Mariculture intelligent growth monitoring and analysis method and system

Abstract

The invention discloses an intelligent growth monitoring and analysis method and system for mariculture, and the method comprises the following steps: firstly, collecting breeding environment data in real time through a first sensor group; and acquiring growth data of the breeding object through the second sensor group. Then, the cultivation environment data and the growth data are transmitted to a cloud platform by using a data acquisition system, and the data are preprocessed by the cloud platform to generate a standardized data model; based on the standardized data, a deep learning algorithm is used for analyzing the growth trend of the breeding object and predicting the growth change in a period of time in the future. Through combining the breeding environment and the growth data, the cloud platform performs risk assessment, generates early warning information in real time, and identifies potential risk factors. According to the early warning information, the cloud platform automatically adjusts breeding environment parameters and management measures, and healthy growth of breeding objects is ensured. According to the method, intelligent decision support is provided for breeding management, and the breeding efficiency and the growth quality of bred objects are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of marine aquaculture, and particularly to an intelligent growth monitoring and analysis method and system for marine aquaculture. Background Art

[0002] With the increase in the global population, the demand for food is also growing day by day. In particular, marine food, as an important source of protein, has received increasing attention. Marine aquaculture, as an important source of aquatic food supply, has become an indispensable part of global agricultural production. However, with the expansion of the scale of marine aquaculture, how to scientifically and effectively manage the aquaculture environment and aquaculture objects has become an urgent challenge to be solved. Traditional aquaculture management methods often lack timely and accurate monitoring means when facing the rapidly changing aquaculture environment, and it is difficult to meet the requirements of refined and intelligent management in modern marine aquaculture.

[0003] Although the existing intelligent monitoring systems for marine aquaculture have made certain progress, there are still some problems in practical applications. First of all, most of the data monitoring systems in the existing technologies only focus on single-type data and lack comprehensive analysis of various environmental factors and growth factors. In addition, there is still a certain lag in the real-time feedback and processing of data in the existing systems, and it is difficult to quickly adjust the environment and management measures during the aquaculture process, and it fails to truly achieve precise dynamic management.

[0004] Therefore, how to improve the accuracy, real-time performance and multi-dimensional analysis ability of the intelligent monitoring system for marine aquaculture has become an urgent problem to be solved. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide an intelligent growth monitoring and analysis method and system for marine aquaculture to solve the above technical problems.

[0006] To achieve the above object, in a first aspect, an intelligent growth monitoring and analysis method for marine aquaculture is provided, which includes:

[0007] Real-time collecting aquaculture environment data through a first sensor group;

[0008] Real-time obtaining the growth data of aquaculture objects through a second sensor group;

[0009] Transmitting the aquaculture environment data and the growth data of the aquaculture objects to a cloud platform through a data collection system, and performing data processing by the cloud platform to obtain standardized data;

[0010] Based on the standardized data, analyzing and predicting the growth trend of aquaculture objects through a deep learning algorithm to obtain the predicted growth trend of aquaculture objects within a future period of time;

[0011] Based on the predicted growth trend, combined with the aquaculture environment data and the growth data of the aquaculture object, conduct risk assessment and generate warning information in real time; adjust the aquaculture environment and aquaculture management measures according to the warning information to ensure the healthy growth of the aquaculture object.

[0012] In a second aspect, a marine aquaculture intelligent growth monitoring and analysis system is provided, which includes:

[0013] A first sensor group for real-time collection of aquaculture environment data;

[0014] A second sensor group for real-time acquisition of the growth data of the aquaculture object;

[0015] A data acquisition system for receiving the aquaculture environment data collected by the first sensor group and the growth data of the aquaculture object collected by the second sensor group, and transmitting them to the cloud platform;

[0016] A cloud platform for receiving the aquaculture environment data and the growth data of the aquaculture object, performing data processing, and generating standardized data;

[0017] A deep learning analysis module located on the cloud platform for analyzing and predicting the growth trend of the aquaculture object based on the standardized data through deep learning algorithms, and obtaining the predicted growth trend of the aquaculture object within a future period of time;

[0018] A risk analysis module located on the cloud platform for conducting risk assessment and generating real-time warning information according to the predicted growth trend, the aquaculture environment data and the growth data of the aquaculture object;

[0019] An adjustment module located on the cloud platform for adjusting the aquaculture environment and aquaculture management measures according to the warning information to ensure the healthy growth of the aquaculture object.

[0020] The above technical solution has the following beneficial technical effects:

[0021] The marine aquaculture intelligent growth monitoring and analysis method of the embodiment of the present invention can achieve accurate prediction and dynamic monitoring of the growth trend of aquaculture objects by integrating multiple sensors to collect aquaculture environment data and aquaculture object growth data in real time and using deep learning algorithms to analyze and predict the data. Through the real-time risk assessment and warning system, environmental changes or abnormalities in the growth of aquaculture objects can be detected in a timely manner, and then the aquaculture environment and management measures can be automatically adjusted to ensure the healthy growth of aquaculture objects under the best conditions. This method effectively improves the intelligent level of aquaculture management, optimizes resource allocation, reduces the dependence on manual intervention, and improves aquaculture efficiency. Description of the Drawings

[0022] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation to the present invention. Among them:

[0023] Figure 1 is a flowchart of a method for intelligent growth monitoring and analysis in marine aquaculture according to an embodiment of the present invention;

[0024] Figure 2 is a specific flowchart of step S2 in the embodiment of the present invention;

[0025] Figure 3 is a specific flowchart of step S3 in the embodiment of the present invention;

[0026] Figure 4 is a specific flowchart of step S4 in the embodiment of the present invention;

[0027] Figure 5 is a specific flowchart of step S5 in the embodiment of the present invention;

[0028] Figure 6 is a specific flowchart of step S53 in the embodiment of the present invention;

[0029] Figure 7 is a functional block diagram of a system for intelligent growth monitoring and analysis in marine aquaculture according to an embodiment of the present invention;

[0030] Figure 8 is a schematic structural diagram of a computer system according to an embodiment of the present invention. Detailed implementation manners

[0031] The following describes exemplary embodiments of the present invention with reference to the accompanying drawings. Various details of the embodiments of the present invention are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, descriptions of well-known functions and structures are omitted in the following.

[0032] The objective of the embodiment of the present invention is to provide a method and system for intelligent growth monitoring and analysis in marine aquaculture. By integrating various sensor data acquisitions and deep learning algorithms, the growth trends of the aquaculture environment and aquaculture objects are analyzed in real time, and dynamic monitoring and prediction of the healthy growth of the aquaculture environment and aquaculture objects are realized.

[0033] Embodiment 1

[0034] As Figure 1 shown, a method for intelligent growth monitoring and analysis in marine aquaculture according to an embodiment of the present invention includes the following steps:

[0035] S1: Real-time collect aquaculture environment data through the first sensor group.

[0036] In this embodiment, the first sensor group includes multiple sensors for collecting data of the aquaculture environment in real time. Specifically, the sensor group includes a marine noise level sensor, a trace element concentration sensor, a water transparency sensor, a temperature gradient sensor, etc. Each sensor transmits the collected data to the data acquisition system in real time through a wireless network or a wired network. Taking the marine noise level sensor as an example, the sensor reflects the ecological changes in the aquaculture environment in real time by monitoring the underwater noise intensity, helping to identify potential environmental stress factors. Other sensors, such as the trace element concentration sensor, ensure that the aquaculture objects obtain sufficient nutrition by continuously monitoring the concentration of trace elements (such as iron, zinc, etc.) in the water. The water transparency sensor helps to monitor the concentration of suspended substances in the water to ensure the cleanliness and health of the water quality.

[0037] S2: Obtain the growth data of the aquaculture objects in real time through the second sensor group.

[0038] Specifically, the second sensor group is responsible for collecting the growth data of the aquaculture objects in real time, mainly including the weight, body length, morphological characteristics, etc. of the aquaculture objects. In this embodiment, the second sensor group adopts devices such as an underwater buoy weighing device, a high-definition camera, and an infrared sensor. The underwater buoy weighing device records the weight of the aquaculture objects in real time through the change in buoyancy. The high-definition camera is used to take images of the aquaculture objects and analyze the body length through image processing algorithms. The infrared sensor detects the activity status of the aquaculture objects and provides data on the activity frequency and movement range. These data are wirelessly transmitted to the data acquisition system and further integrated and processed together with the aquaculture environment data. The aquaculture objects include one or more of fish, shellfish, and seaweed.

[0039] S3: Transmit the aquaculture environment data and the growth data of the aquaculture objects to the cloud platform through the data acquisition system, and the cloud platform processes the data to obtain standardized data.

[0040] Specifically, in the data acquisition stage, all data from the first sensor group and the second sensor group are transmitted to the cloud platform through the data acquisition system. The data acquisition system uses standard data transmission protocols, such as MQTT or HTTP, to upload the real-time data collected by the sensors to the cloud platform. The data received by the cloud platform includes environmental data (such as water temperature, salinity, dissolved oxygen concentration) and growth data (such as weight, body length, activity status). The system automatically identifies and stores the timestamp of each piece of data to ensure that the data is stored in chronological order. During the data transmission process, the system ensures the security and integrity of the data and uses encryption and compression technologies to optimize the data transmission efficiency.

[0041] S4: Based on the standardized data, analyze and predict the growth trend of the aquaculture objects through deep learning algorithms to obtain the predicted growth trend of the aquaculture objects in a future period of time.

[0042] Specifically, after receiving the aquaculture environment data and the growth data of the aquaculture objects, the cloud platform processes the data through a deep learning analysis module. This module first standardizes the original data, including data cleaning, removing outliers, and filling in missing data. Then, the standardized data is input into the deep learning model, which analyzes the growth trend of the aquaculture objects by learning the relationship between the growth trend and environmental factors in the historical data. The deep learning model uses algorithms such as LSTM (Long Short-Term Memory Network) for time series prediction to predict the growth trend of the aquaculture objects in the future for a period of time, including changes in key indicators such as weight and body length. This prediction can provide a scientific basis for aquaculture management to help managers make adjustments in advance.

[0043] S5: According to the predicted growth trend, combined with the aquaculture environment data and the growth data of the aquaculture objects, conduct risk assessment and generate warning information in real time; adjust the aquaculture environment and aquaculture management measures according to the warning information to ensure the healthy growth of the aquaculture objects.

[0044] Specifically, after obtaining the predicted growth trend of the aquaculture objects, the cloud platform conducts risk assessment based on the prediction results and the current environmental data. The risk analysis module uses data analysis models (such as decision trees, support vector machines, etc.) for analysis to identify potential risk factors. For example, when the predicted growth trend shows abnormalities, such as a slowdown in growth rate, and the water quality or temperature in the environmental data is abnormal, the system will issue a warning indicating that there are problems with the aquaculture environment, which will affect the growth of the aquaculture objects. The cloud platform generates warning information in real time and transmits the alarm to the aquaculture managers through the notification system so that they can take corresponding measures in a timely manner.

[0045] Specifically, after receiving the warning information, the cloud platform automatically or recommends adjusting the aquaculture environment and management measures. For example, if the dissolved oxygen concentration in the water body is too low, the system can automatically adjust the working state of the air pump to increase the dissolved oxygen concentration in the water; if the water temperature is too high, the system can adjust the water flow or take cooling measures. In addition, the system can also adjust aquaculture management measures such as feed input amount, feeding frequency, and aquaculture density to optimize the growth environment of the aquaculture objects. Managers can optimize and adjust the aquaculture environment according to the real-time suggestions of the cloud platform to ensure the healthy growth of the aquaculture objects under the best conditions.

[0046] In some embodiments, the aquaculture environment data includes any combination of ocean noise level, trace element concentration, change in aquaculture water depth, microbial community structure, temperature gradient change, density of phytoplankton in water, heavy metal concentration in water body, water transparency, water flow turbulence, and gas exchange rate; the growth data of the aquaculture objects includes any combination of weight, body length, morphological characteristics, activity status, body fat content, change in growth rate, and muscle fiber density.

[0047] In this embodiment, the aquaculture environment data includes multiple parameters for comprehensively monitoring the health and suitability of the aquaculture environment. Specifically, the marine noise level sensor monitors the underwater noise intensity of the aquaculture water area in real time. By analyzing the noise level, it can be determined whether there are external noise sources in the environment that affect the health of the aquaculture objects. The trace element concentration sensor is used to monitor the concentration of trace elements in the water in real time, such as iron, zinc, etc. These elements are important for the growth of the aquaculture objects. The aquaculture water depth sensor monitors the change in water depth, especially the depth fluctuations generated during the aquaculture process. This information helps to evaluate the change in water flow and its impact on the growth of the aquaculture objects.

[0048] The microbial community structure sensor helps to judge the health status of the water quality by monitoring the types and quantities of microorganisms in the water. For example, the change in the bacterial population in the water body will affect the water quality or the health of the aquaculture objects. The temperature gradient sensor monitors the temperature difference at different depths of the water body, which is conducive to analyzing whether the water body has undergone uneven mixing, affecting the distribution and growth environment of the aquaculture objects. The phytoplankton density sensor in the water monitors the number of phytoplankton in the water in real time. The phytoplankton density directly affects the oxygen production and water quality balance in the water and is an important factor affecting the growth of the aquaculture objects.

[0049] The water heavy metal concentration sensor is used to monitor the possible heavy metal pollution in the water (such as lead, mercury, cadmium, etc.). These substances are toxic to the aquaculture objects, and excessive concentrations will directly affect the health of the aquaculture objects. The water transparency sensor monitors the clarity of the water. The transparency reflects the concentration of suspended substances in the water and has an important impact on underwater photosynthesis and the ecological environment. The water flow turbulence sensor monitors the turbulence degree of the water flow. The change in turbulence degree will affect the activities of the aquaculture objects and the distribution of feed. The gas exchange rate sensor is used to monitor the gas exchange between the water surface and the air (such as carbon dioxide and oxygen). This parameter evaluates the ventilation performance and dissolved oxygen level of the water body to ensure that the aquaculture objects obtain sufficient oxygen.

[0050] In this embodiment, the growth data of the aquaculture objects includes parameters in multiple dimensions for comprehensively monitoring the health and growth of the aquaculture objects. The weight sensor records the weight of the aquaculture objects in real time through the underwater buoy weighing device. The weight is a key indicator reflecting the health status and growth progress of the aquaculture objects. The body length data is obtained through a high-definition camera. The camera takes pictures of the aquaculture objects, and the body length is calculated by combining image processing algorithms. The body length data is an important indicator of the growth state and can reflect the body shape change of the aquaculture objects.

[0051] Morphological features are automatically extracted through image recognition methods. The system uses image analysis technology to obtain the morphological features of the aquaculture objects (such as the shape of the dorsal fin, body size ratio, etc.). By comparing with different growth stages, the morphological changes of the aquaculture objects are analyzed to evaluate their growth status. The activity state is monitored by an infrared sensor to obtain the activity frequency and movement range of the aquaculture objects. This information is helpful for evaluating the activity, health status, and environmental adaptability of the aquaculture objects.

[0052] The body fat content is measured by scanning the aquaculture object with an ultrasonic probe to measure the thickness of the fat layer, and the fat content is calculated in combination with a set mathematical model. The fat content is an important indicator for evaluating the nutritional status and health level of the aquaculture object. The change in growth rate is calculated by regularly recording the changes in the weight and body length of the aquaculture object. By monitoring the growth rate of the aquaculture object, abnormal growth or non-conforming situations can be identified in a timely manner for effective intervention.

[0053] The muscle fiber density is obtained by taking pictures of the muscle part of the aquaculture object with a high-definition camera and extracting the density of muscle fibers in combination with an image processing algorithm. The muscle development directly affects the quality and market value of the aquaculture object.

[0054] In some embodiments, the first sensor group includes any combination of: a marine noise level sensor, a trace element concentration sensor, a sensor for the change in the depth of the aquaculture water area, a microbial community structure sensor, a temperature gradient change sensor, a phytoplankton density sensor in water, a heavy metal concentration sensor in water, a water transparency sensor, a water flow turbulence sensor, and a gas exchange rate sensor.

[0055] The marine noise level sensor is connected to the data acquisition system through an underwater cable and is used to monitor the underwater noise intensity. This marine noise level sensor is mainly used to evaluate the noise environment of the aquaculture water area. The noise intensity can affect the behavior and health of the aquaculture objects, especially for some sensitive marine aquaculture organisms. In practical applications, the marine noise level sensor can capture the underwater noise signal in real time and convert it into an electrical signal, which is transmitted to the data acquisition system through a cable connection. The data acquisition system can process the data to obtain the specific value of the underwater noise.

[0056] The trace element concentration sensor is connected to the data acquisition system through a digital interface and is used to monitor the concentration of trace elements in water in real time. Trace elements (such as iron, zinc, selenium, etc.) are beneficial to the growth and health of the aquaculture objects. This sensor helps to ensure that the water quality meets the physiological needs of the aquaculture objects by monitoring the concentration of trace elements in water. The sensor is connected to the data acquisition system through a high-precision digital interface to ensure real-time and efficient data transmission.

[0057] Aquaculture water depth change sensor, which is connected to the data acquisition system through an analog signal interface or a digital signal interface, is used to monitor the change of aquaculture water depth; this sensor is used to monitor the change of aquaculture water depth to help evaluate the change of water flow and its impact on the aquaculture environment. The change of water depth will affect the distribution of water flow and the uniformity of water quality, thus affecting the growth of aquaculture objects. By monitoring the depth change in real time, the sensor can provide data support to help aquaculture managers adjust the aquaculture environment to ensure that the depth change will not have a negative impact on the health of aquaculture objects.

[0058] Microbial community structure sensor, which is connected to the data acquisition system through an optical fiber interface, is used to monitor the types and quantities of microorganisms in water. This sensor is used to monitor the types and quantities of microorganisms in water, especially the ratio of beneficial and harmful microorganisms. The microbial community can affect water quality and the health of aquaculture objects. Through the optical fiber interface, the sensor can obtain relevant data of microorganisms in water in real time and transmit the data to the data acquisition system to analyze the microbial population structure in water, helping aquaculture managers understand the water quality status and thus make corresponding adjustments.

[0059] Temperature gradient change sensor, which is connected to the data acquisition system through a serial interface, is used to monitor the temperature difference between different water layers. The change of water temperature gradient in water will affect the vertical mixing of water and the oxygen distribution, and further affect the growth conditions of aquaculture objects. This sensor can monitor the temperature change between different depth water layers in real time and transmit the data to the data acquisition system through the serial interface. The system will evaluate the temperature distribution of water according to the real-time data and provide data support to obtain the thermal distribution status of water and take necessary measures to ensure the appropriate temperature.

[0060] Density sensor of phytoplankton in water, which is connected to the data acquisition system through an analog signal or a digital signal, is used to monitor the density of phytoplankton in water in real time and transmit the monitoring data to the data acquisition system. Phytoplankton is the main source of oxygen in water. This sensor ensures the balance and health of water quality by monitoring the quantity of phytoplankton in water in real time. The data acquisition system receives and processes the transmitted data and provides the density change trend of phytoplankton for aquaculture managers, which is conducive to adjusting water quality management measures to maintain a suitable aquaculture environment.

[0061] Heavy metal concentration sensor in water, which is connected to the data acquisition system through a digital signal interface, is used to monitor the heavy metal concentration in water. Heavy metals (such as lead, mercury, cadmium, etc.) are extremely harmful to the growth of aquaculture objects. Long-term exposure to high concentrations of heavy metals will cause aquaculture objects to get sick or die. This sensor can detect the heavy metal pollution in water in real time and transmit the data to the data acquisition system in real time through digital signal transmission for aquaculture personnel to take necessary measures, such as water quality filtration or water source replacement, to ensure the safety of the aquaculture environment.

[0062] The water transparency sensor is connected to the data acquisition system through a cable and is used to monitor the water transparency, which reflects the concentration of suspended substances in the water. A water body with lower transparency indicates more suspended particles in the water, which affects the photosynthesis of the aquaculture objects and the water quality. The transparency sensor transmits real-time data to the data acquisition system through the cable. The system can judge the clarity of the water quality based on this data and guide the aquaculture management personnel to take water purification measures to ensure the healthy growth of the aquaculture objects.

[0063] The water flow turbulence sensor is connected to the data acquisition system and is used to monitor the turbulence of the water flow. The turbulence affects the activity behavior and growth environment of the aquaculture objects. The turbulence has an important impact on the activity behavior of the aquaculture objects and their growth environment. Too high or too low turbulence will affect the swimming and food acquisition of the aquaculture objects. By monitoring the turbulence of the water flow in real time, this sensor provides real-time data to help the aquaculture management personnel understand the water flow conditions, so as to adjust the aquaculture environment to meet the activity needs of the aquaculture objects.

[0064] The gas exchange rate sensor is connected to the data acquisition system through an analog signal or digital signal interface and is used to monitor the gas exchange rate between the water surface and the air. The gas exchange rate is used to evaluate the aeration performance and dissolved oxygen level of the water body. The gas exchange rate is a key indicator for evaluating the dissolved oxygen level and aeration performance of the water body. The dissolved oxygen concentration directly affects the growth and health of the aquaculture objects. The sensor evaluates the aeration performance of the water body by detecting the exchange rate of oxygen and carbon dioxide. Through real-time monitoring, the data acquisition system can effectively evaluate the oxygen supply situation of the water body and provide timely adjustment suggestions for the aquaculture personnel to ensure that the oxygen level in the water is suitable for the healthy growth of the aquaculture objects.

[0065] As Figure 2 shown, in some embodiments, the growth data of the aquaculture objects in step S2 is obtained in the following manner:

[0066] S21: Record the weight of the aquaculture objects in real time through an underwater buoy weighing device; each aquaculture object is equipped with an identifier with a radio frequency identification tag. When the aquaculture object approaches the weighing device, the radio frequency identification tag is automatically activated and transmits information to the data acquisition system through a wireless signal to ensure accurate identification of the aquaculture object; the buoy weighing device consists of multiple buoys and load sensors. When the aquaculture object enters the buoy weighing device, the load sensors monitor the load change on the buoys in real time; the load sensors capture the load increment caused by the entry of the aquaculture object based on the buoyancy change of the buoys; calculate the weight of the aquaculture object by analyzing the data of the load sensors and transmit the weight data to the data acquisition system in real time;

[0067] S22: Use a camera to take pictures of the aquaculture objects to obtain images of the aquaculture objects, and combine image processing algorithms to obtain the body lengths of the aquaculture objects from the images of the aquaculture objects;

[0068] Specifically, first, the system installs multiple high-definition cameras in the aquaculture area. The cameras are arranged at different angles of the aquaculture pond to ensure that the aquaculture objects can be photographed from multiple perspectives. The cameras are protected by underwater waterproof enclosures to meet the shooting requirements of the underwater environment. The cameras take real-time pictures at fixed positions and transmit the image data to the data acquisition system through optical fibers or wireless signals.

[0069] Secondly, the received image data is first preliminarily processed by the data acquisition system, including improving image clarity, removing noise, etc. Subsequently, the image data will be transmitted to the image processing module, and the image processing algorithm starts to analyze the image. The specific image processing algorithm includes the following steps: In the image preprocessing step, perform denoising, enhancing contrast, adjusting brightness, etc. of the image to ensure that the aquaculture objects in the image are clearly visible. In the edge detection step, use an edge detection algorithm (such as the Canny edge detection algorithm) to extract the contours of the aquaculture objects. Through edge detection, the algorithm can accurately distinguish the edges of the aquaculture objects and extract the relevant features of the body length. In the feature extraction step, according to the edge information in the image, the algorithm further extracts the body length of the aquaculture objects. By comparing different feature points in the image, the algorithm determines the maximum longitudinal distance of the aquaculture objects, that is, the body length. In the size calibration step, in order to convert the pixel data in the image into actual physical dimensions, the system performs size calibration through a preset calibration object (such as a scale with a fixed size) to ensure that the measured body length in the image can be accurately converted into the actual body length of the aquaculture objects.

[0070] Finally, the body length data calculated by the image processing algorithm will be transmitted to the cloud platform in real time through the data acquisition system. The cloud platform further analyzes and stores the acquired data, and at the same time integrates it with other growth data (weight, morphological characteristics, etc.) to provide comprehensive growth data support for aquaculture management personnel.

[0071] S23: Obtain images of aquaculture objects in real time through a camera, use image recognition methods to analyze the images of aquaculture objects, automatically extract morphological characteristics, and compare the images at different growth stages to analyze the morphological changes of aquaculture objects;

[0072] Specifically, to ensure comprehensive monitoring of the morphological characteristics of the aquaculture objects, multiple high-definition cameras are installed at different positions in the aquaculture pond. The arrangement angles and heights of the cameras are ensured to be able to photograph the aquaculture objects in all directions, especially their main morphological characteristics such as body shape, dorsal fin, and tail fin. The cameras have high resolution and strong light compensation capabilities to ensure clear capture of the images of the aquaculture objects under different water quality and light conditions. The cameras are protected by underwater waterproof enclosures to ensure long-term stable operation of the cameras in the underwater environment. During the aquaculture process, the cameras continuously collect the images of the aquaculture objects in real time and transmit the image data to the data acquisition system. The data acquisition system sends the collected image data to the cloud platform through wireless or wired connections for subsequent processing and analysis.

[0073] After receiving the image data, the cloud platform preprocesses the images through the image processing module. First, image denoising, brightness, and contrast enhancement are performed to ensure that the outlines of the aquaculture objects in the images are clear and facilitate subsequent morphological feature extraction. This process uses denoising algorithms (such as Gaussian filtering) and image enhancement algorithms (such as histogram equalization), effectively improving the image quality.

[0074] The preprocessed images enter the image recognition algorithm for morphological feature extraction. The image recognition method mainly uses deep learning technologies such as convolutional neural networks (CNNs). Through the trained model, the aquaculture objects in the images are automatically recognized, and their main morphological characteristics are extracted. For example, the system can automatically extract the body length, dorsal fin shape, tail fin length, and other key parts of the aquaculture objects through algorithms. These morphological characteristics are important bases for evaluating the health and growth of the aquaculture objects. This process further accurately identifies the growth status of the aquaculture objects by comparing with the morphological characteristics of different aquaculture objects in the database. The system can judge whether there are deformities or abnormal growth conditions based on information such as the appearance and proportion of the aquaculture objects.

[0075] The system conducts comparative analysis based on the image data of the aquaculture objects at different growth stages to monitor the morphological changes of the aquaculture objects. By comparing the current images with the historical images, the growth pattern of the aquaculture objects is analyzed to check whether it conforms to the normal growth trend. Specifically, the system will compare the changes in the body shape, proportion, posture, etc. of the aquaculture objects to identify the subtle changes during their growth process. For example, during the growth process of the aquaculture objects, as the body length increases, the shape of the dorsal fin and the body proportion will change to a certain extent. The system can automatically detect these changes by comparing the images at different stages and evaluate the health status and growth quality of the aquaculture objects. If it is detected that the morphological changes of the aquaculture objects do not conform to the normal expectations, the system will automatically issue an alarm to prompt the aquaculture personnel to conduct further inspections or adjust the aquaculture management measures.

[0076] Through morphological change analysis, the key growth data of the aquaculture objects (such as body length, morphological feature changes, etc.) are transmitted to the data acquisition system and further uploaded to the cloud platform for storage and management. The cloud platform can generate growth reports of the aquaculture objects based on these data, including morphological change trend charts, warning information, etc. These data and reports can provide valuable information for aquaculture management personnel to make scientific decisions by the system, such as whether to adjust the feed formula, change the aquaculture environment or check the health status of the aquaculture objects.

[0077] S24: Monitor the activity frequency and movement range of the aquaculture objects through an infrared sensor, provide a real-time video stream through a video monitoring system, and analyze the video stream in combination with image analysis software to determine the swimming speed, swimming pattern and activity level of the aquaculture objects;

[0078] Specifically, in this embodiment, the infrared sensor is used to monitor the activity frequency and movement range of the aquaculture objects in real time. The infrared sensor detects the thermal radiation of objects in the water to identify the movement trajectory and activity range of the aquaculture objects. When the aquaculture objects move in the water, they emit specific infrared radiation signals, and the infrared sensor can capture these signals and convert them into electrical signals. The sensor continuously monitors the position and activity of the aquaculture objects, capturing their movement frequency, that is, the number of activities of the aquaculture objects within a certain period of time and the change of the movement range. The data of the sensor are transmitted to the data acquisition system in real time by wireless or wired means. The data acquisition system is responsible for recording the movement information of the aquaculture objects and storing it for subsequent analysis. This data is helpful to understand the activity level of the aquaculture objects and further analyze whether there are health problems.

[0079] To further improve the monitoring accuracy, this embodiment is also equipped with a video monitoring system, which consists of multiple high-definition cameras installed at different positions in the aquaculture area to ensure that the whole picture of the aquaculture objects can be captured from multiple perspectives. The cameras are protected by underwater waterproof enclosures and can adapt to the low light and clarity requirements in the underwater environment. The video monitoring system continuously collects real-time images of the aquaculture objects and transmits these image data to the data processing system in real time through the network. The data transmission uses a high-speed data transmission protocol to ensure that the video data can reach the processing module in time without affecting the real-time monitoring effect due to delay.

[0080] The received real-time video stream is processed by image analysis software, which can analyze the aquaculture objects in the video images through computer vision technology. The image analysis software first performs frame processing on the video stream, that is, extracts each frame of the image and processes it separately. Then, the system identifies the movement of the aquaculture objects in the video stream through image recognition algorithms (such as motion detection algorithms). To extract the swimming speed of the aquaculture objects, the system will track the position information of the aquaculture objects at multiple time points, calculate the displacement per unit time, and thus obtain the swimming speed. At the same time, through the changes between consecutive frames, the system can identify the swimming patterns of the aquaculture objects (such as uniform swimming, jumping, fast swimming, etc.). In addition, the image analysis software can also determine whether the aquaculture objects are in a static state, analyze their activity level, and calculate the activity index of the aquaculture objects during the monitoring period.

[0081] The data such as the swimming speed, swimming pattern, and activity level of the aquaculture objects obtained through the image analysis software are transmitted to the cloud platform in real time for further processing. The cloud platform will integrate these data with other environmental and growth data to generate a comprehensive analysis report. The report includes the movement trend, activity level change, and health risk of the aquaculture objects. By regularly monitoring the movement state of the aquaculture objects, the system can help aquaculture managers determine whether there are abnormal situations for the aquaculture objects. For example, a significant decrease in activity level indicates that the aquaculture objects are sick or stressed.

[0082] Combined with the real-time data of the infrared sensor and the video monitoring system, aquaculture managers can view the real-time movement of the aquaculture objects on the cloud platform. When it is found that the activity frequency of the aquaculture objects is abnormally low or the swimming pattern is abnormal, the system will automatically generate a warning message to prompt aquaculture managers to further check the health status of the aquaculture objects. In addition, according to the monitoring data, the system can provide optimization suggestions for the aquaculture environment, such as improving the water flow or adjusting the aquaculture density, to improve the activity level and growth health of the aquaculture objects.

[0083] S25: Non-invasively scan the body of the aquaculture object through an ultrasonic probe to obtain the fat layer thickness data, and calculate the body fat content of the aquaculture object according to the set mathematical model and the fat layer thickness data;

[0084] Specifically, in this embodiment, an ultrasonic probe is used to scan the body of the cultured object in a non-invasive manner. The ultrasonic probe uses high-frequency sound waves to penetrate the body surface of the cultured object, and obtains the thickness data of the fat layer of the cultured object through the reflection and echo detection of the sound waves. The ultrasonic probe can obtain the thickness of the fat layer in the body of the cultured object in real time and accurately without causing any harm to the cultured object. The probe is installed at a designated position in the culture pond to ensure that it can be scanned when the cultured object passes by, and the stability of the equipment in the underwater environment is ensured by the protective shell installed underwater. The ultrasonic probe transmits the collected signal data to the data acquisition system through wireless or wired transmission. The system uses the echo signal fed back by the probe to calculate the thickness of the fat layer. The probe emits an ultrasonic signal, penetrates the epidermis of the cultured object, and is reflected back to the probe after encountering the fat layer. The probe receives the reflected echo signal, converts it into a digital signal, and transmits it to the data acquisition system for processing.

[0085] The received ultrasonic signal data is parsed by the processing module of the data acquisition system. The system first uses the known sound speed and reflection time to convert the echo signal into the thickness data of the fat layer. Since the ultrasonic signal propagates at different speeds in different tissues, there are differences in the reflected echoes of the fat layer, muscle layer and other tissue layers. The system can accurately distinguish these tissues and calculate the thickness of the fat layer. This calculation method is based on the propagation characteristics of sound waves in the fat layer, combined with the specific size and type of the cultured object, to accurately calculate the actual thickness of the fat layer. The system can ensure the accuracy of the measurement through different scanning angles and positions to avoid errors caused by angle problems.

[0086] In this embodiment, the system uses a set mathematical model to calculate the body fat content of the farmed object based on the thickness data of the fat layer. The mathematical model establishes a relationship model between the thickness of the fat layer and the body fat content by studying the data of a large number of farmed objects. The relationship model takes into account factors such as the body shape, type and growth stage of the farmed object, and can calculate the fat content of the farmed object based on the thickness of the fat layer. The calculation formula for the fat content is based on experimental data and regression analysis, and estimates the overall body fat content through the relationship between the thickness of the fat layer and the proportion of fat in the body. This mathematical model provides the system with accurate fat content estimation capabilities, and can be dynamically adjusted according to the type and growth stage of the farmed object.

[0087] The above mathematical model can be fitted into the following form based on regression analysis or experimental data:

[0088] Fat content = a1·fat layer thickness + a2·body shape factor + a3·species factor + a4·growth stage factor + b; where a1, a2, a3, and a4 are regression coefficients obtained from experimental data, and body shape factor, species factor, and growth stage factor are quantitative descriptions based on the specific characteristics of the breeding object, corresponding to the body shape, species, and growth stage of the breeding object, respectively. b is a constant term used to correct the deviation of the mathematical model. This functional relationship is obtained by fitting experimental data, and the fat content in the body can be estimated based on different types of breeding objects, different growth stages (such as infancy, growth, etc.) and fat layer thickness.

[0089] The data of fat layer thickness and body fat content are transmitted to the cloud platform through the data acquisition system. The cloud platform receives, processes and stores this data in real time, and integrates and analyzes it together with other growth data (such as weight, body length, activity status, etc.). The cloud platform generates a growth report of the farmed object based on this data, providing a comprehensive assessment of the health status of the farmed object. Farming managers can view the fat content of each farmed object through the cloud platform and compare it with the normal growth standard. If the fat content of the farmed object is found to be abnormal (for example, too low or too high), the system will automatically issue an early warning to remind the farming manager that there is malnutrition, feed ratio problems or illness. Based on this data, the management system can adjust the amount, type or breeding environment of feed, optimize breeding management measures, and ensure the healthy growth of the farmed objects.

[0090] S26: Calculate the growth rate change by regularly recording the weight and length changes of the cultured objects;

[0091] S27: photographing the muscle parts of the breeding object through a camera, and extracting the muscle fiber density of the breeding object using an image processing algorithm.

[0092] Specifically, in this embodiment, in order to accurately obtain the muscle fiber density of the cultured object, a camera is installed at a specific position in the culture pond, specifically for photographing the muscle parts of the cultured object. Multiple high-resolution cameras are protected by an underwater waterproof housing to ensure stable operation in an underwater environment. These cameras are reasonably arranged to photograph the cultured object from multiple angles, especially focusing on its muscle parts, ensuring that the captured images are clear and can accurately display the details of the muscle tissue. When the cultured object is swimming or still, the camera will capture images of its muscle parts in real time. Through regular shooting, the system can obtain images of the muscle parts of the cultured object at different time points, providing data support for subsequent image processing and muscle fiber density extraction.

[0093] After image acquisition, the image data captured by the camera will be transmitted to the image processing system. The image processing system analyzes the acquired images through image processing algorithms to extract the characteristics of the muscle parts of the aquaculture objects. First, the system preprocesses the images, including denoising, enhancing contrast, and sharpening the images, etc., to ensure that the images of the muscle parts are clearer and the details are more prominent. This process uses image enhancement techniques, such as histogram equalization, edge detection, etc., to ensure that the detailed parts of the images (such as the structure and arrangement of muscle fibers) are accurately captured. Then, the image processing algorithm uses feature extraction techniques to extract the edge information of the muscle areas in the images through segmentation algorithms, and further analyzes the arrangement and density of muscle fibers. The algorithm identifies the structure of muscle fibers by detecting the texture features, pixel distribution, and color differences in the images, and extracts the eigenvalue related to the muscle fiber density. In this process, methods such as convolutional neural networks or texture-based analysis algorithms are used.

[0094] After extracting the muscle fiber characteristics, the system will calculate the muscle fiber density of the aquaculture object based on the preset calculation model and the muscle fiber data in the image. This density value is obtained by analyzing the number, arrangement, and occupied area of muscle fibers in the image. According to the extracted fiber characteristics, the algorithm will calculate the number of muscle fibers per unit area, thereby obtaining the muscle fiber density of the aquaculture object. To ensure the accuracy of the calculation, the system utilizes a trained machine learning model, which is trained based on a large amount of image data of aquaculture objects and can maintain a high accuracy under different lighting and water quality environments. Through this model, the system can provide reliable muscle fiber density values to help aquaculture management personnel evaluate the muscle development of aquaculture objects.

[0095] After calculating the muscle fiber density data, these data will be transmitted to the data acquisition system in real time and further uploaded to the cloud platform for storage and analysis. The cloud platform integrates and analyzes the muscle fiber density data with other aquaculture data (such as body weight, body length, activity status, etc.) to form a growth report of the aquaculture object. The cloud platform can analyze the muscle development of the aquaculture object through data comparison and compare it with the normal growth standards. If the muscle fiber density is abnormal, the system will automatically issue a warning to remind aquaculture management personnel to check the health status of the aquaculture object. These data can not only help evaluate the growth and development status of aquaculture objects, but also help optimize the aquaculture environment and feed management plan. When the muscle fiber density is low, the system recommends adjusting the protein and amino acid ratios in the feed or optimizing the aquaculture environment to promote muscle growth.

[0096] As Figure 3 shown, in some embodiments, step S3 specifically includes:

[0097] S31: The breeding environment data and growth data received by the cloud platform are first preprocessed, which includes removing missing values, correcting outliers, data normalization, and format unification;

[0098] Specifically, there may be some missing or incomplete values in the data, which will affect the accuracy of subsequent analysis. The cloud platform will remove the missing values according to the preset rules. During data collection, due to equipment failures, environmental interference, and other reasons, some data points will have outliers. The cloud platform detects the outliers in the data and corrects them to ensure the accuracy of the data. The correction methods include using the average value of adjacent data points to replace the outliers, or correcting them according to the distribution of historical data. The data collected by different types of sensors have different dimensions and ranges. The purpose of data normalization is to convert the data with different dimensions to a unified standard range. The normalization methods include linear normalization and Z-score standardization.

[0099] S32: The cloud platform integrates and processes the preprocessed breeding environment data and growth data to form a unified data set; the integration process includes: the cloud platform aligns different types of environmental data with the growth data of the breeding object according to the timestamp of each piece of data; the cloud platform annotates the time-aligned data according to the sensor type, data source, and breeding object identifier to ensure the accurate classification of the data;

[0100] Specifically, the breeding environment data and growth data come from different sensors, and their collection time points may not be consistent. For accurate analysis, the cloud platform will align different types of environmental data (such as temperature, humidity, etc.) with the growth data of the breeding object (such as weight, body length, etc.) according to the timestamp of each piece of data. The timestamp is the time mark of each data record.

[0101] Specifically, the data source refers to the location or system where the data is collected. Different data sources indicate that the environments or devices for data collection are different. In the breeding environment, the data sources can include different breeding ponds, different areas, or different breeding objects. The breeding object identifier is an ID that uniquely identifies each breeding object, such as an RFID tag or other identification methods. Each breeding object (such as fish, shellfish, etc.) has a unique identifier, which is used to track the growth data of the object. Through the breeding object identifier, the cloud platform can distinguish the growth data of the breeding object from other environmental data and ensure that the breeding object corresponding to each piece of data is unique.

[0102] S33: The cloud platform performs standardization processing on the integrated data. According to different data types and units, it performs numerical conversion to obtain standardized data.

[0103] Specifically, the data collected by different sensors have different units and ranges. Standardization processing is to make these data have a unified comparison standard. The cloud platform will perform numerical conversion according to the type and unit of each data. For example, convert the water temperature from Celsius to Fahrenheit, or convert the weight of the aquaculture object from kilograms to grams.

[0104] As Figure 4 shown, in some embodiments, step S4 specifically includes:

[0105] S41: The cloud platform uses historical data to train the deep learning model and optimize the parameters of the deep learning model.

[0106] In this embodiment, the cloud platform first trains the deep learning model by collecting and sorting out a large amount of historical data. The historical data includes past aquaculture environment data and the growth data of aquaculture objects, such as water temperature, dissolved oxygen, salinity, the weight, body length, and morphological changes of aquaculture objects. These historical data provide a large number of samples for training the deep learning model, enabling the model to learn the growth laws of aquaculture objects under different environmental conditions.

[0107] To optimize the accuracy of the model, the cloud platform uses this historical data for deep learning training and adopts a suitable neural network structure (such as Convolutional Neural Network CNN or Long Short-Term Memory Network LSTM) to model. During the training process, the model will continuously adjust its internal parameters (such as weights and biases) to minimize the error between the predicted value and the actual growth data. The optimization process usually uses the backpropagation algorithm and adjusts the model parameters through optimization techniques such as gradient descent method to improve the prediction accuracy.

[0108] The training data includes environmental variables and growth records of aquaculture objects in multiple aquaculture cycles, enabling the model to not only understand static data (such as water temperature) but also process dynamic environmental changes and their impacts on the growth of aquaculture objects. Through repeated training, the model gradually masters the complex correlations between data and can predict the future growth trends of aquaculture objects.

[0109] S42: After completing the training of the deep learning model, the cloud platform uses the cross-validation method or an independent test set to verify the effect of the deep learning model.

[0110] After completing the training, the cloud platform will use the cross-validation method or an independent test set to verify the effect of the deep learning model. Cross-validation is to divide the historical data set into multiple subsets. The data set can be divided into K parts, where K - 1 parts are used to train the model, and the remaining 1 part is used to test the effect of the model. This process will be repeated K times, and each time a different part is selected as the test set to ensure that the performance of the model on all data is stable and reliable.

[0111] The independent test set is a part of the data set divided from historical data, which is specifically used to test the generalization ability of the model. The data in the test set does not participate in the training of the model, ensuring that the test results can truly reflect the model's prediction ability for unknown data. Through these verification methods, the cloud platform can evaluate whether the trained deep learning model is overfitting and its prediction accuracy.

[0112] During the verification process, the cloud platform will adjust the training strategy according to the model's performance on the test set, perform hyperparameter optimization (such as learning rate, number of network layers, etc.) or add more sample data to further improve the model's generalization ability and prediction effect.

[0113] S43: The cloud platform inputs the latest aquaculture environment data and the growth data of aquaculture objects into the trained and verified deep learning model to obtain the predicted growth trend of aquaculture objects in the future for a period of time.

[0114] After the model is trained and verified, the cloud platform inputs the latest aquaculture environment data and the growth data of aquaculture objects collected in real time into the trained and verified deep learning model. The input data includes the current water temperature, salinity, dissolved oxygen concentration, and the latest growth data such as the weight and body length of aquaculture objects. These input data are compared and matched with the historical data patterns learned by the model, thereby generating the growth prediction of aquaculture objects in the future for a period of time.

[0115] The deep learning model will calculate the predicted growth trend of aquaculture objects in the future for a period of time based on these input data, including indicators such as weight, body length, and morphological changes. The deep learning model predicts the healthy growth trend of aquaculture objects by considering the current environmental state and the growth rules of aquaculture objects, and calculates the future growth rate and growth results.

[0116] This prediction result will help aquaculture management personnel make timely adjustment decisions. For example, if the prediction shows that the growth rate of aquaculture objects is slow in the next few days, the management personnel can adjust the aquaculture environment (water temperature, oxygen concentration, etc.) or change the feeding amount of feed according to the suggestions of the deep learning model to ensure that the aquaculture objects can maintain a good growth state.

[0117] As Figure 5 shown, in some embodiments, step S5 specifically includes:

[0118] S51: The cloud platform uses the data analysis model to perform risk assessment based on the predicted growth trend, the aquaculture environment data, and the growth data of the aquaculture objects, and identifies the existing risks.

[0119] In this embodiment, the cloud platform uses a deep learning model trained with historical data, as well as real-time collected aquaculture environment data and aquaculture object growth data, to conduct risk assessment through a data analysis model. Specifically, first, the cloud platform combines the predicted future growth trend of the aquaculture object by the deep learning model with the current aquaculture environment data and the actual growth data of the aquaculture object as input items. The aquaculture environment data includes water temperature, salinity, dissolved oxygen concentration, ammonia nitrogen concentration, etc., while the growth data of the aquaculture object includes weight, body length, morphological characteristics, etc.

[0120] The cloud platform uses a data analysis model (such as multiple regression analysis, decision tree, support vector machine, etc.) to evaluate these input data and identify risk factors affecting the healthy growth of aquaculture objects. For example, the system will analyze the impact of water temperature changes on the growth rate of aquaculture objects, or analyze the impact of dissolved oxygen concentration in water on the health of aquaculture objects. If it is found that the predicted growth trend of the aquaculture object does not match the current environmental conditions and there are environmental stress factors (such as too high temperature or water pollution), the system will identify the corresponding risk and mark it as a potential growth risk.

[0121] S52: According to the results of the risk assessment, the cloud platform generates warning information in real time.

[0122] Specifically, after completing the risk assessment, the cloud platform generates real-time warning information based on the assessment results. This process triggers the alarm mechanism in a timely manner based on the identified potential risks and current environmental changes. For example, if the cloud platform identifies that the water temperature is too high, or the growth rate of the aquaculture object is significantly lower than the predicted value, the system will generate warning information according to the preset rules to prompt the aquaculture management personnel of the existing risks.

[0123] The content of the warning information includes: risk type, which includes environmental risk, nutritional risk, disease risk, etc.; severity level of the risk, which includes minor, medium, and severe; time when the risk occurs, which refers to when the risk may occur according to the prediction to help management personnel make timely responses; recommended countermeasures, for example, if it is found that the temperature is too high, the system recommends reducing the water temperature or improving ventilation.

[0124] The cloud platform sends the warning information to the aquaculture management personnel in real time through various methods such as message notification, email, or SMS to ensure that they can obtain it in a timely manner and take necessary countermeasures.

[0125] S53: According to the warning information, the cloud platform adjusts the aquaculture environment and management measures.

[0126] After receiving the warning information, the cloud platform automatically or manually adjusts the aquaculture environment and management measures according to the recommended countermeasures to mitigate or eliminate risks. For example, when the cloud platform detects that the water temperature is too high, the system can automatically adjust the working state of the water pump in the aquaculture pond according to the preset rules, enhance the water flow, and promote cooling; or the system can increase the oxygen supply to improve the dissolved oxygen level in the water and avoid hypoxia of the aquaculture objects.

[0127] In addition, based on the warning information, the cloud platform can also recommend that the aquaculture management personnel adjust or automatically adjust the feed feeding amount or feeding frequency. For example, if the weight growth rate of the aquaculture objects is lower than the predicted value, the system recommends increasing the feed feeding amount to ensure that the nutritional needs of the aquaculture objects are met. For the situation of water pollution or microbial community imbalance, the cloud platform can prompt for water purification or drug addition treatment to ensure that the water quality returns to normal.

[0128] After adjusting the environment and management measures, the cloud platform will continue to monitor the state of the aquaculture environment and aquaculture objects, and evaluate the effect of the adjustment measures according to the newly obtained data. If the adjustment measures fail to solve the problem, the cloud platform will issue a new warning and put forward further optimization suggestions again, forming a closed-loop dynamic adjustment process to ensure the healthy growth of the aquaculture objects.

[0129] As Figure 6 shown, in some embodiments, step S53 specifically includes:

[0130] S531: According to the warning information, the cloud platform monitors the aquaculture environment parameters in real time and adjusts the aquaculture environment according to the warning result; the adjustment content includes any one or more of the following: adjusting the water flow rate, controlling the water temperature, optimizing the water quality, and adjusting the oxygen concentration.

[0131] In this embodiment, the cloud platform monitors multiple key parameters in the aquaculture environment in real time according to the generated warning information, such as water flow rate, water temperature, water quality, and dissolved oxygen concentration. If the warning information indicates an environmental risk (such as too high water temperature or water pollution), the cloud platform will automatically initiate the corresponding adjustment measures.

[0132] For example, if the system monitors that the water temperature exceeds the set safety range, the cloud platform will adjust the operation of the water pump or fan through the intelligent control system to increase the water flow and air circulation, so as to achieve the cooling effect; if the water quality detection shows that the concentration of harmful substances in the water is too high (such as too high ammonia nitrogen concentration), the system can start the water treatment equipment, such as an aerator, a filtration device, or a water purification device, to improve the water quality.

[0133] In addition, the cloud platform will automatically adjust the oxygen supply according to the change of oxygen concentration. For example, when the dissolved oxygen concentration in the water is too low, the cloud platform will increase the oxygen content in the water by automatically controlling the air pump to avoid slow growth or health problems of the aquaculture objects due to lack of oxygen.

[0134] S532: According to the warning information, the cloud platform automatically adjusts the aquaculture management measures, which include feed feeding amount, feeding frequency, aquaculture density, and disease prevention plan.

[0135] In addition to the automatic adjustment of the environment, the cloud platform will also automatically optimize the aquaculture management measures according to the warning information to ensure that the aquaculture objects can be fully taken care of under different circumstances. For example, if the system analyzes the growth data of the aquaculture objects and finds that their growth rate slows down, or the warning information shows that the water temperature or oxygen concentration is not within the ideal range, the cloud platform can automatically adjust the feeding amount and frequency of the feed to meet the nutritional needs of the aquaculture objects and promote their healthy growth.

[0136] In some cases, the cloud platform can also adjust the aquaculture density based on environmental changes. For example, when the water temperature is too high or the dissolved oxygen concentration is insufficient, the cloud platform recommends reducing the aquaculture density and reducing the number of aquaculture objects per unit water body to reduce the environmental burden and optimize the aquaculture conditions. In addition, the cloud platform can also automatically adjust the disease prevention plan according to the risk assessment results. If the warning information shows that there are problems with the water quality or signs of certain diseases, the cloud platform will automatically start the corresponding prevention and control measures, such as adding drugs or starting the disinfection program.

[0137] These automatic adjustments can quickly respond to changes in the aquaculture environment and objects, reduce the delay of manual operations, and improve the accuracy and efficiency of marine aquaculture management.

[0138] S533: The cloud platform will evaluate the effect of the adjustment measures according to the feedback of the aquaculture environment and the status of the aquaculture objects, and decide whether further adjustment is needed to ensure the healthy growth of the aquaculture objects.

[0139] Specifically, after the environment and management are adjusted, the cloud platform will continuously monitor the status of the aquaculture environment and aquaculture objects to evaluate the effect of the adjustment measures. By receiving real-time aquaculture environment parameters (such as water temperature, dissolved oxygen concentration, ammonia nitrogen concentration, etc.) and the growth data of aquaculture objects (such as weight, body length, health status, etc.), the cloud platform can analyze the change trends of these data and judge whether the adjustment measures have achieved the expected effect.

[0140] For example, if the adjusted water flow rate and oxygen concentration have returned to the normal range, and the weight gain of the aquaculture objects has recovered to the expected level, the system will consider the adjustment successful. However, if the data indicates that the health status of the aquaculture objects has not improved, the cloud platform will continue to optimize the adjustment measures, such as increasing the operating time of the water treatment equipment, further adjusting the water flow, or increasing the feed feeding amount.

[0141] Through such a closed-loop adjustment process, the cloud platform ensures that the aquaculture objects always maintain a healthy state in the dynamic changes of the aquaculture environment, thereby improving the aquaculture efficiency and the growth quality of the aquaculture objects.

[0142] Embodiment 2

[0143] As Figure 7 shown, this embodiment provides an intelligent growth monitoring and analysis system for marine aquaculture, which includes:

[0144] The first sensor group is used to collect aquaculture environment data in real time;

[0145] The second sensor group is used to obtain the growth data of the aquaculture objects in real time;

[0146] The data acquisition system is used to receive the aquaculture environment data collected by the first sensor group and the growth data of the aquaculture objects collected by the second sensor group, and transmit them to the cloud platform;

[0147] The cloud platform is used to receive the aquaculture environment data and the growth data of the aquaculture objects, perform data processing, and generate standardized data;

[0148] The deep learning analysis module is located on the cloud platform and is used to analyze and predict the growth trend of the aquaculture objects based on the standardized data through deep learning algorithms, and obtain the predicted growth trend of the aquaculture objects in a future period of time;

[0149] The risk analysis module is located on the cloud platform and is used to perform risk assessment and generate real-time warning information according to the predicted growth trend, the aquaculture environment data, and the growth data of the aquaculture objects;

[0150] The adjustment module is located on the cloud platform and is used to adjust the aquaculture environment and aquaculture management measures according to the warning information to ensure the healthy growth of the aquaculture objects.

[0151] In some embodiments, the first sensor group includes any of the following sensors connected to the data acquisition system:

[0152] The marine noise level sensor is used to monitor the underwater noise intensity;

[0153] The trace element concentration sensor is used to monitor the trace element concentration in the water;

[0154] Aquaculture water depth sensor, used to monitor the depth of aquaculture water;

[0155] Microbial community structure sensor, used to monitor the types and quantities of microorganisms in water;

[0156] Temperature gradient sensor, used to monitor the temperature differences in different water layers;

[0157] Phytoplankton density sensor in water, used to monitor the phytoplankton density in water and transmit the monitoring data to the data acquisition system;

[0158] Heavy metal concentration sensor in water, used to monitor the heavy metal concentration in water;

[0159] Water transparency sensor, used to monitor the water transparency, and the water transparency reflects the concentration of suspended substances in water;

[0160] Water flow turbulence sensor, used to monitor the turbulence of water flow, and the turbulence affects the activity behavior and growth environment of aquaculture objects;

[0161] Gas exchange rate sensor, used to monitor the gas exchange rate between the water surface and the air, and the gas exchange rate is used to evaluate the ventilation performance and dissolved oxygen level of water.

[0162] In some embodiments, the second sensor group includes any of the following multiple sensors connected to the data acquisition system:

[0163] Underwater buoy weighing device, used to record the weight of aquaculture objects in real time;

[0164] Camera, used to take images of aquaculture objects and extract the body length of aquaculture objects in combination with image processing algorithms;

[0165] Image recognition device, used to analyze the images of aquaculture objects, automatically extract morphological features, and analyze morphological changes;

[0166] Infrared sensor, used to monitor the activity frequency and movement range of aquaculture objects;

[0167] Video monitoring system, used to provide a real-time video stream and analyze the swimming speed, swimming mode, and activity level of aquaculture objects in combination with image analysis software;

[0168] Ultrasonic probe, used to scan aquaculture objects and obtain fat layer thickness data for calculating the body fat content;

[0169] Regular recording device, used to record the weight changes and body length changes of aquaculture objects to calculate the growth rate changes;

[0170] Muscle fiber density extraction device, used to extract the muscle fiber density of aquaculture objects.

[0171] In some embodiments, the cloud platform includes:

[0172] A data processing module, configured to receive and process aquaculture environment data and aquaculture object growth data, perform data cleaning, preprocessing, time alignment, and annotation processing to ensure accurate classification of the data;

[0173] A deep learning analysis module, configured to train a deep learning algorithm based on the standardized data, analyze the growth trend of the aquaculture object, and predict future growth conditions;

[0174] A risk analysis module, configured to perform real-time risk assessment based on the predicted growth trend, aquaculture environment data, and growth data, and generate warning information;

[0175] An adjustment module, configured to adjust the aquaculture environment and aquaculture management measures according to the warning information to ensure the healthy growth of the aquaculture object.

[0176] In some embodiments, the data processing module further includes:

[0177] A data cleaning and preprocessing sub-module, configured to remove missing data, correct abnormal data, and perform standardization processing;

[0178] A data integration sub-module, configured to align different types of data according to timestamps and perform annotation classification on the data.

[0179] In some embodiments, the deep learning analysis module includes:

[0180] A model training module, configured to train a deep learning model using historical data to optimize the prediction ability of the model;

[0181] A verification module, configured to verify the effect of the trained deep learning model using a test set or cross-validation method.

[0182] In some embodiments, the risk analysis module includes:

[0183] A risk identification sub-module, configured to identify risks related to the growth trend of the aquaculture object and environmental data;

[0184] A warning information generation sub-module, configured to generate real-time warning information according to the risk assessment result.

[0185] In some embodiments, the adjustment module includes:

[0186] An environment adjustment module, configured to monitor and adjust the parameters of the aquaculture environment in real time, including adjusting the water flow rate, controlling the water temperature, optimizing the water quality, etc.;

[0187] A management adjustment module for automatically or manually adjusting aquaculture management measures according to early warning information, including adjusting feed input amount, breeding density, disease prevention, etc.

[0188] Those skilled in the art can clearly understand that, for the convenience and conciseness of description, only the above-mentioned division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated here.

[0189] The embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements any one of the above methods.

[0190] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above method embodiments of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc. Of course, there are other ways of readable storage media, such as quantum memory, graphene memory, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0191] The present invention also provides an electronic device. The electronic device according to an embodiment of the present invention includes: one or more processors; a storage device for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement the method provided by the present invention.

[0192] Reference is made below Figure 8 to FIG., which shows a schematic structural diagram of a computer system 800 suitable for implementing the electronic device according to an embodiment of the present invention. Figure 8 The electronic device shown is merely an example and should not impose any limitation on the functions and scope of use of the embodiments of the present invention.

[0193] As Figure 8 shown, the computer system 800 includes a central processing unit (CPU) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage section 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the computer system 800 are also stored. The CPU 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0194] The following components are connected to the I / O interface 805: an input section 806 including a keyboard, a mouse, etc.; an output section 807 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section 808 including a hard disk, etc.; and a communication section 809 including a network interface card such as a LAN card, a modem, etc. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the I / O interface 805 as required. A removable medium 811, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 810 as required so that a computer program read therefrom is installed into the storage section 808 as required.

[0195] It should be noted that the computer-readable medium shown in the present invention can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the above two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of a computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device.

[0196] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent growth monitoring and analysis of marine aquaculture, characterized in that: The method comprises the following steps: S1: Collecting aquaculture environment data in real time through the first sensor group; S2: Acquire the growth data of the breeding object in real time through the second sensor group; S3: transmitting the breeding environment data and the growth data of the breeding objects to a cloud platform through a data acquisition system, and the cloud platform processes the data to obtain standardized data; S4: Based on the standardized data, the growth trend of the breeding objects is analyzed and predicted by a deep learning algorithm to obtain a predicted growth trend of the breeding objects in the future; S5: Based on the predicted growth trend, combined with the breeding environment data and the growth data of the breeding objects, risk assessment is performed to generate early warning information in real time; the breeding environment and breeding management measures are adjusted according to the early warning information to ensure the healthy growth of the breeding objects.

2. The method according to claim 1, characterized in that The aquaculture environment data include any one of the following: ocean noise level, trace element concentration, depth change of aquaculture waters, microbial community structure, temperature gradient change, phytoplankton density in water, heavy metal concentration in water bodies, water transparency, water flow turbulence and gas exchange rate; The growth data of the cultured object includes any one of body weight, body length, morphological characteristics, activity status, body fat content, growth rate changes and muscle fiber density.

3. The method according to claim 2, characterized in that The first sensor group includes any number of the following sensors connected to the data acquisition system: Ocean noise level sensor, used to monitor underwater noise intensity; Trace element concentration sensor, used to monitor the concentration of trace elements in water; Aquaculture water depth change sensor, used to monitor the depth change of aquaculture waters; Microbial community structure sensor, used to monitor the types and quantities of microorganisms in water; Temperature gradient change sensor, used to monitor the temperature difference between different water layers; A phytoplankton density sensor in water is used to monitor the phytoplankton density in water and transmit the monitoring data to a data acquisition system; Water body heavy metal concentration sensor, used to monitor the heavy metal concentration in water; Water transparency sensors, used to monitor water transparency, which indicates the concentration of suspended matter in the water; Water flow turbulence sensor, used to monitor the turbulence of water flow, which affects the activity behavior and growth environment of the aquaculture objects; Gas exchange rate sensor, used to monitor the gas exchange rate between the water surface and the air, said gas exchange rate is used to evaluate the aeration performance and dissolved oxygen level of the water body.

4. The method according to claim 2, characterized in that: The growth data of the cultured object in step S2 is obtained in the following manner: S21: Record the weight of the cultured object in real time through an underwater buoy weighing device; S22: photographing the breeding object with a camera to obtain an image of the breeding object, and obtaining the body length of the breeding object from the image of the breeding object using an image processing algorithm; S23: acquiring images of the farming objects in real time through a camera, analyzing the images of the farming objects using an image recognition method, automatically extracting morphological features, and comparing images of different growth stages to obtain morphological changes of the farming objects; S24: Monitor the activity frequency and movement range of the cultured object through infrared sensors, provide basic movement signals, and provide real-time video streams through video monitoring systems. Combined with image analysis software, analyze the video streams to further determine the swimming speed, swimming pattern and activity level of the cultured object; S25: scanning the cultured object with an ultrasonic probe to obtain fat layer thickness data, and calculating the body fat content of the cultured object according to a set mathematical model and the fat layer thickness data; S26: Calculate the growth rate change by regularly recording the weight change and body length change of the cultured object; S27: photographing the muscle parts of the breeding object through a camera, and extracting the muscle fiber density of the breeding object using an image processing algorithm.

5. The method according to claim 1, characterized in that Step S3 specifically includes: S31: The cloud platform preprocesses the received breeding environment data and growth data, including removing missing values, correcting outliers, normalizing data, and unifying formats; S32: The cloud platform integrates the pre-processed aquaculture environment data and growth data to form a unified data set; the integration process includes: the cloud platform time-aligns different types of environmental data with the growth data of the aquaculture object according to the timestamp of each data; the cloud platform labels the time-aligned data according to the sensor type, data source and aquaculture object identifier, thereby identifying the data into different categories; S33: The cloud platform performs standardization on the integrated data, converting values ​​according to different data types and units to obtain standardized data.

6. The method according to claim 1, characterized in that Step S4 specifically includes: S41: The cloud platform uses historical data to train the deep learning model and optimize the parameters of the deep learning model; S42: After completing the deep learning model training, the cloud platform uses a cross-validation method or an independent test set to verify the effect of the deep learning model; S43: The cloud platform inputs the latest breeding environment data and the growth data of the breeding objects into the trained and verified deep learning model to obtain the predicted growth trend of the breeding objects in the future.

7. The method according to claim 1, characterized in that Step S5 specifically includes: S51: The cloud platform uses a data analysis model to perform risk assessment based on the predicted growth trend, the breeding environment data, and the growth data of the breeding object, and identifies the existing risks; S52: Based on the results of the risk assessment, the cloud platform generates early warning information in real time; S53: According to the early warning information, the cloud platform adjusts the breeding environment and management measures.

8. The method according to claim 7, characterized in that Step S53 specifically includes: S531: According to the warning information, the cloud platform monitors the aquaculture environment parameters in real time, and adjusts the aquaculture environment according to the warning results; the adjustment content includes any one or more of the following: adjusting water flow speed, controlling water temperature, optimizing water quality, and adjusting oxygen concentration; S532: According to the early warning information, the cloud platform automatically adjusts the breeding management measures, including any one or more of the feed amount, feeding frequency, breeding density, and disease prevention plan; S533: The cloud platform will evaluate the effectiveness of adjustment measures based on the feedback of the breeding environment and the status of the breeding objects, and decide whether further adjustments are needed to ensure the healthy growth of the breeding objects.

9. The method according to claim 1, characterized in that: The cultured objects include: one or more of fish, shellfish, and seaweed.

10. An intelligent growth monitoring and analysis system for marine aquaculture, characterized in that: include: The first sensor group is used to collect breeding environment data in real time; The second sensor group is used to obtain the growth data of the breeding objects in real time; A data acquisition system, used for receiving the aquaculture environment data collected by the first sensor group and the aquaculture object growth data collected by the second sensor group, and transmitting them to the cloud platform; A cloud platform, used to receive the breeding environment data and the growth data of the breeding objects, perform data processing, and generate standardized data; A deep learning analysis module, located on the cloud platform, is used to analyze and predict the growth trend of the breeding objects through a deep learning algorithm based on the standardized data, and obtain the predicted growth trend of the breeding objects in the future; A risk analysis module, located on the cloud platform, is used to perform risk assessment and generate real-time warning information based on the predicted growth trend, the breeding environment data and the growth data of the breeding object; The adjustment module is located on the cloud platform and is used to adjust the breeding environment and breeding management measures according to the early warning information to ensure the healthy growth of the breeding objects.

Images