Blue-green algae population particle characterization method and system based on artificial intelligence algorithm

Abstract

The invention discloses a cyanobacteria population particle characterization method based on an artificial intelligence algorithm, and the method comprises the following steps: carrying out the omnibearing and multi-angle data collection of a target water area through employing an unmanned plane and an unmanned ship carrier to carry a multispectral camera and water quality monitoring sensor equipment; carrying out denoising, correction and normalization preprocessing operation on the collected data; introducing a deep learning algorithm, constructing a blue-green algae image recognition model, and carrying out model training and optimization through a large amount of labeled blue-green algae image data on the basis of a convolutional neural network; the beneficial effects of the invention are that the method can automatically learn and extract complex features in the blue-green algae image, and achieves the high-precision recognition and classification of blue-green algae blooms; compared with a traditional image processing method, the recognition accuracy and stability are remarkably improved; the model is trained through multispectral image data, information of different spectral bands can be fully utilized, and the influence of environmental factors on the recognition result is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of cyanobacteria colony particle characterization, and in particular relates to a cyanobacteria colony particle characterization method and system based on an artificial intelligence algorithm. Background Art

[0002] Monitoring of cyanobacterial populations plays a vital role in the fields of environmental science and ecology. As an ancient type of microorganism, the excessive proliferation of cyanobacteria in water bodies is often closely related to eutrophication of water bodies. This eutrophication phenomenon not only changes the ecological balance of water bodies, but also may lead to the formation of cyanobacterial blooms, and then release toxins that are harmful to human health, such as microcystins. Therefore, monitoring of cyanobacterial populations is a key step in assessing and preventing water pollution.

[0003] Monitoring of cyanobacteria helps scientists and environmental managers understand the nutritional status of water bodies and the health of ecosystems. Through continuous observation of cyanobacteria, signs of water quality deterioration can be detected in a timely manner, so that corresponding management measures can be taken, such as reducing nitrogen and phosphorus emissions in agricultural runoff, or adjusting the flow dynamics of water bodies to reduce or prevent the occurrence of cyanobacterial blooms. Monitoring of cyanobacteria is also an important part of climate change research. The growth of cyanobacteria is significantly affected by environmental factors such as temperature, light and nutrients. Long-term monitoring of changes in cyanobacteria can reveal the specific impacts of climate change on aquatic ecosystems and provide data support for the formulation of adaptive management strategies.

[0004] Monitoring of cyanobacteria also has practical application value. For example, in aquaculture, excessive growth of cyanobacteria may affect aquaculture water quality and output. By monitoring cyanobacteria, timely early warning information can be provided to aquaculture operators, helping them take measures to protect the aquaculture environment and improve economic benefits. Monitoring of cyanobacteria is also an effective way to increase public environmental awareness and participation. Through education and publicity activities, the public can better understand the hazards and prevention measures of cyanobacterial blooms, and thus take action in their daily lives to reduce water pollution.

[0005] Monitoring of cyanobacterial populations is not only of great significance to scientific research, but also an important part of environmental management and public health protection; by monitoring cyanobacterial populations, we can gain a deeper understanding of the health of aquatic ecosystems, assess the impact of environmental changes, and take effective prevention and response measures.

[0006] Cyanobacteria are widely distributed groups of photosynthetic microorganisms that often dominate warm, nutrient-rich freshwater and saltwater lakes. They can produce a variety of toxins, posing a serious threat to drinking water, irrigation water and fisheries. Therefore, accurately characterizing the characteristics and distribution of cyanobacterial community particles is of great significance for environmental protection, water quality monitoring and ecological balance.

[0007] With the continuous advancement of remote sensing technology, especially the popularization of drone remote sensing technology, cyanobacteria monitoring has become more convenient and efficient. Drones can be equipped with multi-spectral and hyperspectral imaging equipment to conduct large-scale, high-resolution scans of water bodies to obtain key information such as the distribution and abundance of cyanobacteria. After obtaining a large amount of remote sensing data, how to efficiently and accurately process and analyze these data has become a key issue. Artificial intelligence algorithms, especially deep learning algorithms, are widely used in the field of cyanobacteria monitoring due to their powerful data processing and pattern recognition capabilities. By training deep learning models, it is possible to automatically identify and classify the spectral characteristics of cyanobacteria, thereby estimating the relative abundance of cyanobacteria and assessing the risk of cyanobacterial blooms.

[0008] The application of artificial intelligence algorithms in the characterization of cyanobacterial colony particles is the product of the combination of remote sensing technology and artificial intelligence technology. By obtaining the spectral characteristic data of cyanobacteria through unmanned aerial vehicle remote sensing technology, and using artificial intelligence algorithms for efficient and accurate processing and analysis, real-time monitoring and risk assessment of cyanobacterial blooms can be achieved. The application of this technology is of great significance for protecting water quality and maintaining ecological balance.

[0009] In the characterization of cyanobacteria colony particles, existing technologies are mainly based on remote sensing technology and traditional image processing methods. Although these methods can identify and monitor cyanobacteria to a certain extent, they still have the following disadvantages:

[0010] Limited recognition accuracy: Since the spectral characteristics of cyanobacteria are complex and easily affected by environmental factors, traditional image processing methods often have limited accuracy in distinguishing cyanobacteria from other water components. Especially when the drone is equipped with a camera based on the visible light band, it is difficult to accurately identify cyanobacteria blooms due to the lack of infrared band information and the unclear distinction of the green band;

[0011] Affected by environmental factors: When the drone is flying outdoors, the recognition effect is greatly affected by factors such as shooting angle, distance, and lighting conditions. This may cause some cyanobacteria blooms to be similar in color to the strong reflective areas far away from the camera, thereby increasing the possibility of misjudgment.

[0012] Low execution efficiency: In order to improve the recognition rate, some urban management departments frequently replace drone-mounted cameras or mount multiple cameras, but this sacrifices the execution efficiency and range of the drone;

[0013] Poor algorithm adaptability: Traditional image processing methods often rely on preset rules and thresholds. For the identification of cyanobacteria in different waters, different seasons, and different environmental conditions, it is necessary to continuously adjust and optimize the algorithm parameters, which increases the complexity and maintenance cost of the algorithm.

[0014] Lack of intelligent decision-making support: Most existing technologies remain at the stage of data collection and preliminary analysis. They lack intelligent decision-making support systems and are unable to automatically formulate response measures or early warning mechanisms based on the monitoring results of cyanobacteria. Summary of the invention

[0015] The purpose of the present invention is to provide a method and system for characterizing cyanobacterial colony particles based on an artificial intelligence algorithm. By introducing advanced technologies such as artificial intelligence, big data, and the Internet of Things, an intelligent cyanobacterial monitoring and early warning system is constructed, and response measures and early warning mechanisms are automatically formulated according to the monitoring results, providing strong support for environmental management and ecological protection.

[0016] To achieve the above object, the present invention provides the following technical solution: a method for characterizing cyanobacteria colony particles based on artificial intelligence algorithm, comprising the following steps:

[0017] Use drones and unmanned boats equipped with multi-spectral cameras and water quality monitoring sensor equipment to collect data from all directions and angles in the target waters;

[0018] Perform denoising, correction, and normalization preprocessing operations on the collected data;

[0019] Introducing deep learning algorithms, building a cyanobacteria image recognition model, based on convolutional neural networks, and using a large amount of labeled cyanobacteria image data for model training and optimization;

[0020] High-precision water quality monitoring sensors are used to monitor the key parameters of water bodies, such as algae protein density, pH value, electrolytes, chlorophyll and turbidity in real time. The monitoring data are combined with the results of cyanobacteria image recognition to comprehensively analyze the health status of the water body and the distribution of cyanobacteria.

[0021] Build an intelligent cyanobacteria monitoring and early warning system to automatically formulate response measures and early warning mechanisms based on monitoring results; use big data analysis technology to mine and analyze historical monitoring data to discover the laws and trends of cyanobacteria outbreaks and provide a scientific basis for future environmental management; realize remote monitoring and real-time data transmission through Internet of Things technology;

[0022] By combining multiple monitoring methods such as drones, unmanned ships and satellite remote sensing, an integrated air-space-ground and air-space cyanobacteria monitoring network is constructed.

[0023] As a preferred technical solution of the present invention, a variety of monitoring methods such as drones, unmanned ships, and satellite remote sensing are combined to build an integrated air-space-ground cyanobacteria monitoring network, in which drones are used for rapid inspections and key area monitoring, unmanned ships conduct close-range, high-precision water quality monitoring, and satellite remote sensing conducts large-scale periodic monitoring.

[0024] As a preferred technical solution of the present invention, the method for constructing the cyanobacteria image recognition model is as follows:

[0025] Choose the deep learning framework TensorFlow or PyTorch to build a cyanobacteria image recognition model;

[0026] According to the characteristics of cyanobacteria images and the requirements of recognition tasks, the convolutional neural network structure is designed;

[0027] Select a loss function to evaluate the difference between the model prediction results and the true annotations, and define an optimization algorithm to minimize the loss function to optimize the model parameters.

[0028] As a preferred technical solution of the present invention, the network structure includes an input layer, multiple convolutional layers, a pooling layer, a fully connected layer and an output layer.

[0029] As a preferred technical solution of the present invention, the model training includes

[0030] Training process: input the preprocessed image data into the model, calculate the prediction results through forward propagation, and calculate the gradient and update the model parameters through back propagation;

[0031] Hyperparameter Tuning: During the training process, hyperparameters need to be adjusted to optimize the performance of the model.

[0032] As a preferred technical solution of the present invention, the hyperparameters include learning rate, batch size, and convolution kernel size; the adjustment of the hyperparameters is determined based on experience or by a cross-validation method.

[0033] As a preferred technical solution of the present invention, it also includes model evaluation: after the training is completed, the model is evaluated using a test set, and the evaluation indicators include accuracy, recall rate, and F1 score.

[0034] As a preferred technical solution of the present invention, the corresponding contents of the response measures and early warning mechanism automatically formulated according to the monitoring results are as follows:

[0035] Use big data analysis technology to conduct in-depth mining and analysis of historical monitoring data, identify key factors affecting the growth of cyanobacteria and outbreak patterns, and use machine learning algorithms to establish a cyanobacteria outbreak prediction model;

[0036] Input the real-time monitoring data into the prediction model to conduct real-time assessment and prediction of the risk of blue-green algae outbreaks. Through model calculation, the possibility and severity of blue-green algae outbreaks in the future can be predicted;

[0037] Different warning levels will be set based on the predicted results of blue-green algae outbreaks, and the triggering conditions and response measures for each level of warning will be clarified.

[0038] The present invention also discloses a cyanobacteria colony particle characterization system based on artificial intelligence algorithm, comprising:

[0039] Data collection and preprocessing module: Use drones and unmanned boats to carry multi-spectral cameras and water quality monitoring sensor equipment to collect data from all directions and angles in the target waters; perform denoising, correction, and normalization preprocessing on the collected data;

[0040] Cyanobacteria image intelligent recognition module: Introducing deep learning algorithms to build a cyanobacteria image recognition model. This model is based on a convolutional neural network and is trained and optimized through a large amount of labeled cyanobacteria image data to improve the recognition accuracy and generalization ability of the model. It combines infrared and near-infrared multispectral image data to improve the ability to extract spectral features of cyanobacteria and achieve high-precision recognition and classification of cyanobacteria blooms. By continuously optimizing the algorithm model, it can adapt to cyanobacteria recognition tasks under different lighting conditions, shooting angles and distances.

[0041] Real-time monitoring module for water quality parameters: Use high-precision water quality monitoring sensors to monitor the algae protein density, pH value, electrolytes, chlorophyll and turbidity parameters of the water body in real time, combine the monitoring data with the results of cyanobacteria image recognition, and comprehensively analyze the health status of the water body and the distribution of cyanobacteria;

[0042] Blue algae monitoring and early warning system: automatically formulate response measures and early warning mechanisms based on monitoring results, use big data analysis technology to mine and analyze historical monitoring data, and realize remote monitoring and real-time data transmission through Internet of Things technology;

[0043] Integrated air-space-ground collaborative monitoring system: Combines multiple monitoring methods such as drones, unmanned ships, and satellite remote sensing to build an integrated air-space-ground cyanobacteria monitoring network.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention can automatically learn and extract complex features in cyanobacteria images, and realize high-precision recognition and classification of cyanobacteria blooms; compared with traditional image processing methods, it significantly improves the accuracy and stability of recognition;

[0046] The model of the present invention is trained with multispectral image data and can make full use of information from different spectral bands to reduce the impact of environmental factors on the recognition results. At the same time, the model also has a certain generalization ability and can adapt to the cyanobacteria recognition tasks in different waters, different seasons and different environmental conditions.

[0047] With the help of high-efficiency monitoring equipment such as drones and unmanned ships, the technical solution of the present invention can achieve rapid and large-area cyanobacteria monitoring; combined with efficient image processing and recognition algorithms, the execution efficiency of monitoring tasks and the data processing speed are significantly improved;

[0048] It can automatically formulate response measures and early warning mechanisms based on monitoring results, providing intelligent decision-making support for environmental management and ecological protection;

[0049] Through the coordinated work of multiple monitoring methods such as drones, unmanned boats, and satellite remote sensing, comprehensive, three-dimensional, and real-time monitoring of the target waters is achieved; this collaborative monitoring method improves the comprehensiveness and accuracy of monitoring and provides more reliable data support for blue algae control;

[0050] The technical solution of the present invention has strong flexibility and scalability; both the selection of hardware equipment and the development of software systems adopt modular design, which is convenient for customization and deployment according to different needs and scenarios; at the same time, with the continuous advancement of technology and the continuous accumulation of data, the system can also be continuously upgraded and optimized to adapt to more complex monitoring tasks;

[0051] By achieving efficient cyanobacteria monitoring and early warning, the present invention helps to timely discover and control the occurrence of cyanobacteria blooms, reducing their damage to aquatic ecosystems and environmental pollution; at the same time, intelligent decision support also reduces the cost and difficulty of manual intervention, and improves the efficiency and economic benefits of environmental governance. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 The figure is a flow chart of the characterization method of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] Example 1

[0055] See also Figure 1 , which is the first embodiment of the present invention, provides a method and system for characterizing cyanobacteria colony particles based on an artificial intelligence algorithm, comprising the following parts:

[0056] Data collection and preprocessing:

[0057] Use drones, unmanned boats and other carriers equipped with multi-spectral cameras, water quality monitoring sensors and other equipment to collect data from all directions and angles in the target waters;

[0058] Preprocess the collected data, including denoising, correction, normalization and other operations, to improve data quality and the accuracy of subsequent processing;

[0059] Intelligent recognition of cyanobacteria images:

[0060] Introducing deep learning algorithms to build a cyanobacteria image recognition model, the model can automatically learn and extract the spectral features of cyanobacteria, and achieve high-precision recognition and classification of cyanobacteria blooms; using multispectral image data as input, combined with information from infrared, near-infrared and other bands, to improve the accuracy and stability of cyanobacteria recognition; by continuously optimizing the algorithm model, it can adapt to cyanobacteria recognition tasks under different lighting conditions, shooting angles and distances;

[0061] Real-time monitoring of water quality parameters: Use water quality monitoring sensors to monitor key parameters of water bodies such as algae protein density, pH value, electrolytes, chlorophyll and turbidity in real time; combine monitoring data with cyanobacteria image recognition results to comprehensively analyze the health status of water bodies and the distribution of cyanobacteria;

[0062] Intelligent decision support: Build an intelligent cyanobacteria monitoring and early warning system to automatically formulate response measures and early warning mechanisms based on monitoring results; use big data analysis technology to mine and analyze historical monitoring data to discover the laws and trends of cyanobacteria outbreaks and provide a scientific basis for future environmental management; use the Internet of Things technology to achieve remote monitoring and real-time data transmission to ensure the timeliness and accuracy of monitoring results;

[0063] Integrated air-ground-space collaborative monitoring:

[0064] Combine drones, unmanned boats, satellite remote sensing and other monitoring methods to build an integrated air-space-ground blue algae monitoring network; use the flexibility and efficiency of drones to conduct rapid inspections and key area monitoring; use unmanned boats to conduct close-range, high-precision water quality monitoring; use satellite remote sensing to conduct large-scale periodic monitoring; and improve the comprehensiveness and accuracy of blue algae monitoring through data fusion and collaborative analysis;

[0065] Technical implementation details

[0066] Construction of cyanobacteria image recognition model: Using deep learning models such as convolutional neural network (CNN) as the basis for cyanobacteria image recognition; training and optimizing the model through a large amount of labeled cyanobacteria image data to improve the recognition accuracy and generalization ability of the model; introducing advanced technologies such as attention mechanism to improve the model's attention to key features and recognition efficiency;

[0067] Real-time monitoring system for water quality parameters:

[0068] Design high-precision water quality monitoring sensors to ensure the accuracy and stability of monitoring data; develop data acquisition and transmission modules to achieve real-time acquisition and remote transmission of monitoring data; build a data processing and analysis platform to process and analyze monitoring data in real time and generate visual monitoring reports;

[0069] Intelligent decision support system:

[0070] Develop intelligent decision-making algorithms and models to automatically formulate response measures and early warning mechanisms based on monitoring results; build a user-friendly operation interface and visualization platform to facilitate users to view monitoring results and decision-making recommendations; provide data export and report generation functions to facilitate users to conduct subsequent data analysis and decision support;

[0071] Integrated air-ground-space collaborative monitoring network:

[0072] Design collaborative monitoring plans for drones, unmanned ships and satellite remote sensing, and clarify their respective monitoring tasks and data processing procedures; develop data fusion and collaborative analysis algorithms to achieve data sharing and collaborative analysis among different monitoring methods; build a unified data management platform to centrally manage and coordinate monitoring data.

[0073] Example 2

[0074] See also Figure 1 , which is the second embodiment of the present invention, is based on the previous embodiment, except that:

[0075] Build an intelligent blue algae monitoring and early warning system, and automatically formulate response measures and early warning mechanisms based on the monitoring results. The corresponding methods are as follows:

[0076] Using the Internet of Things technology, water quality parameters and meteorological data are collected in real time through sensors, buoys, automatic water quality monitoring stations and other monitoring equipment; the collected raw data are cleaned, denoised, calibrated and standardized to ensure the accuracy and reliability of the data;

[0077] Use big data analysis technology to conduct in-depth mining and analysis of historical monitoring data to identify key factors affecting the growth of cyanobacteria and outbreak patterns; use machine learning algorithms (such as decision trees, random forests, neural networks, etc.) to establish a cyanobacteria outbreak prediction model; input real-time monitoring data into the prediction model to conduct real-time assessment and prediction of cyanobacteria outbreak risks; and use model calculations to predict the possibility and severity of cyanobacteria outbreaks in the future.

[0078] According to the prediction results of blue algae outbreaks, different warning levels (such as level one response, level two response, level three response, etc.) are set, and the triggering conditions and response measures of each level of warning are clarified;

[0079] Level 1 response: When it is predicted that the outbreak of blue algae is large and may seriously affect the safety of drinking water, the level 1 response should be initiated immediately; the response measures include strengthening blue algae monitoring, emergency interception and salvage of blue algae, taking temporary water supply measures, and launching emergency plans;

[0080] Secondary response: When the outbreak area of ​​cyanobacteria is predicted to be moderate and has a certain impact on drinking water safety, the secondary response will be initiated; measures include strengthening monitoring, appropriately salvaging cyanobacteria, and preparing emergency supplies;

[0081] Level 3 response: When it is predicted that the outbreak area of ​​blue algae is small and the impact on drinking water safety is small, the level 3 response will be initiated; the main measures are to strengthen monitoring and early warning, and prepare for emergencies.

[0082] The technical principle of the cyanobacteria image recognition model is mainly based on deep learning algorithms, especially the application of models such as convolutional neural networks (CNN). The following is a detailed description of its technical principle:

[0083] 1. Data Preparation

[0084] Data collection: Use drones, unmanned boats and other carriers equipped with multispectral cameras and other equipment to collect high-resolution images of the target waters; these images contain the reflectance characteristics of cyanobacteria in different spectral bands and are an important basis for identifying cyanobacteria;

[0085] Data annotation: Manually annotate the collected images, distinguish the cyanobacteria area from the non-cyanobacteria area in the image, and mark the type or concentration level of cyanobacteria; these annotated data will be used for model training;

[0086] Data preprocessing: Preprocess the labeled image data, including denoising, correction, normalization and other operations to improve data quality and the accuracy of subsequent processing; at the same time, it may also be necessary to enhance the image to improve the visibility and distinction of the cyanobacteria area;

[0087] 2. Model Construction

[0088] Choose a deep learning framework: Choose a suitable deep learning framework (such as TensorFlow, PyTorch, etc.) to build a cyanobacteria image recognition model; these frameworks provide a rich set of neural network layers, optimization algorithms, and tool libraries to help quickly build and train models;

[0089] Design network structure: Design a suitable convolutional neural network structure according to the characteristics of cyanobacteria images and the requirements of the recognition task; the network structure usually includes input layer, multiple convolutional layers, pooling layer, fully connected layer and output layer; the convolutional layer is used to extract feature information from the image, the pooling layer is used to reduce the dimension and complexity of the feature map, and the fully connected layer is used to map the extracted feature information to the output category;

[0090] Define loss function and optimization algorithm: Select an appropriate loss function (such as cross entropy loss function) to evaluate the difference between the model prediction results and the true annotations, and define an optimization algorithm (such as gradient descent algorithm) to minimize the loss function, thereby optimizing the model parameters;

[0091] 3. Model Training

[0092] Training process: input the preprocessed image data into the model, calculate the prediction results through forward propagation, and calculate the gradient and update the model parameters through back propagation; this process will be repeated many times (i.e. multiple training cycles) until the performance of the model on the validation set reaches stability or reaches the preset training rounds;

[0093] Hyperparameter adjustment: During the training process, some hyperparameters (such as learning rate, batch size, convolution kernel size, etc.) may need to be adjusted to optimize the performance of the model; the adjustment of these hyperparameters is usually determined based on experience or through methods such as cross-validation;

[0094] Model evaluation: After training, the model is evaluated using the test set to verify the model's generalization ability and recognition accuracy; evaluation indicators usually include accuracy, recall rate, F1 score, etc.

[0095] IV. Model Application

[0096] Cyanobacteria recognition: Apply the trained model to new cyanobacteria image data to achieve automatic recognition and classification of cyanobacteria; the model will output the corresponding cyanobacteria category or concentration level based on the feature information of the input image;

[0097] Real-time monitoring and early warning: The cyanobacteria image recognition model is combined with monitoring equipment such as drones and unmanned boats to achieve real-time monitoring and early warning of target waters. When cyanobacteria blooms are detected, the system can automatically issue early warning signals and take corresponding countermeasures.

[0098] Data visualization and analysis: The identification results are displayed to users in a visual form, which helps them understand the distribution and changing trends of cyanobacteria. At the same time, the identification results can also be analyzed in depth to reveal the laws and mechanisms of cyanobacteria outbreaks.

[0099] In summary, the technical principles of the cyanobacteria image recognition model are mainly based on the application of deep learning algorithms and convolutional neural networks; by building a suitable network structure, training models and evaluating performance, high-precision recognition and classification of cyanobacteria images can be achieved.

[0100] Although the embodiments of the present invention have been shown and described, as detailed above, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for characterizing cyanobacteria colony particles based on artificial intelligence algorithm, characterized in that: The steps include: Use drones and unmanned boats equipped with multi-spectral cameras and water quality monitoring sensor equipment to collect data from all directions and angles in the target waters; Perform denoising, correction, and normalization preprocessing operations on the collected data; Introducing deep learning algorithms, building a cyanobacteria image recognition model, based on convolutional neural networks, and using a large amount of labeled cyanobacteria image data for model training and optimization; High-precision water quality monitoring sensors are used to monitor the key parameters of water bodies, such as algae protein density, pH value, electrolytes, chlorophyll and turbidity in real time. The monitoring data are combined with the results of cyanobacteria image recognition to comprehensively analyze the health status of the water body and the distribution of cyanobacteria. Build an intelligent cyanobacteria monitoring and early warning system to automatically formulate response measures and early warning mechanisms based on monitoring results; use big data analysis technology to mine and analyze historical monitoring data to discover the laws and trends of cyanobacteria outbreaks and provide a scientific basis for future environmental management; realize remote monitoring and real-time data transmission through Internet of Things technology; By combining multiple monitoring methods such as drones, unmanned ships and satellite remote sensing, an integrated air-space-ground and air-space cyanobacteria monitoring network is constructed.

2. The method for characterizing cyanobacteria colony particles based on an artificial intelligence algorithm according to claim 1, characterized in that: By combining multiple monitoring methods such as drones, unmanned boats, and satellite remote sensing, an integrated air-space-ground and air-space cyanobacteria monitoring network is constructed. Drones are used for rapid inspections and key area monitoring, unmanned boats conduct close-range, high-precision water quality monitoring, and satellite remote sensing conducts large-scale periodic monitoring.

3. The method for characterizing cyanobacteria colony particles based on an artificial intelligence algorithm according to claim 1, characterized in that: The method for constructing the cyanobacteria image recognition model is as follows: Choose the deep learning framework TensorFlow or PyTorch to build a cyanobacteria image recognition model; According to the characteristics of cyanobacteria images and the requirements of recognition tasks, the convolutional neural network structure is designed; Select a loss function to evaluate the difference between the model prediction results and the true annotations, and define an optimization algorithm to minimize the loss function to optimize the model parameters.

4. The method for characterizing cyanobacteria colony particles based on an artificial intelligence algorithm according to claim 3, characterized in that: The network structure includes an input layer, multiple convolutional layers, a pooling layer, a fully connected layer, and an output layer.

5. The method for characterizing cyanobacteria colony particles based on an artificial intelligence algorithm according to claim 1, characterized in that: The model training includes Training process: input the preprocessed image data into the model, calculate the prediction results through forward propagation, and calculate the gradient and update the model parameters through back propagation; Hyperparameter Tuning: During the training process, hyperparameters need to be adjusted to optimize the performance of the model.

6. The method for characterizing cyanobacteria colony particles based on an artificial intelligence algorithm according to claim 5, characterized in that: The hyperparameters include learning rate, batch size, and convolution kernel size; the adjustment of the hyperparameters is determined based on experience or through a cross-validation method.

7. The method for characterizing cyanobacteria colony particles based on an artificial intelligence algorithm according to claim 1, characterized in that: It also includes model evaluation: after training, the model is evaluated using the test set. The evaluation indicators include accuracy, recall, and F1 score.

8. The method for characterizing cyanobacteria colony particles based on an artificial intelligence algorithm according to claim 1, characterized in that: The corresponding contents of the response measures and early warning mechanism automatically formulated according to the monitoring results are as follows: Use big data analysis technology to conduct in-depth mining and analysis of historical monitoring data, identify key factors affecting the growth of cyanobacteria and outbreak patterns, and use machine learning algorithms to establish a cyanobacteria outbreak prediction model; Input the real-time monitoring data into the prediction model to conduct real-time assessment and prediction of the risk of blue-green algae outbreaks. Through model calculation, the possibility and severity of blue-green algae outbreaks in the future can be predicted; Different warning levels will be set based on the predicted results of blue-green algae outbreaks, and the triggering conditions and response measures for each level of warning will be clarified.

9. A cyanobacteria colony particle characterization system based on artificial intelligence algorithm, characterized by: include Data collection and preprocessing module: Use drones and unmanned boats to carry multi-spectral cameras and water quality monitoring sensor equipment to collect data from all directions and angles in the target waters; perform denoising, correction, and normalization preprocessing on the collected data; Cyanobacteria image intelligent recognition module: Introducing deep learning algorithms to build a cyanobacteria image recognition model. This model is based on a convolutional neural network and is trained and optimized through a large amount of labeled cyanobacteria image data to improve the recognition accuracy and generalization ability of the model. It combines infrared and near-infrared multispectral image data to improve the ability to extract spectral features of cyanobacteria and achieve high-precision recognition and classification of cyanobacteria blooms. By continuously optimizing the algorithm model, it can adapt to cyanobacteria recognition tasks under different lighting conditions, shooting angles and distances. Real-time monitoring module for water quality parameters: Use high-precision water quality monitoring sensors to monitor the algae protein density, pH value, electrolytes, chlorophyll and turbidity parameters of the water body in real time, combine the monitoring data with the results of cyanobacteria image recognition, and comprehensively analyze the health status of the water body and the distribution of cyanobacteria; Blue algae monitoring and early warning system: automatically formulate response measures and early warning mechanisms based on monitoring results, use big data analysis technology to mine and analyze historical monitoring data, and realize remote monitoring and real-time data transmission through Internet of Things technology; Integrated air-space-ground collaborative monitoring system: Combines multiple monitoring methods such as drones, unmanned ships, and satellite remote sensing to build an integrated air-space-ground cyanobacteria monitoring network.

Images