Shrimp swarm health status assessment method based on deep clustering analysis

Abstract

The present invention relates to the technical field of data processing, and particularly to a method for evaluating the health status of shrimp populations based on deep clustering analysis. The method includes: monitoring health parameters; determining the shrimp population dispersion grid; determining the temporary grid; identifying non-shrimp population abnormal grids; calculating the actual health index; adjusting the grid step size and evaluating health. By combining physiological and behavioral data such as real-time density, saturation, and activity, the present invention can accurately monitor and evaluate the health status of shrimp populations. By gradually excluding abnormal grids and screening out more representative grids, it avoids evaluation biases caused by environmental fluctuations or measurement errors, not only realizes the dynamic monitoring of the health status of shrimp populations, but also automatically adjusts the grid step size through the comparison between the ideal health index and the actual health index, thereby improving the accuracy and reliability of health evaluation, and effectively solving the problem of inaccurate health evaluation results due to unstable environmental parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method for evaluating the health status of a shrimp group based on deep clustering analysis. Background Art

[0002] With the increasing demand for biological asset management, how to achieve efficient, accurate, and real-time evaluation of the health status of biological groups has become an urgent problem to be solved. Especially when facing complex environmental changes and various internal and external factors, how to accurately monitor and evaluate the health status of biological groups through automated technology is an important challenge for achieving refined management and optimizing resource allocation.

[0003] The patent document with the publication number CN117636146A discloses a smart identification and evaluation system and method for inventorying live shrimp assets. The method includes: capturing shrimp images under various lighting conditions; processing and identifying the shrimp activity images through a miniature biological body shape recognition and counting algorithm to accurately identify the shape, size, color, species, and quantity information of the shrimp; evaluating the live shrimp assets according to the shape, size, color, species, quantity information of the shrimp and environmental parameters; sending an alarm to the user when the health status or quantity of the shrimp is abnormal; integrating, analyzing, and storing the shrimp information and the evaluation results of the live shrimp assets.

[0004] It can be seen that the smart identification and evaluation method for inventorying live shrimp assets has the following problems: This method relies on capturing shrimp images under various lighting conditions, and the change of lighting affects the image quality, resulting in recognition errors or inaccurate evaluation results. In low-light or strong-light environments, the accuracy of image processing is significantly affected, leading to data deviation; when evaluating live shrimp assets, this method relies on environmental parameters such as water quality and temperature, resulting in the evaluation results being greatly affected by environmental changes, and it is difficult to obtain stable and consistent evaluation results especially under extreme environmental conditions. Summary of the Invention

[0005] For this reason, the present invention provides a method for evaluating the health status of a shrimp group based on deep clustering analysis to overcome the problem of inaccurate health evaluation results caused by unstable environmental parameters in the prior art through deep clustering analysis.

[0006] To achieve the above object, the present invention provides a method for evaluating the health status of a shrimp group based on deep clustering analysis, including:

[0007] Monitoring the real-time density, real-time saturation, and real-time activity of the shrimp group in each grid to be evaluated in the monitoring area constructed based on a preset grid step size;

[0008] Determining a number of shrimp group dispersion grids according to the real-time density;

[0009] Determine a number of temporary grids according to the real-time saturation of the shrimp swarm dispersion grid within a preset determined duration;

[0010] Determine a number of non-shrimp swarm abnormal grids according to the real-time activity and the real-time density of any two adjacent temporary grids;

[0011] Determine the actual shrimp swarm health index according to the real-time density and the real-time saturation of the non-shrimp swarm abnormal grids;

[0012] Determine the ideal shrimp swarm health index according to a preset clustering analysis model, the real-time density, the real-time saturation, and the real-time activity of each grid to be evaluated;

[0013] Adjust the preset grid step size according to the ideal shrimp swarm health index and the actual shrimp swarm health index;

[0014] Determine the actual shrimp swarm health index based on the grid step size to form a shrimp swarm health assessment value.

[0015] Further, the determining a number of non-shrimp swarm abnormal grids according to the real-time activity and the real-time density of any two adjacent temporary grids includes:

[0016] Calculate the difference between the two real-time densities to form a density change value;

[0017] Calculate the difference between the two real-time activities to form an activity change value;

[0018] Determine a change consistency according to the density change value and the activity change value;

[0019] Determine a number of shrimp swarm abnormal grids according to the change consistency and the real-time density;

[0020] Exclude the shrimp swarm abnormal grids from all the grids to be evaluated to determine a number of the non-shrimp swarm abnormal grids.

[0021] Further, the determining a change consistency according to the density change value and the activity change value includes:

[0022] Draw a density change curve according to the density change value within a preset drawing duration;

[0023] Draw an activity change curve according to the activity change value within the preset drawing duration;

[0024] Calculate the cosine similarity of the density change curve and the activity change curve to form a change consistency.

[0025] Further, determining a number of abnormal shrimp swarm grids according to the change consistency and the real-time density includes:

[0026] When the change consistency is less than a preset consistency threshold, determining that the two adjacent temporary grids are difference grids, and forming a number of difference grids;

[0027] Determining a number of abnormal shrimp swarm grids according to the real-time density of the difference grids.

[0028] Further, determining a number of abnormal shrimp swarm grids according to the real-time density of the difference grids includes:

[0029] Calculating the standard deviation of the difference between the real-time densities of any two adjacent difference grids to form a difference density fluctuation value;

[0030] When the difference density fluctuation value is greater than a preset difference fluctuation threshold, determining that the difference grid is the abnormal shrimp swarm grid, and forming a number of abnormal shrimp swarm grids.

[0031] Further, determining the actual shrimp swarm health index according to the real-time density and the real-time saturation of the non-abnormal shrimp swarm grids includes:

[0032] Performing standardization processing on the real-time density to form a standard density;

[0033] Performing standardization processing on the real-time saturation to form a standard saturation;

[0034] Performing weighted summation on the standard density, the standard saturation, a preset density weight, and a preset saturation weight to form an actual shrimp swarm health index.

[0035] Further, adjusting the preset grid step according to the ideal shrimp swarm health index and the actual shrimp swarm health index includes:

[0036] Calculating the relative deviation between the ideal shrimp swarm health index and the actual shrimp swarm health index to form an index deviation;

[0037] Adjusting the preset grid step according to the index deviation.

[0038] Further, adjusting the preset grid step according to the index deviation includes:

[0039] Calculating the standard deviation of the index deviation within a preset adjustment duration to form a deviation fluctuation value;

[0040] When the deviation fluctuation value is greater than a preset deviation fluctuation value threshold, reducing the preset grid step according to the relative deviation between the deviation fluctuation value and the preset deviation fluctuation value threshold and a preset adjustment coefficient.

[0041] Further, the determining a number of shrimp swarm dispersion grids according to the real-time density includes:

[0042] When the real-time density is less than a preset density threshold, it is determined that the grid to be evaluated is a shrimp swarm dispersion grid, and a number of shrimp swarm dispersion grids are formed.

[0043] Further, the determining a number of temporary grids according to the real-time saturation degree of the shrimp swarm dispersion grids includes:

[0044] Calculate the standard deviation of the real-time saturation degree to form a saturation degree fluctuation value;

[0045] When the saturation degree fluctuation value is greater than a preset saturation degree fluctuation value threshold, it is determined that the shrimp swarm dispersion grid is the temporary grid, and a number of temporary grids are formed.

[0046] Compared with the prior art, the beneficial effects of the present invention are as follows: by combining physiological and behavioral data such as real-time density, saturation degree, and activity, the health status of the shrimp swarm can be accurately monitored and evaluated. By gradually excluding abnormal grids, more representative grids are screened out, avoiding evaluation biases caused by environmental fluctuations or measurement errors. It not only realizes the dynamic monitoring of the health status of the shrimp swarm, but also automatically adjusts the grid step size through the comparison of the ideal health index and the actual health index, thereby improving the accuracy and reliability of the health assessment. In addition, the adaptive model based on deep clustering analysis can process complex and variable monitoring data, has good robustness and flexibility, and effectively solves the problem of inaccurate health assessment results caused by unstable environmental parameters.

[0047] Further, by calculating the density and activity differences between adjacent grids and combining the judgment of the change consistency, non-shrimp swarm abnormal grids that do not meet the health standards can be accurately identified. This method can effectively exclude abnormal data and ensure the accuracy of the health assessment.

[0048] Further, by calculating the cosine similarity of the density and activity change curves, abnormal situations with poor change synchronization can be effectively identified. By dynamically comparing the changes in density and activity, potential abnormal conditions can be captured more sensitively, providing higher reliability and response speed for the health monitoring of the shrimp swarm.

[0049] Further, by dynamically judging the change consistency and analyzing the real-time density of the difference grids, areas with abnormal health status can be accurately identified, improving the determination accuracy of abnormal grids and avoiding misjudgment caused by accidental fluctuations.

[0050] Furthermore, by calculating the differential density fluctuation value, the grid regions with abnormal density fluctuations are effectively identified, thus accurately capturing the shrimp groups with large fluctuations in health status, helping to exclude misjudgments caused by natural fluctuations or short-term changes, and improving the accuracy of health assessment and the sensitivity of the system.

[0051] Furthermore, through standardization and weighting, the health status of the shrimp group is comprehensively evaluated, effectively avoiding errors caused by fluctuations in a single index. At the same time, the impacts of density and saturation can be comprehensively considered to provide a more comprehensive and accurate health index.

[0052] Furthermore, by adjusting the preset grid step size according to the index deviation, the grid division can be dynamically optimized, improving the sensitivity and accuracy of health assessment.

[0053] Furthermore, by dynamically adjusting the preset grid step size, it is possible to effectively respond to fluctuations in the health index, reduce the instability caused by environmental changes or measurement errors, ensure more refined grid division, and make the health assessment results more accurate.

[0054] Furthermore, by determining the shrimp group dispersion grid according to the real-time density, the regions with low density can be effectively identified, thus accurately dividing the shrimp group dispersion grid, which helps to optimize the health assessment process, ensure more accurate assessment results, and avoid the impact of low-density regions on the overall health judgment.

[0055] Furthermore, by introducing the judgment of the saturation fluctuation value, the grids with large fluctuations in health status can be effectively identified, and then the grid division can be adjusted to ensure the accuracy and meticulousness of the assessment, which helps to improve the timeliness and accuracy of shrimp group health monitoring. Description of the Drawings

[0056] Figure 1 It is the flowchart of the method for assessing the health status of shrimp groups based on deep clustering analysis in this embodiment;

[0057] Figure 2 It is the decision logic diagram for determining the abnormal grids of shrimp groups in this embodiment;

[0058] Figure 3 It is the decision logic diagram for adjusting the preset grid step size in this embodiment;

[0059] Figure 4 It is the decision logic diagram for determining the shrimp group dispersion grid in this embodiment. Detailed Implementation Manner

[0060] In order to make the objectives and advantages of the present invention clearer, the present invention will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0061] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present invention and do not limit the protection scope of the present invention.

[0062] It should be noted that in the description of the present invention, the terms indicating the direction or positional relationship such as "upper", "lower", "left", "right", "inner", "outer", etc. are based on the direction or positional relationship shown in the drawings. This is only for convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of the present invention.

[0063] In addition, it should also be noted that in the description of the present invention, unless otherwise clearly specified and defined, the terms "installation", "connection", "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0064] Please refer to Figure 1 as shown, which is a flowchart of the method for evaluating the health status of shrimp populations based on deep clustering analysis in this embodiment;

[0065] This embodiment provides a method for evaluating the health status of shrimp populations based on deep clustering analysis, including:

[0066] Monitoring the real-time density, real-time saturation, and real-time activity of shrimp populations in each grid to be evaluated within the monitoring area constructed based on a preset grid step size;

[0067] Determining a number of shrimp population dispersion grids according to the real-time density;

[0068] Determining a number of temporary grids according to the real-time saturation of the shrimp population dispersion grids within a preset determination time period;

[0069] Determining a number of non-shrimp population abnormal grids according to the real-time activity and real-time density of any two adjacent temporary grids;

[0070] Determining the actual shrimp population health index according to the real-time density and real-time saturation of the non-shrimp population abnormal grids;

[0071] Determining the ideal shrimp population health index according to a preset clustering analysis model, the real-time density, real-time saturation, and real-time activity of each grid to be evaluated;

[0072] Adjust the preset grid step according to the ideal shrimp swarm health index and the actual shrimp swarm health index;

[0073] Determine the actual shrimp swarm health index based on the grid step to form a shrimp swarm health assessment value.

[0074] The preset grid step refers to the step for dividing each grid in the monitoring area, which is used to refine the distribution and sampling density of the monitoring area. It depends on the scale of the monitoring area, the resolution of the monitoring equipment, and the required evaluation accuracy. It is usually set between 1 meter and 10 meters. In this embodiment, it is set to 5 meters, which can balance accuracy and computational efficiency, ensure that the dynamic changes of the shrimp swarm can be effectively captured, and avoid excessive computational complexity caused by over-refinement.

[0075] The preset determination duration refers to the time window set during the monitoring process to obtain stable health assessment data. It depends on the activity cycle of the shrimp swarm, the frequency of health changes, and the acquisition frequency of monitoring data. It is usually set between 1 hour and 24 hours. In this embodiment, it is set to 6 hours, which can balance the timeliness and accuracy of the data, effectively reflect the health change trend of the shrimp swarm, reduce the interference caused by short-term fluctuations, and improve the stability of the evaluation results.

[0076] The real-time density refers to the number density of the shrimp swarm in each monitoring grid. The position and number of the shrimp swarm in the monitoring area are captured in real time through an underwater camera or a laser scanning device. Through an image processing algorithm, the number of shrimps in each grid is identified and calculated, and then the density value is calculated in combination with the area of the grid.

[0077] The real-time saturation reflects the density or spatial occupancy degree of the shrimp swarm at a certain time point, which is usually obtained through an optical sensor or a sonar system. By analyzing the spatial distribution of the shrimp swarm in the grid area, the saturation index is calculated. The saturation is usually expressed as the ratio of the occupied area to the grid area.

[0078] The real-time activity measures the activity frequency or movement intensity of the shrimp swarm within a specified time period, which is achieved by randomly sampling several shrimps in each grid of the shrimp swarm on the water surface and combining video monitoring with a motion detection algorithm. By analyzing the movement speed and change trajectory of the shrimp swarm, its activity is calculated. The activity value is usually positively correlated with the swimming frequency and speed of the shrimp swarm.

[0079] The preset clustering analysis model used in this embodiment is the K-means clustering model. By analyzing the real-time density, saturation, and activity of each monitoring grid, this model divides each grid to be evaluated into multiple clusters, quickly identifying shrimp groups in different health conditions. The K-means model is suitable for processing large-scale data, can efficiently classify the state of shrimp groups in real-time monitoring, and has a relatively fast calculation speed, facilitating a quick response to health assessments. By setting the preset number of clusters K and thresholds, shrimp group states such as healthy and abnormal can be identified based on the clustering results, providing a basis for health assessments.

[0080] The actual shrimp group health index is a comprehensive indicator calculated through the real-time monitoring and calculation of key health parameters such as the density, saturation, and activity of the shrimp group in a specific monitoring area. It reflects the health status of the shrimp group in the current environment, takes into account the influence of abnormal grids (non-shrimp group abnormal grids), and performs weighted calculations based on the real-time density and saturation of these grids to obtain an accurate health assessment result.

[0081] The shrimp group health assessment value is used to comprehensively measure the health status of the shrimp group, helping managers promptly identify abnormal situations and intervene. This assessment value quantifies the health status of the shrimp group by analyzing key parameters such as real-time density, saturation, and activity, in combination with the clustering analysis model.

[0082] By real-time monitoring the physiological and behavioral characteristics of the shrimp group such as density, saturation, and activity, and using deep clustering analysis to evaluate the health status of the shrimp group. First, divide the shrimp group dispersion grids according to the real-time density, and further screen the temporary grids and abnormal grids through saturation fluctuations and activity differences. Then, combine the density and saturation of non-shrimp group abnormal grids to calculate the actual shrimp group health index. At the same time, based on the preset clustering model, calculate the ideal health index through the real-time data of each grid. Finally, compare the differences between the actual and ideal health indices, and adjust the grid step size to optimize the accuracy of the health assessment, forming the final shrimp group health assessment value.

[0083] By combining physiological and behavioral data such as real-time density, saturation, and activity, it is possible to accurately monitor and evaluate the health status of the shrimp group. By gradually excluding abnormal grids and screening out more representative grids, it avoids assessment biases caused by environmental fluctuations or measurement errors. It not only realizes the dynamic monitoring of the health status of the shrimp group, but also automatically adjusts the grid step size through the comparison of the ideal health index and the actual health index, thereby improving the accuracy and reliability of the health assessment. In addition, the adaptive model based on deep clustering analysis can handle complex and variable monitoring data, has good robustness and flexibility, and effectively solves the problem of inaccurate health assessment results due to unstable environmental parameters.

[0084] Specifically, determining a number of non-shrimp-school abnormal grids based on the real-time activity and the real-time density of any two adjacent temporary grids includes:

[0085] Calculate the difference between the two real-time densities to form a density change value;

[0086] Calculate the difference between the two real-time activities to form an activity change value;

[0087] Determine a change consistency based on the density change value and the activity change value;

[0088] Determine a number of shrimp-school abnormal grids based on the change consistency and the real-time density;

[0089] Exclude the shrimp-school abnormal grids from all the grids to be evaluated to determine a number of the non-shrimp-school abnormal grids.

[0090] By calculating the differences in real-time density and real-time activity between any two adjacent temporary grids, a density change value and an activity change value are formed. Subsequently, a change consistency is calculated based on these two change values to reflect the coordination degree of the density and activity changes. Based on the relationship between the change consistency and the real-time density, shrimp-school abnormal grids are further identified. Finally, the abnormal grids are excluded from all the grids to be evaluated, and ultimately the non-shrimp-school abnormal grids are determined, which mark the areas with inconsistent health states.

[0091] By calculating the differences in density and activity between adjacent grids and combining the judgment of change consistency, non-shrimp-school abnormal grids that do not meet the health standards can be accurately identified. This method can effectively exclude abnormal data and ensure the accuracy of health assessment.

[0092] Specifically, determining the change consistency based on the density change value and the activity change value includes:

[0093] Draw a density change curve based on the density change value within a preset drawing duration;

[0094] Draw an activity change curve based on the activity change value within the preset drawing duration;

[0095] Calculate the cosine similarity between the density change curve and the activity change curve to form the change consistency.

[0096] The preset drawing duration refers to the time window selected during the health assessment for drawing the density and activity change curves. It depends on the monitoring frequency, the periodicity of data changes, and the time scale of changes in the health status of the shrimp population. It is usually set between 10 minutes and 1 hour to ensure that sufficient change information can be captured. In this embodiment, it is set to 30 minutes, which can effectively reflect the dynamic changes in the health status of the shrimp population without being disturbed by short-term fluctuations, ensuring the stability and reliability of the health assessment.

[0097] Within the preset duration, the change curves of the density change value and the activity change value are respectively drawn. Then, the cosine similarity of these two change curves is calculated to obtain the change consistency. The cosine similarity reflects the coordination between the density change and the activity change, and is used to evaluate whether they change synchronously. Through this process, the relationship between density and activity can be quantitatively analyzed to further judge the health status of the shrimp population.

[0098] By calculating the cosine similarity of the density and activity change curves, abnormal situations with poor change synchronization can be effectively identified. By dynamically comparing the changes in density and activity, potential abnormal situations can be captured more sensitively, providing higher reliability and response speed for the health monitoring of the shrimp population.

[0099] Please continue to refer to Figure 2 as shown, which is the determination logic diagram for identifying abnormal grids of the shrimp population in this embodiment;

[0100] Specifically, the determination of several shrimp population abnormal grids based on the change consistency and the real-time density includes:

[0101] When the change consistency is less than the preset consistency threshold, it is determined that the two adjacent temporary grids are difference grids, forming several difference grids;

[0102] Several shrimp population abnormal grids are determined according to the real-time density of the difference grids.

[0103] The preset consistency threshold is a standard value used to judge whether the density and activity changes of adjacent grids are consistent. It depends on the natural fluctuation range of the shrimp population activities in the monitoring area and the sensitivity of health status changes. It is usually set between 0.8 and 0.95. In this embodiment, it is set to 0.9 to ensure that obvious health abnormalities can be accurately detected while reducing misjudgments of short-term fluctuations, improving the reliability and stability of the health assessment.

[0104] When the change consistency is less than the preset consistency threshold, it is determined that two adjacent temporary grids are different grids, forming a number of different grids. Next, according to the real-time density of these different grids, the shrimp swarm abnormal grids are further screened out. These abnormal grids mark the areas with significant differences in health assessment, and there may be abnormal situations in the health of the shrimp swarm.

[0105] By dynamically judging the change consistency and analyzing the real-time density of different grids, the areas with abnormal health conditions can be accurately identified, improving the determination accuracy of abnormal grids and avoiding misjudgment caused by accidental fluctuations.

[0106] Specifically, the determining a number of shrimp swarm abnormal grids according to the real-time density of the different grids includes:

[0107] Calculating the standard deviation of the difference in the real-time density between any two adjacent different grids to form a differential density fluctuation value;

[0108] When the differential density fluctuation value is greater than the preset differential fluctuation threshold, it is determined that the different grid is the shrimp swarm abnormal grid, forming a number of shrimp swarm abnormal grids.

[0109] The preset differential fluctuation threshold is a standard value used to judge whether the density fluctuation of different grids is abnormal, depending on the density fluctuation range of the shrimp swarm in the monitoring area and the sensitivity of the system to abnormal changes. It is usually set between 0.05 and 0.15, and is set to 0.1 in this embodiment, which can better capture large-scale health fluctuations and avoid being overly sensitive to minor changes, thereby improving the stability and accuracy of the assessment results.

[0110] First, calculate the standard deviation of the difference in the real-time density between any adjacent different grids to form a differential density fluctuation value. Then, if the differential density fluctuation value is greater than the preset differential fluctuation threshold, it is determined that these different grids are shrimp swarm abnormal grids. These abnormal grids may indicate large fluctuations or abnormalities in the health status of the shrimp swarm, so they are further excluded or marked as areas that need to be monitored key.

[0111] By calculating the differential density fluctuation value, the grid areas with abnormal density fluctuations are effectively identified, thus accurately capturing the shrimp swarms with large health fluctuations, helping to exclude misjudgments caused by natural fluctuations or short-term changes, and improving the accuracy of health assessment and the sensitivity of the system.

[0112] Specifically, the determining the actual shrimp swarm health index according to the real-time density and the real-time saturation of the non-shrimp swarm abnormal grids includes:

[0113] Performing normalization processing on the real-time density to form a standard density;

[0114] Standardize the real-time saturation to form a standard saturation;

[0115] Perform a weighted sum of the standard density, the standard saturation, a preset density weight, and a preset saturation weight to form an actual shrimp population health index.

[0116] The preset density weight refers to the weight ratio of density in the total assessment when calculating the shrimp population health index, which depends on the impact of density on the health of the shrimp population and the sensitivity of density changes to the health index in the actual farming environment. It is usually set between 0.3 and 0.7. In this embodiment, it is set to 0.5, which can maintain its reasonable weighting effect when density has a greater impact on the health index, while avoiding over-reliance on a single factor.

[0117] The preset saturation weight refers to the weight ratio of saturation in the total assessment when calculating the shrimp population health index, which depends on the impact of saturation on the health of the shrimp population and the indication of saturation fluctuations on the health status in the actual farming environment. It is usually set between 0.3 and 0.7. In this embodiment, it is set to 0.5, which can fully reflect the contribution of saturation changes to health assessment and ensure the accuracy of the assessment results.

[0118] First, standardize the real-time density and real-time saturation of non-abnormal grids of the shrimp population to obtain the standard density and standard saturation. Then, combine the preset density weight and saturation weight to perform a weighted sum of the standard density and standard saturation to form an actual shrimp population health index. In this way, the density and saturation information of the shrimp population is quantified, and its health status is evaluated through weighting, ensuring that the assessment results can fully reflect the overall health level of the shrimp population.

[0119] Through standardization and weighting, comprehensively evaluate the health status of the shrimp population, effectively avoiding errors caused by fluctuations in a single indicator, and at the same time being able to comprehensively consider the impacts of density and saturation, providing a more comprehensive and accurate health index.

[0120] Specifically, the adjusting the preset grid step according to the ideal shrimp population health index and the actual shrimp population health index includes:

[0121] Calculate the relative deviation between the ideal shrimp population health index and the actual shrimp population health index to form an index deviation;

[0122] Adjust the preset grid step according to the index deviation.

[0123] According to the difference between the ideal shrimp group health index and the actual shrimp group health index, calculate the relative deviation between the two to form an index deviation. This deviation reflects the accuracy of the evaluation result at the current grid step size and indicates the difference between the actual health status and the ideal status. Based on this deviation, adjust the preset grid step size, thereby optimizing the health evaluation process, improving the accuracy of grid division, ensuring that subsequent evaluation results are more accurate, and timely reflecting the health changes of the shrimp group.

[0124] By adjusting the preset grid step size according to the index deviation, the grid division can be dynamically optimized, and the sensitivity and accuracy of the health evaluation can be improved.

[0125] Please continue to refer to Figure 3 as shown, which is the decision logic diagram for adjusting the preset grid step size in this embodiment;

[0126] Specifically, the adjusting the preset grid step size according to the index deviation includes:

[0127] Calculate the standard deviation of the index deviation within the preset adjustment duration to form a deviation fluctuation value;

[0128] When the deviation fluctuation value is greater than the preset deviation fluctuation value threshold, reduce the preset grid step size according to the relative deviation between the deviation fluctuation value and the preset deviation fluctuation value threshold and the preset adjustment coefficient.

[0129] The preset deviation fluctuation value threshold is a standard value used to determine whether the deviation fluctuation value exceeds the normal range, which depends on the stability of the breeding environment, the accuracy requirements of the health evaluation, and the expected health fluctuation range of the shrimp group. It is usually set between 0.1 and 0.5. In this embodiment, it is set to 0.3, which can effectively avoid excessive adjustment of the grid step size within the reasonable fluctuation range of the health index change, and at the same time ensure timely adjustment in the case of large fluctuations, thereby improving the accuracy and stability of the evaluation result.

[0130] First, calculate the standard deviation within the preset adjustment duration to obtain the deviation fluctuation value. This value reflects the fluctuation of the deviation and represents the change range between the health index and the ideal value. When the deviation fluctuation value is greater than the preset deviation fluctuation value threshold, it indicates that the stability of the evaluation result is poor. At this time, according to the relative deviation between the deviation fluctuation value and the preset deviation fluctuation value threshold and the preset adjustment coefficient, the preset grid step size is reduced to improve the accuracy and reliability of the evaluation.

[0131] By dynamically adjusting the preset grid step size, it can effectively respond to the fluctuation of the health index, reduce the instability caused by environmental changes or measurement errors, ensure that the grid division is more refined, and the health evaluation result is more accurate.

[0132] Please continue to refer to Figure 4As shown, it is a decision logic diagram for determining the shrimp swarm dispersion grid in this embodiment;

[0133] Specifically, the determining of several shrimp swarm dispersion grids according to the real-time density includes:

[0134] When the real-time density is less than the preset density threshold, it is determined that the grid to be evaluated is a shrimp swarm dispersion grid, and several shrimp swarm dispersion grids are formed.

[0135] The preset density threshold refers to the standard value of the real-time density used to judge the shrimp swarm dispersion grid, which depends on the normal growth density range of the shrimp swarm, the capacity of the aquaculture environment, and the impact on the health of the shrimp swarm. It is usually set between 0.2 and 1.0. In this embodiment, it is set to 0.5, which can effectively distinguish the areas with too low density, avoid misjudgment caused by too low density, and at the same time ensure a reasonable division of the shrimp swarm distribution.

[0136] First, monitor the density of each grid to be evaluated. When the real-time density of some grids is lower than the preset density threshold, the system determines these grids as shrimp swarm dispersion grids. Through this process, the areas with lower density and more dispersed shrimp swarms can be effectively identified and divided into shrimp swarm dispersion grids for further analysis and adjustment.

[0137] By determining the shrimp swarm dispersion grid according to the real-time density, the areas with lower density can be effectively identified, so as to accurately divide the shrimp swarm dispersion grid, which helps to optimize the health assessment process, ensure the more accurate assessment result, and avoid the influence of low-density areas on the overall health judgment.

[0138] Specifically, the determining of several temporary grids according to the real-time saturation degree of the shrimp swarm dispersion grid includes:

[0139] Calculate the standard deviation of the real-time saturation degree to form a saturation fluctuation value;

[0140] When the saturation fluctuation value is greater than the preset saturation fluctuation value threshold, it is determined that the shrimp swarm dispersion grid is the temporary grid, and several temporary grids are formed.

[0141] The preset saturation fluctuation value threshold refers to the critical value used to judge whether the saturation fluctuation is significant during the evaluation process, which depends on the degree of influence of the saturation on the health of the shrimp swarm and the normal range of saturation fluctuation in the aquaculture environment. It is usually set between 0.05 and 0.2 to ensure that the grids with obvious changes in the health state can be captured. In this embodiment, it is set to 0.1, which can effectively screen out the grids with significant fluctuations and make timely adjustments to the health assessment, avoiding being overly sensitive or insensitive, and ensuring the rationality and accuracy of the assessment result.

[0142] Calculate the saturation fluctuation value based on the standard deviation of the real-time saturation. When the fluctuation value exceeds the preset saturation fluctuation value threshold, determine the corresponding shrimp swarm dispersion grid as a temporary grid, thereby forming several temporary grids. This process utilizes the fluctuation of saturation to identify potential areas of health problems.

[0143] By introducing the judgment of the saturation fluctuation value, grids with large fluctuations in health status can be effectively identified, and then the grid division can be adjusted to ensure the accuracy and meticulousness of the evaluation, which helps to improve the timeliness and accuracy of shrimp swarm health monitoring.

[0144] So far, the technical solution of the present invention has been described in combination with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the protection scope of the present invention.

[0145] The above are only the preferred embodiments of the present invention and are not used to limit the present invention; for those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A shrimp population health status assessment method based on deep cluster analysis, characterized in that: include: Monitor the real-time density, real-time saturation and real-time activity of shrimp populations in each grid to be evaluated within the monitoring area constructed based on the preset grid step size; Determine a number of shrimp swarm dispersion grids according to the real-time density; Determine a number of temporary grids according to the real-time saturation of the shrimp swarm dispersion grid within a preset determined time period; Determine a number of non-shrimp group abnormal grids according to the real-time activity and the real-time density of any two adjacent temporary grids; Determine an actual shrimp population health index according to the real-time density and the real-time saturation of the non-shrimp population abnormal grid; Determine an ideal shrimp population health index according to a preset cluster analysis model, the real-time density, the real-time saturation, and the real-time activity of each grid to be evaluated; Adjust the preset grid step size according to the ideal shrimp health index and the actual shrimp health index; Determine the actual shrimp health index based on the grid step length to form a shrimp health assessment value; Determining the actual shrimp health index according to the real-time density and the real-time saturation of the non-shrimp abnormal grid includes: Standardizing the real-time density to form a standard density; Standardizing the real-time saturation to form a standard saturation; Performing weighted summation on the standard density, the standard saturation, the preset density weight, and the preset saturation weight to form an actual shrimp population health index; The adjusting the preset grid step size according to the ideal shrimp health index and the actual shrimp health index comprises: Calculating the relative deviation between the ideal shrimp population health index and the actual shrimp population health index to form an index deviation; adjusting the preset grid step size according to the exponential deviation; The adjusting the preset grid step size according to the exponential deviation comprises: Calculate the standard deviation of the index deviation within the preset adjustment period to form a deviation fluctuation value; When the deviation fluctuation value is greater than a preset deviation fluctuation value threshold, the preset grid step size is reduced according to a relative deviation between the deviation fluctuation value and the preset deviation fluctuation value threshold and a preset adjustment coefficient.

2. The shrimp population health status assessment method based on deep cluster analysis according to claim 1, characterized in that: The step of determining a number of non-shrimp group abnormal grids according to the real-time activity and the real-time density of any two adjacent temporary grids comprises: Calculating the difference between the two real-time densities to form a density change value; Calculate the difference between the two real-time activities to form an activity change value; Determining a change consistency according to the density change value and the activity change value; Determine a number of shrimp group abnormal grids according to the change consistency and the real-time density; The shrimp group abnormal grid is excluded from all the grids to be evaluated to determine a number of non-shrimp group abnormal grids.

3. The shrimp population health status assessment method based on deep cluster analysis according to claim 2 is characterized in that: Determining the change consistency according to the density change value and the activity change value includes: Draw a density change curve according to the density change value within a preset drawing time; Draw an activity change curve according to the activity change value within the preset drawing time; The cosine similarity of the density change curve and the activity change curve is calculated to form a change consistency.

4. The shrimp population health status assessment method based on deep cluster analysis according to claim 3 is characterized in that: Determining a number of shrimp group abnormal grids according to the change consistency and the real-time density includes: When the change consistency is less than a preset consistency threshold, determining that the two adjacent temporary grids are difference grids, and forming a plurality of difference grids; A plurality of shrimp group abnormal grids are determined according to the real-time density of the difference grids.

5. The shrimp population health status assessment method based on deep cluster analysis according to claim 4, characterized in that: The step of determining a plurality of shrimp group abnormal grids according to the real-time density of the difference grids comprises: Calculating the standard deviation of the difference between the real-time densities of any two adjacent difference grids to form a difference density fluctuation value; When the difference density fluctuation value is greater than a preset difference fluctuation threshold, the difference grid is determined to be the shrimp group abnormal grid, and a plurality of shrimp group abnormal grids are formed.

6. The shrimp population health status assessment method based on deep cluster analysis according to claim 5, characterized in that: Determining a plurality of shrimp swarm dispersion grids according to the real-time density comprises: When the real-time density is less than a preset density threshold, the grid to be evaluated is determined to be a shrimp swarm dispersion grid, and a plurality of shrimp swarm dispersion grids are formed.

7. The shrimp population health status assessment method based on deep cluster analysis according to claim 6, characterized in that: The determining of a plurality of temporary grids according to the real-time saturation of the shrimp swarm dispersion grids comprises: Calculating the standard deviation of the real-time saturation to form a saturation fluctuation value; When the saturation fluctuation value is greater than a preset saturation fluctuation value threshold, the shrimp swarm dispersion grid is determined to be the temporary grid, and a plurality of temporary grids are formed.

Images