overall health index, so that larger tubers or volumes will not overly affect the calculation of the health index and maintain a more balanced assessment.

[0092] The higher the health index, the healthier the growth condition of Gastrodia elata, the better the overall growth of the tuber, and the smaller the influence of pests and diseases; on the contrary, a lower health index means that there may be growth problems or a greater influence of pests and diseases. By establishing this health index, researchers and farmers can more intuitively evaluate the growth condition of Gastrodia elata and provide a basis for subsequent management and decision-making.

[0093] The size of the tuber is usually directly related to the growth health of the plant. Larger tubers often indicate good nutrient accumulation in the plant, thus promoting healthy growth. The increased tuber size leads to an increase in the health index. The volume of the tuber is related to the internal tissues and nutrient reserves. An increase in volume means an improvement in the growth potential and survival ability of the plant, which usually corresponds to a better health state. Accordingly, the increased tuber volume also raises the health index. The proportion of the pest and disease area is an important negative indicator reflecting the health of the plant. The larger the pest and disease area, the worse the health condition of the plant, affecting its growth and development and resulting in a decline in the health index. The logarithmic function is used to represent the positive impact of tuber size on the health index. As the tuber size increases, the health index also increases, reflecting a positive promotion effect on plant growth. Similarly, the increase in tuber volume increases the health index in the form of a logarithmic function, reflecting the growth potential of the plant. The proportion of the pest and disease area adopts the form of -α3·(e P -1), showing its negative impact on the health index. As the proportion of pests and diseases increases, the health index decreases, indicating that the health of the plant is threatened.

[0094] Pests and diseases are one of the key factors affecting plant health. The presence of pests and diseases directly damages plant growth and may cause problems such as growth stagnation and a decline in fruit quality. The severity of pests and diseases is often the primary indicator for judging the health condition of plants, so its weight should be relatively high. When the proportion of pests and diseases reaches a certain level, its impact on plant health may increase sharply and even lead to plant death. Therefore, the change in the proportion of pests and diseases has a significant and crucial impact on the health index. The tuber size is still an important manifestation of plant growth. Larger tubers usually mean higher nutrient accumulation and growth potential. However, considering the direct threat of pests and diseases, the weight of tuber size can be appropriately reduced. The tuber volume is closely related to the nutrient reserves and growth potential of the plant. Its importance lies in supporting healthy growth. However, compared with the negative impact of pests and diseases, the weight of tuber volume can still be slightly lower. Generally, the impact of tuber volume on health is more significant than that of tuber size. The tuber volume is the total volume of the tuber, which is directly related to the nutrient storage capacity and growth potential of the plant. A larger volume can improve the overall growth condition of the plant, affecting its stress resistance, nutrient supply, and growth rate. Therefore, it is reasonable to assign it a greater weight. =. The constraint condition of α1 + α2 + α3 = 1 ensures that the health index is a normalized indicator, ensuring that the total contribution of the weighted combination of multiple factors is 100%.

[0095] Step 5: Generate a comprehensive growth index based on the health condition, environmental data, and planting density of Gastrodia elata in the current sampling area within each community set to analyze the growth condition of Gastrodia elata in each grid, and screen out the grids with abnormal growth conditions based on the comprehensive growth index.

[0096] In this embodiment, analyzing the environmental data specifically includes:

[0097] According to the environmental data of each current sampling area, calculate the comprehensive environmental index to evaluate the environment within the soil area for Gastrodia elata growth. The calculation formula is:

[0098]

[0099] where H is the comprehensive environmental index, M is the soil humidity of the current sampling area, T is the temperature of the current sampling area, L is the light intensity of the current sampling area, M max , T max , L max are the maximum suitable values of soil humidity, temperature, and light intensity, and d, e, f are the weight coefficients of soil humidity, temperature, and light intensity, where 0 < f < e < d < 1;

[0100] The maximum suitable value of soil humidity is set to 70% (volume moisture). Excessive humidity may cause root hypoxia, while too low humidity will affect the growth of Gastrodia elata. The suitable growth temperature of Gastrodia elata is generally between 15°C and 25°C. Too high or too low temperature will affect its growth and development. Too high temperature will lead to increased water evaporation and even root burning phenomenon, while too low temperature will cause the growth rate to slow down, the root growth to become slow, and the ability to absorb water and nutrients to decrease, etc. Therefore, the maximum suitable growth temperature is usually set to 20°C. Gastrodia elata is a shade-loving plant, and the suitable light intensity is usually between 2000 and 4000 Lux. Too strong light will cause the soil to dry, and too low light intensity will also affect the growth and development of Gastrodia elata. Therefore, the maximum suitable value of light intensity is set to 3000 Lux.

[0101] In this embodiment, screening out the current sampling areas with abnormal growth conditions specifically includes:

[0102] The formula for calculating the comprehensive growth index of each current sampling area is:

[0103]

[0104] where I is the comprehensive growth index, H is the health index, D is the planting density of the current sampling area, and λ1, λ2, λ3 are the weight coefficients of the health index, comprehensive environmental index, and planting density of Gastrodia elata, where 0 < λ3 ≤ λ2 < λ1 < 1, and λ1 + λ2 + λ3 = 1, and θ is the adjustment coefficient;

[0105] Compare the comprehensive growth index of each current sampling area with a preset growth threshold. If the comprehensive growth index of the current sampling area at the current moment is higher than the preset growth threshold, it indicates that the growth condition of Gastrodia elata in this current sampling area is good, that is, it indicates that all grids in the community set corresponding to this current sampling area are in a good growth state of Gastrodia elata, then normal monitoring is maintained. If the comprehensive growth index of the current sampling area at the current moment is lower than the preset growth threshold, it indicates that the growth state of Gastrodia elata in this current sampling area is poor, that is, it indicates that all grids in the community set corresponding to this current sampling area are in a poor growth state of Gastrodia elata, then all grids in the community set of this current sampling area are marked as grids with abnormal growth, and a warning of poor growth condition is sent. Moreover, for the grids marked as having abnormal growth, the comprehensive environmental index of their adjacent grids is also judged. If the comprehensive environmental index of an adjacent grid is lower than the preset environmental threshold, then this adjacent grid is also marked as abnormal and a warning of poor growth condition is sent.

[0106] The health index is in exponential form, which can emphasize the influence of the health index on the comprehensive growth index. If the health condition of the plant is good and the value of the health index H is high, then will increase significantly, reflecting the key role of the health state in the growth process. The comprehensive environmental index uses a logarithmic function to smooth the influence of environmental data and reduce the interference of extreme values on the comprehensive growth index. The characteristic of the logarithmic function is that as the environmental conditions improve, the influence will show a decreasing effect. The highlight is that it can ensure that even when the environmental data E is 0, the index will not be undefined. In addition, adding λ2 can serve as a weight coefficient, facilitating flexible adjustment under different environmental conditions. The form of can effectively control the influence of planting density, ensuring that its proportion in the growth index is higher at low density and tends to saturate at high density. This form makes it so that even when the planting density increases to a relatively high level, the growth rate of the comprehensive growth index will gradually decrease, reflecting the possible resource competition and pressure caused by excessive density. The value range of the adjustment coefficient θ is [0,1]. A smaller value, such as 0.1 or 0.2, indicates that the influence of planting density on the growth index is small, while a larger value (close to 1) indicates that the influence of planting density on growth is significant. Here, a smaller value can be set to reflect that the influence of planting density on this comprehensive growth index is relatively small.

[0107] The health index H is usually the most important indicator reflecting the plant condition. Healthy plants play a central role in growth and physiological activities. Therefore, the weight λ1 should be set to the maximum to reflect its importance. Environmental conditions (such as soil humidity, temperature, light, etc.) directly affect the growth and health of plants, but their impact on planting density may be more direct and obvious. Changes in environmental conditions will cause significant changes in the health status of plants. Therefore, λ2 can be set to be greater than or equal to λ3. The planting density reflects the number of Gastrodia elata tubers in the area. Excessive planting density will make it difficult to meet the nutrients required by the Gastrodia elata tubers in this area. Compared with the health index and environmental data, its impact is relatively small. Therefore, its weight can also be set relatively small. Therefore, θ>1 here to reflect that the higher the planting density, the lower the comprehensive impact index, and vice versa, the lower the planting density, the higher the comprehensive impact index.

[0108] By comparing the comprehensive growth index with a preset growth threshold, grids with abnormal growth states can be quickly identified. When the comprehensive growth index is lower than the threshold, the system can issue an early warning in a timely manner to ensure that growers can quickly take measures, such as adjusting irrigation, fertilization, or improving other environmental conditions. This timely response is crucial for ensuring the healthy growth of Gastrodia elata. After an abnormal mark is made on a certain grid, the system will also judge the comprehensive environmental index of adjacent grids. This linkage monitoring mechanism can effectively prevent the spread of potential diseases or adverse growth conditions and ensure the healthy management of the entire planting area. In addition, by paying attention to adjacent grids, management strategies can be adjusted in a timely manner to avoid large-area losses. By establishing the relationship between the comprehensive growth index and the environment and plant physiology, growers can make decisions based on scientific data rather than relying on traditional experience. This flexible management method helps to continuously improve the production process and adapt to different growth conditions and market demands.

[0109] Please refer to Figure 2 , the present invention further provides a Gastrodia elata growth monitoring system based on image recognition. The Gastrodia elata growth monitoring system based on image recognition is used to implement the above-mentioned Gastrodia elata growth monitoring method based on image recognition, including:

[0110] A data acquisition module, used to divide the soil area where Gastrodia elata grows into several grids and collect the environmental data of each grid at the current moment. The environmental data includes soil humidity, temperature, and light intensity;

[0111] A model construction module, used to obtain multiple Gastrodia elata growth images with known growth data, use the Gastrodia elata growth images as inputs and the corresponding growth data as outputs, and construct and train a growth monitoring model based on a convolutional neural network;

[0112] The planting density analysis module is used to divide all grids into multiple community sets according to the environmental data of each grid, select a current sampling area within each community set, and determine the planting density of the current sampling area based on the number of gastrodia tuber and the sampling area within the current sampling area;

[0113] The real-time monitoring module is used to obtain the growth images of gastrodia within the current sampling area and input them into the growth monitoring model to obtain the growth data of gastrodia within the current sampling area, and analyze the health status of gastrodia within the current sampling area based on the growth data;

[0114] The comprehensive analysis module is used to generate a comprehensive growth index based on the health status, environmental data, and planting density of gastrodia in the current sampling area within each community set to analyze the growth status of gastrodia in each grid, and screen out the grids with abnormal growth status based on the comprehensive growth index.

[0115] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data and performing software simulation to obtain a formula closest to the actual situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0116] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0117] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, and may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0118] As described above, only the specific implementation manners of this application are provided, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application.

Claims

1. A method for monitoring the growth of Gastrodia elata based on image recognition, characterized in that, The specific steps include: Step 1: Divide the soil area where Gastrodia elata grows into several grids, and collect the environmental data of each grid at the current moment. The environmental data includes soil humidity, temperature, and light intensity. Step 2: Obtain multiple Gastrodia elata growth images with known growth data. Using the Gastrodia elata growth images as inputs and the corresponding growth data as outputs, construct and train a growth monitoring model based on a convolutional neural network. Step 3: Divide all grids into multiple community sets according to the environmental data of each grid. Select a current sampling area within each community set, and determine the planting density of the current sampling area based on the number of tubers and the sampling area of Gastrodia elata within the current sampling area. Step 4: Obtain the growth image of Gastrodia elata within the current sampling area, and input it into the growth monitoring model to obtain the growth data of Gastrodia elata within the current sampling area, and analyze the health status of Gastrodia elata within the current sampling area based on the growth data. Step 5: Generate a comprehensive growth index according to the health status, environmental data, and planting density of Gastrodia elata in the current sampling area within each community set to analyze the growth status of Gastrodia elata in each grid, and screen out the grids with abnormal growth status based on the comprehensive growth index.

2. The method for monitoring the growth of Gastrodia elata based on image recognition according to claim 1, characterized in that, Specifically, collecting the environmental data of each grid at the current moment includes: Evenly divide the planting soil area of Gastrodia elata into multiple grids, and arrange soil humidity sensors, temperature sensors, and light sensors at the grid centers to collect the environmental data of the Gastrodia elata growth area.

3. A Gastrodia elata growth monitoring method based on image recognition according to claim 1, characterized in that, Specifically, constructing and training the growth monitoring model includes: Obtain multiple Gastrodia elata growth images with known growth data. Each group of Gastrodia elata growth images includes images of Gastrodia elata taken from various angles. The growth data includes tuber surface area, tuber volume, and the proportion of the area with pests and diseases on the tuber. Take the average value of the tuber surface areas of each group of Gastrodia elata growth images as the tuber surface area of the corresponding Gastrodia elata. The proportion refers to the ratio of the surface area of the tuber with pests and diseases to the tuber surface area. The formula for calculating the proportion of the area with pests and diseases on the tuber is: Among them, P is the proportion of the area with pests and diseases on the tuber, N d is the surface area of the area with pests and diseases in the tuber, N t is the surface area of the tuber; Label the growth data for each group of Gastrodia elata growth images. Perform preprocessing on each labeled group of Gastrodia elata growth images. The preprocessing includes removing blurred and duplicate images, randomly rotating, flipping, cropping, and adjusting the brightness of the images, and finally unifying the image size. Using the Gastrodia elata growth images after labeling and preprocessing as inputs, select a convolutional neural network as the basic model, and the corresponding growth data as labels, train the convolutional neural network model to construct a growth monitoring model.

4. The method for monitoring the growth of Gastrodia elata based on image recognition according to claim 1, characterized in that, Specifically, determining the planting density of the grid includes: Take each grid as a node, and use the environmental data corresponding to each node as the attributes of the node to construct an undirected graph. If two nodes are adjacent, add an edge in the undirected graph to form an undirected graph network. Place each node in an independent community. For each node in the undirected graph network, move it successively to the communities where all its neighbor nodes are located, and calculate the environmental change increment corresponding to each move. Specifically, it includes: First, calculate the average value of the environmental data between a node and any of its neighbor nodes, and then analyze the change between the node and the average value to obtain the environmental change increment. The calculation formula is: Among them, △E(ij) is the environmental change increment between the i-th node and the j-th node, is the average soil humidity between the i-th node and the j-th node, M(i) is the soil humidity of the i-th node, is the average temperature between the i-th node and the j-th node, T(i) is the temperature of the i-th node, is the average light intensity between the i-th node and the j-th node, L(i) is the light intensity of the i-th node, and both i and j are nodes in the undirected graph, and i and j are neighbor nodes; Move the node to the neighboring community with the minimum added value of environmental change. Repeat moving the nodes until the community affiliations of all nodes no longer change. At this time, randomly select a node in each community as the current sampling area corresponding to all nodes in the community set, and obtain the planting density of the current sampling area. The formula for calculating the planting density is as follows: Among them, D is the planting density of the current sampling area, n is the number of tubers in the current sampling area, and A is the area of the current sampling area; Regard the planting density of the current sampling area as the planting density of the community set where the current sampling area is located. The growth area of Gastrodia elata can be classified according to the planting density: if D = 0, it is judged as an area without Gastrodia elata; if 0 < D < T1, it is judged as a sparse planting area; if T1 ≤ D < T2, it is judged as a lush planting area; if D ≥ T2, it is judged as an overcrowded planting area. When the growth area of Gastrodia elata is in an area without Gastrodia elata and a sparse planting area, send a signal to the planter to increase the planting of Gastrodia elata in this grid. When the growth area of Gastrodia elata is in an overcrowded planting area, send a signal to the planter to reduce the planting of Gastrodia elata in this grid. T1 is the sparse threshold, and T is the dense threshold.

5. A Gastrodia elata growth monitoring method based on image recognition according to claim 1, characterized in that, Analyzing the health status of Gastrodia elata in the current sampling area specifically includes: Obtain the growth image of Gastrodia elata in the current sampling area, input it into the trained growth monitoring model, obtain the growth data of Gastrodia elata in the current sampling area output by the model, and generate a health index based on the growth data of Gastrodia elata to evaluate the health status of Gastrodia elata in the current sampling area. The calculation formula is as follows: H = α1·ln(1 + S) + α2·ln(1 + V) - α3·(e P - 1) Among them, H is the health index, S is the tuber size, V is the tuber volume, P is the proportion of the area with pests and diseases on the tuber, and α1, α2, and α3 are the weight coefficients of each item, 0 < α1 < α2 < α3 < 1, and α1 + α2 + α3 = 1; regard the health index of Gastrodia elata in the current sampling area as the health index of the grid where the current sampling area is located.

6. The method for monitoring the growth of Gastrodia elata based on image recognition according to claim 1, wherein, Analyzing the environmental data specifically includes: According to the environmental data of each current sampling area, calculate the environmental comprehensive index to evaluate the environment in the soil area where Gastrodia elata grows. The calculation formula is as follows: Among them, E is the comprehensive environmental index, M is the soil humidity of the current sampling area, T is the temperature of the current sampling area, L is the light intensity of the current sampling area, M max , T max , L max are the maximum suitable values of soil humidity, temperature, and light intensity, and d, e, and f are the weight coefficients of soil humidity, temperature, and light intensity, where 0 < f < e < d < 1.

7. The method for monitoring the growth of Gastrodia elata based on image recognition according to claim 6, wherein, Screening out the current sampling areas with abnormal growth conditions specifically includes: The formula for calculating the comprehensive growth index of each current sampling area is: Among them, I is the comprehensive growth index, H is the health index, D is the planting density of the current sampling area, and λ1, λ2, and λ3 are the weight coefficients of the health index, environmental comprehensive index, and planting density of Gastrodia elata, 0 < λ3 ≤ λ2 < λ1 < 1, and λ1 + λ2 + λ3 = 1, and θ is the adjustment coefficient; Compare the comprehensive growth index of each current sampling area with a preset growth threshold. If the comprehensive growth index of the current sampling area at the current moment is higher than the preset growth threshold, it indicates that the growth condition of Gastrodia elata in this current sampling area is good, that is, it indicates that all grids within the community set corresponding to this current sampling area are in a good growth state of Gastrodia elata, then normal monitoring is maintained. If the comprehensive growth index of the current sampling area at the current moment is lower than the preset growth threshold, it indicates that the growth condition of Gastrodia elata in this current sampling area is poor, that is, it indicates that all grids within the community set corresponding to this current sampling area are in a poor growth state of Gastrodia elata, then all grids within the community set of this current sampling area are marked as grids with abnormal growth, and a warning of poor growth condition is sent. And for the grids marked with abnormal growth, the comprehensive environmental index of their adjacent grids is also judged. If the comprehensive environmental index of an adjacent grid is lower than the preset environmental threshold, then this adjacent grid is also marked as abnormal and a warning of poor growth condition is sent.

8. A Gastrodia elata growth monitoring system based on image recognition, characterized in that, The described Gastrodia elata growth monitoring system based on image recognition is used to implement the Gastrodia elata growth monitoring method according to any one of claims 1 - 7, and includes: A data acquisition module, which is used to divide the soil area where Gastrodia elata grows into several grids and collect the environmental data of each grid at the current moment. The environmental data includes soil humidity, temperature, and light intensity; A model construction module, which is used to obtain multiple Gastrodia elata growth images with known growth data, use the Gastrodia elata growth images as inputs and the corresponding growth data as outputs, and construct and train a growth monitoring model based on a convolutional neural network; A planting density analysis module, which is used to divide all grids into multiple community sets according to the environmental data of each grid, select a current sampling area within each community set, and determine the planting density of the current sampling area based on the number of tubers of Gastrodia elata and the sampling area in the current sampling area; A real-time monitoring module, which is used to obtain the growth image of Gastrodia elata in the current sampling area and input it into the growth monitoring model to obtain the growth data of Gastrodia elata in the current sampling area, and analyze the health condition of Gastrodia elata in the current sampling area based on the growth data; A comprehensive analysis module, which is used to generate a comprehensive growth index according to the health condition, environmental data, and planting density of Gastrodia elata in the current sampling area within each community set to analyze the growth condition of Gastrodia elata in each grid, and screen out the grids with abnormal growth conditions based on the comprehensive growth index.

Images