Shrimp and rice symbiosis visual monitoring method based on computer vision

Abstract

The invention discloses a shrimp and rice symbiosis visual monitoring method based on computer vision, and relates to the technical field of intelligent agriculture. The problems that the prior art mainly depends on satellite remote sensing data, multi-dimensional data such as terrain, biomass and water quality lacks effective association, and the aspects such as real-time early warning response and resource optimization configuration have remarkable defects are solved. According to the method, multi-modal data of unmanned aerial vehicle aerial photography, underwater laser scanning, water quality sensor and the like are integrated, a rice biomass estimation, crayfish behavior recognition and water quality prediction model is established, and a differential early warning system is constructed in combination with terrain complexity grading, so that early recognition of risks of crayfish stress behaviors, water quality mutation and the like is realized; the unmanned aerial vehicle and the fixed camera work cooperatively, dynamic cruise and high-definition collection of an abnormal area are achieved, and the intelligent management level of the shrimp and rice symbiotic system is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of smart agriculture, and particularly to a visual monitoring method for shrimp-rice symbiosis based on computer vision. Background Technique

[0002] As an efficient ecological agriculture model, shrimp-rice symbiosis realizes the recycling of resources and the improvement of economic benefits through the coordinated development of rice planting and crayfish farming. The Chinese patent application with the publication number CN115170977A discloses a visual monitoring method and system for shrimp-rice symbiosis based on multi-spectral images, including: image acquisition and preprocessing, acquiring images with a cloud cover of less than 30% through multi-spectral satellites, and performing masking processing on the acquired images; feature calculation and time series analysis, selecting the normalized difference vegetation index as the discrimination index between the shrimp-rice symbiosis area and other ground objects, and fitting through the time series harmonic analysis method to remove pollution points; selecting typical phases for RGB image synthesis, highlighting the color of the shrimp-rice symbiosis area in the multi-spectral image through the time series fitting result and typical phase characteristics; monitoring the prominent color of the shrimp-rice symbiosis area in the multi-spectral image. It solves the problem that it is difficult to accurately distinguish the existing shrimp-rice symbiosis area from other surrounding areas.

[0003] However, the current shrimp-rice symbiosis system still has the following problems:

[0004] The existing technology mainly relies on satellite remote sensing data, and it is difficult to comprehensively obtain the spatio-temporal change information of the rice growth status, crayfish behavior and water quality environment; moreover, there is a lack of effective association of multi-dimensional data such as terrain, biomass, and water quality, and the coupling mechanism inside the system cannot be revealed, and there are significant deficiencies in real-time early warning response and resource optimization allocation. Summary of the Invention

[0005] The purpose of the present invention is to provide a visual monitoring method for shrimp-rice symbiosis based on computer vision, which realizes the comprehensive monitoring and precise management of the shrimp-rice symbiosis system through multi-source data fusion and intelligent analysis technology, so as to solve the problems raised in the above background technique.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A visual monitoring method for shrimp-rice symbiosis based on computer vision, including:

[0008] Based on an image acquisition device, the growth image data of shrimp and rice in the monitoring area is collected in real time, and target feature recognition is performed based on the growth image data, and the growth status of shrimp and rice is respectively judged based on the recognition results;

[0009] Build a water quality parameter prediction model, evaluate the water quality indicators in the monitoring area, and establish a growth model of the shrimp-rice symbiotic system by combining the statistical analysis of the identification results. Predict the symbiotic trend of shrimp and rice in the future for a period of time based on the growth model of the shrimp-rice symbiotic system.

[0010] Furthermore, the monitoring area specifically includes:

[0011] Monitoring area division: According to the actual terrain and area of the shrimp-rice symbiotic fields, divide into no less than four monitoring sub-areas, number each monitoring sub-area, and at the same time, determine the boundary coordinates of each monitoring sub-area;

[0012] Monitoring equipment installation: Determine the equipment type and quantity for each monitoring sub-area according to the location characteristics of each monitoring sub-area, including determining the installation location and quantity of cameras and the installation location and quantity of lighting equipment, and the quantity and location of the deployed underwater laser scanning arrays;

[0013] Data network construction: Based on the wireless network, establish a data transmission channel for the monitoring equipment in each monitoring sub-area, build a monitoring sensor network, and perform data sharing and real-time data transmission based on the monitoring sensor network.

[0014] Furthermore, the growth image data also includes:

[0015] Collect by low-altitude flight of an unmanned aerial vehicle (UAV). The monitoring equipment carried by the UAV performs image collection according to a predetermined shooting frequency to obtain remote sensing image data of the paddy field;

[0016] When the fixed camera detects abnormal growth in a local area, the UAV generates a cruise path based on the boundary coordinates of the corresponding monitoring area, and at the same time, adjusts the shooting frequency and camera resolution;

[0017] Integrate the aerial image data obtained by the UAV and the monitoring image data obtained by the monitoring equipment to generate the growth image data of shrimp and rice in the monitoring area.

[0018] Furthermore, the target feature recognition specifically includes:

[0019] Extract the image data of different growth stages of rice from the growth image data, and extract the color features, texture features and shape features of the rice;

[0020] Establish the relationship between the color features, texture features and shape features of rice and biomass, estimate the rice biomass, and obtain the biomass value corresponding to each monitoring area;

[0021] Among them, the model expression of the rice biomass estimation is:

[0022] y = β0 + β1x1 + β2x2 +... + βm x m +α

[0023] where y is the biomass, and x1, x2,..., x m are rice characteristics, β0, β1,..., β m are model coefficients, and α is the error term;

[0024] Obtain the underwater three-dimensional point cloud data of the monitoring area through the deployed underwater laser scanning array;

[0025] Generate a three-dimensional model of the crayfish habitat based on the underwater three-dimensional point cloud data, and determine the number of crayfish per unit area by combining the monitoring image data obtained by the monitoring equipment in each monitoring sub-area.

[0026] Furthermore, determining the number of crayfish per unit area specifically includes:

[0027] Preprocess the obtained underwater three-dimensional point cloud data, perform surface reconstruction on the preprocessed three-dimensional point cloud data, convert the discrete point cloud data into a continuous surface model, and obtain the three-dimensional model of the crayfish habitat;

[0028] Preprocess the obtained monitoring image data, and perform crayfish target detection on the preprocessed video images;

[0029] Perform spatial registration on the three-dimensional model of the crayfish habitat and the monitoring image data, determine the three-dimensional space area corresponding to each video image, count the number of crayfish detected in the area, and determine the number of crayfish per unit area in combination with the area of the area.

[0030] Furthermore, counting the number of crayfish detected in the area also includes:

[0031] Extract the biological characteristics of each crayfish from the monitoring image data and establish a unique identifier for each crayfish;

[0032] Establish a time axis according to the data length of the collected monitoring image data, perform frame-by-frame processing on the monitoring image data, determine the initial position corresponding to each crayfish, determine the actual position of each initial position in the three-dimensional space area, and analyze the position characteristics of the crayfish;

[0033] Track the moving positions corresponding to each crayfish in consecutive frames, establish the moving characteristics corresponding to the crayfish in combination with the initial positions. At the same time, extract the data ratio of each crayfish in consecutive frames and generate the pause characteristics of each crayfish;

[0034] Based on the moving characteristics of each crayfish combined with the corresponding pause characteristics, obtain the behavior characteristics corresponding to each crayfish, and determine whether the crayfish is in a stress behavior state based on the behavior characteristics;

[0035] Among them, whether the crayfish is in a stress behavior state is determined by a comprehensive stress behavior determination index, and the formula is as follows:

[0036] S = ω1×v avg + ω2×a max + ω3×Δθ avg

[0037] Among them, ω1, ω2, and ω3 are weight coefficients, and ω1 + ω2 + ω3 = 1. v avg is the average speed of the crayfish, a max is the maximum acceleration, and Δθ avg is the average change amount of the attitude angle within a certain period of time;

[0038] When the comprehensive stress behavior determination index exceeds the preset threshold, it is determined that the crayfish is in a stress behavior state.

[0039] Furthermore, the target feature recognition also includes:

[0040] Based on the aerial image data obtained by the drone, obtain the basic terrain of the monitoring area, measure the number of geometric units required to cover the terrain at different scales, quantify the roughness of the terrain surface, and extract hydrological features;

[0041] Based on the basic terrain and hydrological features of the monitoring area, determine the weights of the terrain factors and generate a comprehensive terrain complexity index;

[0042] Based on the distribution characteristics of the comprehensive terrain complexity index, divide the monitoring area into four levels, including extremely low complexity, low complexity, medium complexity, and high complexity.

[0043] Furthermore, based on the classification results of the monitoring area, perform spatial matching and time series analysis on the terrain complexity data, biomass, and water quality parameters to determine the association rules between terrain complexity and biomass fluctuations and water quality mutations, and construct a differential monitoring strategy.

[0044] Furthermore, establishing a growth model of the shrimp-rice symbiotic system also includes: based on four typical scenarios of normal growth, extreme weather, pest outbreaks, and management interventions, simulate the shrimp-rice symbiotic responses under different scenarios, and based on three-dimensional dynamic visualization technology, display the simulation results in real time, including biomass change curves, crayfish distribution heat maps, and water quality parameter fluctuation cloud maps.

[0045] Furthermore, predicting the symbiotic trend of shrimp and rice in the future also includes: calculating the health index of the monitoring area based on the terrain complexity data, rice biomass value, and crayfish density data, correlating with the water quality parameters output by the water quality parameter prediction model, and quantitatively evaluating the symbiotic effect intensity of different monitoring areas;

[0046] Among them, the symbiotic effect intensity index is obtained through the following formula:

[0047]

[0048] Among them, P is the symbiotic effect intensity index, B i is the rice biomass value, W i is the water quality index, A i is the crayfish density data, T i is the comprehensive index of terrain complexity, ω i is the weight coefficient, and ∈ is the smoothing factor.

[0049] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0050] By integrating multi-modal data such as UAV aerial photography, underwater laser scanning, and water quality sensors, the limitations of traditional single-index monitoring are broken through, rice biomass estimation, crayfish behavior recognition, and water quality prediction models are established. Combining with the terrain complexity level division, a differentiated early warning system is constructed to realize the early identification of risks such as crayfish stress behavior and water quality mutation, simulate the system response under different scenarios, generate three-dimensional visualization results, and provide a scientific basis for dynamically adjusting management measures. The UAV and fixed cameras cooperate to realize dynamic cruising and high-definition acquisition of abnormal areas, ensuring no dead ends in monitoring, and effectively improving the intelligent management level of the shrimp-rice symbiotic system. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a flowchart of the visual monitoring method for shrimp-rice symbiosis based on computer vision of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0053] To solve the technical problems that the prior art mainly relies on satellite remote sensing data, and there are significant deficiencies in the effective correlation of multi-dimensional data such as terrain, biomass, and water quality, and real-time early warning response and resource optimization allocation, please refer to Figure 1 , the following technical solutions are provided in this embodiment:

[0054] A visual monitoring method for shrimp-rice symbiosis based on computer vision, including:

[0055] Based on the image acquisition device, the growth image data of shrimp and rice in the monitoring area is collected in real time, and target feature recognition is carried out based on the growth image data. Based on the recognition results, the growth conditions of shrimp and rice are judged respectively;

[0056] A water quality parameter prediction model is established to evaluate the water quality indicators in the monitoring area. Combining the statistical analysis of the recognition results, a growth model of the shrimp-rice symbiotic system is established. Based on four typical scenarios of normal growth, extreme weather, pest and disease outbreaks, and management interventions, the responses of the shrimp-rice symbiotic system under different scenarios are simulated. Based on three-dimensional dynamic visualization technology, the simulation results are displayed in real time, including the biomass change curve, the heat map of crayfish distribution, and the cloud map of water quality parameter fluctuations;

[0057] Based on the growth model of the shrimp-rice symbiotic system, predict the symbiotic trend of shrimp and rice in the future for a period of time;

[0058] In this embodiment, predicting the symbiotic trend of shrimp and rice in the future for a period of time further includes: calculating the health index of the monitoring area based on the terrain complexity data, rice biomass value, and crayfish density data, associating it with the water quality parameters output by the water quality parameter prediction model, and quantitatively evaluating the symbiotic effect intensity of different monitoring areas;

[0059] Among them, the symbiotic effect intensity index is obtained through the following formula:

[0060]

[0061] Among them, P is the symbiotic effect intensity index, B i is the rice biomass value, W i is the water quality index, A i is the crayfish density data, T i is the comprehensive terrain complexity index, ω i is the weight coefficient, and ∈ is the smoothing factor.

[0062] In this embodiment, by integrating multi-dimensional data such as terrain complexity, biomass, water quality, and crayfish density, the comprehensive dynamic monitoring of the shrimp-rice symbiotic system is realized, breaking through the limitations of traditional single-index monitoring. A rice biomass estimation model, a crayfish behavior recognition model, and a water quality parameter prediction model are established. Combining the terrain complexity level division, a differential early warning threshold system is constructed. Through the correlation analysis of the coupling relationship between terrain, biology, and environment, the early recognition and accurate early warning of risks such as crayfish stress behavior and water quality mutation are realized, and three-dimensional visual biomass change curves, crayfish distribution heat maps, and water quality fluctuation cloud maps are generated, providing data support for scientifically formulating management measures such as feeding strategies and water change frequencies.

[0063] In this embodiment, the monitoring area specifically includes:

[0064] Monitoring area division: Based on the actual terrain and area of the shrimp-rice symbiotic fields, divide into no less than four monitoring sub-areas, number each monitoring sub-area, and at the same time, determine the boundary coordinates of each monitoring sub-area to ensure that the images collected subsequently can be accurately corresponding to specific positions;

[0065] Monitoring equipment installation: Determine the types and quantities of equipment in each monitoring sub-area according to the location characteristics of each monitoring sub-area, including determining the installation positions and quantities of cameras and the installation positions and quantities of lighting equipment, and the quantity and position of the deployed underwater laser scanning arrays. For example, install panoramic cameras at high positions on the edge of the field for overall monitoring, install close-up cameras inside the paddy field near the active areas of crayfish and where the rice is dense. The lighting equipment can be automatically turned on or off according to the change of ambient light to ensure clear images can be obtained at night or in dim light;

[0066] Data network construction: Establish a data transmission channel for the monitoring equipment in each monitoring sub-area based on the wireless network, build a monitoring sensor network, and conduct data sharing and real-time data transmission based on the monitoring sensor network.

[0067] In this embodiment, the growth image data further includes:

[0068] Collect by low-altitude flight of an unmanned aerial vehicle (UAV). The monitoring equipment carried by the UAV performs image collection according to a predetermined shooting frequency to obtain remote sensing image data of the paddy field;

[0069] When the fixed camera detects abnormal growth in a local area, the UAV generates a cruise path based on the boundary coordinates of the corresponding monitoring area, and at the same time, adjusts the shooting frequency and camera resolution. For example, dynamically adjust according to the abnormal level (1-5 levels): Resolution: 5 cm / pixel (conventional) → 0.8 cm / pixel (highest), Frame rate: 2 fps (conventional) → 15 fps (highest), Flight altitude: 50 m (conventional) → 10 m (emergency);

[0070] Integrate the aerial image data obtained by the UAV and the monitoring image data obtained by the monitoring equipment to generate the growth image data of shrimp and rice in the monitoring area.

[0071] In this embodiment, by finely dividing the monitoring areas and installing equipment in the shrimp-rice symbiotic fields, an efficient and comprehensive monitoring system is constructed. Combining with the wireless network to achieve real-time data transmission and sharing, it ensures accurate positioning and real-time response of the monitoring. Moreover, remote sensing images are obtained by drones, providing macroscopic images of the paddy fields and supplementing the information of areas that cannot be monitored by fixed cameras. The drones can dynamically adjust the shooting parameters to provide detailed images with high resolution and high frame rate, thus enhancing the emergency handling ability. It not only improves the intelligent level of agricultural management but also provides a scientific basis for agricultural production, achieving the goal of precision agricultural management and significantly enhancing the efficiency and quality of agricultural production.

[0072] In this embodiment, the target feature recognition specifically includes:

[0073] Based on the growth image data, extract the image data of different growth stages of rice, and extract the color features (such as statistics such as mean and standard deviation in the RGB color space), texture features (such as contrast, correlation, energy, and homogeneity), and shape features of rice to reflect the morphology and growth status of rice plants;

[0074] Establish the relationships between the color features, texture features, and shape features of rice and biomass, estimate the biomass of rice, and obtain the biomass values corresponding to each monitoring area;

[0075] Among them, the model expression for the rice biomass estimation is:

[0076] y = β0 + β1x1 + β2x2 +... + β m x m + α

[0077] Among them, y is the biomass, x1, x2,..., x m are the rice characteristics, β0, β1,..., β m are the model coefficients, and α is the error term; solve the model coefficients by methods such as the least squares method, so as to obtain the biomass values of each monitoring area, and analyze the distribution law and change trend of biomass by calculating statistical parameters such as the average value and standard deviation of biomass;

[0078] Obtain the underwater three-dimensional point cloud data of the monitoring area through the deployed underwater laser scanning array;

[0079] Generate a three-dimensional model of the crayfish habitat based on the underwater three-dimensional point cloud data, determine the number of crayfish per unit area in combination with the monitoring image data obtained by the monitoring equipment of each monitoring sub-area, and distinguish the normal activity and stress behavior states through the behavior pattern recognition algorithm.

[0080] In this embodiment, the morphological and growth conditions of rice plants are reflected by extracting the color, texture, and shape features of rice, and then the relationship between these features and biomass is established to estimate the rice biomass, improving the accuracy and precision of crop growth monitoring and promoting the scientific and refined management of farmland;

[0081] In this embodiment, determining the number of crayfish per unit area specifically includes:

[0082] Preprocess the acquired underwater three-dimensional point cloud data, perform surface reconstruction on the preprocessed three-dimensional point cloud data, convert the discrete point cloud data into a continuous surface model, and obtain the three-dimensional model of the crayfish habitat;

[0083] Preprocess the acquired monitoring image data, such as performing grayscale conversion, histogram equalization, etc., to enhance the image contrast and clarity for target detection, and perform crayfish target detection on the preprocessed video images;

[0084] Register the three-dimensional model of the crayfish habitat with the monitoring image data, determine the three-dimensional spatial area corresponding to each video image, count the number of crayfish detected in the area, and determine the number of crayfish per unit area in combination with the area of the area.

[0085] In this embodiment, counting the number of crayfish detected in the area further includes:

[0086] Extract the biological characteristics of each crayfish from the monitoring image data and establish a unique identifier for each crayfish;

[0087] Establish a time axis based on the data length of the acquired monitoring image data, perform frame processing on the monitoring image data, determine the initial position corresponding to each crayfish, determine the actual position of each initial position in the three-dimensional spatial area, and analyze the position characteristics of the crayfish;

[0088] Track the moving positions corresponding to each crayfish in consecutive frames, establish the moving characteristics corresponding to the crayfish in combination with the initial positions. At the same time, extract the data ratio of each crayfish in consecutive frames to generate the pause characteristics of each crayfish;

[0089] Based on the moving characteristics of each crayfish combined with the corresponding pause characteristics, obtain the behavioral characteristics corresponding to each crayfish, and determine whether the crayfish is in a stress behavior state based on the behavioral characteristics to achieve accurate tracking of the individual trajectories of crayfish;

[0090] Among them, whether the crayfish is in a stress behavior state is determined by a comprehensive stress behavior determination index, and the formula is as follows:

[0091] S = ω1×v avg +ω2×a max+ω3×Δθ avg

[0092] where ω1, ω2, and ω3 are weight coefficients, and ω1 + ω2 + ω3 = 1, v avg is the average speed of crayfish, a max is the maximum acceleration, and Δθ avg is the average change in the attitude angle within a certain period of time. When the comprehensive judgment index of the stress behavior exceeds the preset threshold, it is determined that the crayfish is in a stress behavior state.

[0093] In this embodiment, by preprocessing the underwater three-dimensional point cloud data and performing surface reconstruction, combined with the preprocessing of the monitoring images and the crayfish target detection, the accurate estimation of the number of crayfish per unit area is achieved; at the same time, by extracting the biological characteristics of the crayfish, establishing unique identifiers, analyzing the position and movement characteristics, and generating pause characteristics, the behavior patterns of the crayfish can be finely analyzed and the stress behavior state can be intelligently determined, thereby providing a scientific basis for crayfish farming, timely discovering abnormal behaviors, reducing losses, improving the farming efficiency, and overall enhancing the intelligent level and production efficiency of the farming management.

[0094] In this embodiment, the target feature recognition further includes:

[0095] Obtaining the basic terrain of the monitoring area based on the aerial image data obtained by the drone, including slope, aspect, undulation degree, etc., measuring the number of geometric units required to cover the terrain at different scales, quantifying the roughness of the terrain surface, and extracting hydrological features;

[0096] Determining the weights of terrain factors such as slope, aspect, and terrain undulation degree extracted based on DEM data (slope: 0.35, undulation degree: 0.25, roughness: 0.20, other factors: 0.20) based on the basic terrain and hydrological features of the monitoring area, and generating a comprehensive terrain complexity index;

[0097] Dividing the monitoring area into four levels based on the distribution characteristics of the comprehensive terrain complexity index, including extremely low complexity, low complexity, medium complexity, and high complexity.

[0098] Based on the classification results of the monitoring area, spatial matching and time series analysis are carried out on the terrain complexity data, biomass, and water quality parameters to determine the association rules between terrain complexity and biomass fluctuations, water quality mutations. For example, a strong association rule of "high terrain complexity + low dissolved oxygen → the probability of crayfish stress behavior increases by 37%" is discovered, and a differentiated monitoring strategy is constructed. In high-biomass areas: deploy high-density visual sensor nodes and implement high-frequency monitoring every 10 minutes. In complex terrain areas: configure multispectral imagers and micro-meteorological stations to strengthen the collaborative monitoring of environmental parameters, predict high-risk periods based on historical data, intelligently allocate monitoring resources, construct a terrain complexity - warning threshold mapping table, and automatically adjust the warning thresholds of water quality, biomass, and other indicators according to the terrain complexity level. More stringent threshold standards are adopted in complex terrain areas, and appropriately relaxed in simple areas.

[0099] In this embodiment, the basic terrain information of the monitoring area is obtained through UAV aerial image data, the roughness and hydrological characteristics of the terrain surface are quantified, and then a comprehensive terrain complexity index is generated to divide the monitoring area into different levels. By combining the spatial matching and time series analysis of terrain complexity with biomass and water quality parameters, the association rules between terrain complexity and biomass fluctuations, water quality mutations are determined, and a differentiated monitoring strategy is constructed accordingly, realizing the intelligent allocation of monitoring resources and the precise control of early warning, effectively improving the pertinence and efficiency of monitoring, reducing the breeding risk, and enhancing the intelligent level of agricultural management.

[0100] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A computer vision-based visual monitoring method for shrimp-rice symbiosis, characterized in that, Including: Based on the image acquisition device, the growth image data of shrimp and rice in the monitoring area is collected in real time, and target feature recognition is carried out based on the growth image data. Based on the recognition results, the growth conditions of shrimp and rice are judged respectively; A water quality parameter prediction model is established to evaluate the water quality indicators in the monitoring area. Combining the statistical analysis of the recognition results, a growth model of the shrimp-rice symbiotic system is established, and the symbiotic trend of shrimp and rice in the future period is predicted based on the growth model of the shrimp-rice symbiotic system.

2. The computer vision-based visual monitoring method for shrimp-rice symbiosis according to claim 1, characterized in that, The monitoring area specifically includes: Monitoring area division: According to the actual terrain and area of the shrimp-rice symbiotic field, at least four monitoring sub-areas are divided, and each monitoring sub-area is numbered. At the same time, the boundary coordinates of each monitoring sub-area are determined; Monitoring equipment installation: According to the location characteristics of each monitoring sub-area, the equipment type and quantity of each monitoring sub-area are determined, including determining the installation location and quantity of cameras and the installation location and quantity of lighting equipment, and the quantity and location of the deployed underwater laser scanning arrays; Data network construction: Based on the wireless network, a data transmission channel is established for the monitoring equipment in each monitoring sub-area, a monitoring sensor network is built, and data sharing and real-time data transmission are carried out based on the monitoring sensor network.

3. The computer vision-based visual monitoring method for shrimp-rice symbiosis according to claim 1, characterized in that, The growth image data also includes: Based on the low-altitude flight of the unmanned aerial vehicle (UAV) for collection, the monitoring equipment carried by the UAV collects images according to a predetermined shooting frequency to obtain remote sensing image data of the paddy field; When the fixed camera detects abnormal growth in a local area, the UAV generates a cruise path based on the boundary coordinates of the corresponding monitoring area, and at the same time, adjusts the shooting frequency and camera resolution; The aerial image data obtained by the UAV is integrated with the monitoring image data obtained by the monitoring equipment to generate the growth image data of shrimp and rice in the monitoring area.

4. The computer vision-based visual monitoring method for shrimp-rice symbiosis according to claim 3, wherein, The target feature recognition specifically includes: Based on the growth image data, the image data of different growth stages of rice is extracted, and the color features, texture features and shape features of rice are extracted; The relationship between the color features, texture features and shape features of rice and biomass is established, and the rice biomass is estimated to obtain the biomass value corresponding to each monitoring area; The underwater three-dimensional point cloud data of the monitoring area is obtained through the deployed underwater laser scanning array; Based on the underwater three-dimensional point cloud data, a three-dimensional model of the crayfish habitat is generated, and the number of crayfish per unit area is determined in combination with the monitoring image data obtained by the monitoring equipment in each monitoring sub-area.

5. The computer vision-based visual monitoring method for shrimp-rice symbiosis according to claim 4, wherein, Determining the number of crayfish per unit area specifically includes: Preprocessing the obtained underwater three-dimensional point cloud data, performing surface reconstruction on the preprocessed three-dimensional point cloud data, and converting the discrete point cloud data into a continuous surface model to obtain a three-dimensional model of the crayfish habitat; Preprocessing the obtained monitoring image data, and performing crayfish target detection on the preprocessed video image; Performing spatial registration on the three-dimensional model of the crayfish habitat and the monitoring image data, determining the three-dimensional spatial area corresponding to each video image, counting the number of crayfish detected in the area, and combining the area to determine the number of crayfish per unit area.

6. The computer vision-based visual monitoring method for shrimp-rice symbiosis according to claim 5, characterized in that Counting the number of crayfish detected in the area also includes: Extract the biometric features of each crayfish from the monitored image data and establish a unique identifier for each crayfish; Establish a timeline based on the data length of the collected monitored image data, perform frame processing on the monitored image data, determine the initial position corresponding to each crayfish, determine the actual position of each initial position in the three-dimensional spatial region, and analyze the position characteristics of the crayfish; Track the moving positions corresponding to each crayfish in consecutive frames, establish the moving characteristics corresponding to the crayfish in combination with the initial positions. At the same time, extract the data proportion of each crayfish in consecutive frames to generate the pause characteristics of each crayfish; Based on the moving characteristics of each crayfish combined with the corresponding pause characteristics, obtain the behavior characteristics corresponding to each crayfish, and determine whether the crayfish is in a stress behavior state based on the behavior characteristics.

7. The computer vision-based visual monitoring method for shrimp-rice symbiosis according to claim 6, characterized in that The target feature recognition also includes: Obtain the basic terrain of the monitoring area based on the aerial image data obtained by the drone, measure the number of geometric units required to cover the terrain at different scales, quantify the roughness of the terrain surface, and extract the hydrological features; Determine the weights of the terrain factors based on the basic terrain and hydrological features of the monitoring area, and generate a comprehensive terrain complexity index; Divide the monitoring area into four levels based on the distribution characteristics of the comprehensive terrain complexity index, including extremely low complexity, low complexity, medium complexity, and high complexity.

8. The computer vision-based visual monitoring method for shrimp-rice symbiosis according to claim 7, characterized in that Based on the grading results of the monitoring area, perform spatial matching and time series analysis on the terrain complexity data, biomass, and water quality parameters to determine the association rules between terrain complexity and biomass fluctuations, water quality mutations, and construct a differential monitoring strategy.

9. The computer vision-based visual monitoring method for shrimp-rice symbiosis according to claim 1, characterized in that, Establish a growth model of the shrimp-rice symbiotic system, and also include: constructing a growth model of the shrimp-rice symbiotic system based on the association rules between terrain complexity and biomass fluctuations, water quality mutations, simulating the shrimp-rice symbiotic responses under different scenarios, and presenting the simulation results in real time based on three-dimensional dynamic visualization technology, including biomass change curves, crayfish distribution heat maps, and water quality parameter fluctuation cloud maps.

10. The computer vision-based visual monitoring method for shrimp-rice symbiosis according to claim 1, wherein Predict the symbiotic trend of shrimp and rice in the future period, and also include: calculating the health index of the monitoring area based on the terrain complexity data, rice biomass values, and crayfish density data, correlating with the water quality parameters output by the water quality parameter prediction model, and quantitatively evaluating the symbiotic effect intensity of different monitoring areas.

Images