Method for monitoring health state of seaweed bed in ecological restoration area

Abstract

The invention provides an ecological restoration area sea grass bed health state monitoring method, and relates to the technical field of sea grass bed ecological restoration, and the method comprises the steps: obtaining a sea grass bed ecological region, carrying out the region division, and determining a plurality of sea grass bed monitoring regions; according to the image sensor, combined with the image feature detection channel, sea grass state feature analysis is carried out, and a plurality of sea grass state feature analysis results are generated; performing illumination detection feature analysis according to the illumination sensor in combination with the illumination feature analysis channel, and generating a plurality of illumination detection feature analysis results; and carrying out abnormity management on the corresponding sea grass bed monitoring area, carrying out abnormity linkage analysis and carrying out abnormity management. The technical problems that the prior art is limited by resolution and environmental factors, the microscopic state change of the sea grass bed cannot be accurately captured, the health state of the sea grass bed is difficult to comprehensively analyze by integrating multiple monitoring dimensions, and the accuracy and comprehensiveness of state monitoring of the sea grass bed are insufficient are solved.

Description

Technical Field

[0001] The invention relates to the technical field of seagrass bed ecological restoration, and in particular to a method for monitoring the health status of seagrass beds in an ecological restoration area. Background Art

[0002] Seagrass beds are fragile ecosystems that are easily affected by human activities and lead to restoration failures. Therefore, formulating effective management and control measures is the key to achieving good restoration results. However, existing monitoring equipment, such as remote sensing technology and ocean sensors, can provide some ocean data, but these data usually have limitations. For example, remote sensing technology is limited by resolution and environmental factors and cannot accurately capture the microscopic changes in the seagrass beds, resulting in limited accuracy and coverage. In addition, existing seagrass bed monitoring methods usually rely on traditional sampling and manual observation methods, which often rely on expert experience and are not suitable for large-scale or continuous monitoring. Manual monitoring is not only time-consuming and labor-intensive, and difficult to provide real-time data, but also lacks the ability to dynamically monitor the health status and real-time early warning. During the growth cycle of the seagrass beds and changes in environmental conditions, it is difficult to quickly discover problems and intervene, resulting in the inability to repair or adjust protection measures in a timely manner. Summary of the invention

[0003] This application provides a method for monitoring the health status of seagrass beds in ecological restoration areas, aiming to solve the technical problems that the existing technology is limited by resolution and environmental factors, cannot accurately capture the changes in the microscopic state of the seagrass beds, and has difficulty in comprehensively analyzing the health status of the seagrass beds from multiple monitoring dimensions, resulting in insufficient accuracy and comprehensiveness in monitoring the status of the seagrass beds.

[0004] The present application discloses a method for monitoring the health status of seagrass beds in an ecological restoration area, the method comprising: obtaining the ecological area of ​​the seagrass beds in the target ecological restoration area, dividing the area based on the ecological characteristics of the seagrass beds, and determining multiple seagrass bed monitoring areas, wherein the multiple seagrass bed monitoring areas correspond to multiple sets of monitoring equipment groups; performing seagrass status characteristic analysis on the multiple seagrass bed monitoring areas according to the image sensors in the multiple sets of monitoring equipment groups and in combination with image feature detection channels, and generating multiple seagrass status characteristic analysis results, wherein the multiple seagrass status characteristic analysis results include a number of seagrass abnormal status warning information; and The light sensor in the detection equipment group, in combination with the light feature analysis channel, performs light detection feature analysis on the multiple seagrass bed monitoring areas to generate multiple light detection feature analysis results, wherein the multiple light detection feature analysis results include a number of light abnormality point warning information; according to the several seagrass abnormal state warning information and the several light abnormality point warning information, abnormal management of the corresponding seagrass bed monitoring areas is performed; at the same time, abnormal linkage analysis is performed in combination with the several seagrass abnormal state warning information and the several light abnormality point warning information, and abnormal management of the corresponding seagrass bed monitoring areas is performed according to the linkage analysis results.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0006] According to the ecological characteristics of seagrass beds, the target ecological restoration area is divided into multiple seagrass bed monitoring areas, and a corresponding monitoring equipment group is configured for each area. This process ensures that the health status of different areas can be accurately monitored separately, and each monitoring area is equipped with appropriate monitoring equipment, thereby improving the accuracy and comprehensiveness of monitoring; through image sensors combined with image feature detection channels, seagrass status characteristics of multiple seagrass bed monitoring areas are analyzed to generate multiple seagrass status characteristic analysis results, including seagrass abnormal status warning information. This process combines visual data and image processing technology, and can detect the health status of seagrass beds in real time. If seagrass abnormalities are found, early warnings will be issued, thereby improving the sensitivity and accuracy of abnormality detection; through light sensors combined with light feature analysis functions, the light conditions of the seagrass bed area are monitored. Light is a key factor affecting seagrass growth, so accurate monitoring of light is essential. The light detection feature analysis results generated in this step include warning information on abnormal light points, which can issue a warning when the light conditions do not meet the health requirements of the seagrass bed, preventing poor light conditions from having an adverse effect on the growth of seagrass. By combining the abnormal seagrass status warning information and the abnormal light point warning information, an abnormal linkage analysis is performed, which can not only handle a single type of abnormality, but also integrate the mutual influence of multiple factors to conduct more comprehensive abnormal management. The linkage analysis results provide more accurate and comprehensive abnormal management measures for the seagrass bed monitoring area. This linkage analysis can integrate a variety of monitoring data, identify the relationships and potential chain reactions between different factors, and enhance the comprehensiveness of abnormality detection and management. Overall, this method effectively improves the monitoring efficiency, prediction accuracy and emergency management capabilities of seagrass beds, and provides strong technical support for marine ecological restoration and seagrass bed protection.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A flow chart of a method for monitoring the health status of seagrass beds in an ecological restoration area is provided for an embodiment of the present application.

[0009] Figure 2 A schematic diagram of a process for generating a plurality of seagrass abnormal status warning information in a method for monitoring the health status of seagrass beds in an ecological restoration area is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0010] The embodiments of the present application provide a method for monitoring the health status of seagrass beds in ecological restoration areas, thereby solving the technical problems that the prior art is limited by resolution and environmental factors, cannot accurately capture the changes in the microscopic state of the seagrass beds, and has difficulty in comprehensively analyzing the health status of the seagrass beds from multiple monitoring dimensions, resulting in insufficient accuracy and comprehensiveness in monitoring the status of the seagrass beds.

[0011] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0012] like Figure 1 As shown, the embodiment of the present application provides a method for monitoring the health status of seagrass beds in an ecological restoration area, the method comprising:

[0013] The seagrass bed ecological area of ​​the target ecological restoration area is obtained, and the area is divided based on the ecological characteristics of the seagrass bed to determine multiple seagrass bed monitoring areas, wherein the multiple seagrass bed monitoring areas correspond to multiple sets of monitoring equipment groups.

[0014] By interacting with the ecological restoration system, or through remote sensing technology, such as satellite images or drone photography, and ground surveys, the specific location of the target ecological restoration area can be determined, and the seagrass bed ecological area can be obtained, which can be an artificial seagrass bed or a natural seagrass bed.

[0015] Regional divisions are made based on the ecological characteristics of seagrass beds, such as seagrass species, density, coverage, seagrass growth status and water flow environment. For example, based on species differences, different regions may have different types of seagrass, such as red algae, green algae, seagrass plants, etc.; based on growth status differences, regions can be divided into healthy zones, risk zones and degraded zones based on the growth conditions of seagrass, such as health, decline, and death; based on differences in environmental conditions, different monitoring areas are divided based on the effects of environmental factors such as water depth, light intensity, and water flow rate on the growth of seagrass beds.

[0016] Based on the characteristics of the divided areas and combined with monitoring needs, it is determined that different areas require different types of monitoring equipment. For example, healthy areas only require basic light and water quality monitoring, while degraded areas may require more precise image recognition and depth monitoring. Each monitoring area is equipped with appropriate monitoring equipment to ensure that the ecological status of the seagrass bed can be fully monitored.

[0017] According to the image sensors in the multiple monitoring equipment groups, combined with the image feature detection channel, the seagrass status characteristic analysis is performed on the multiple seagrass bed monitoring areas to generate multiple seagrass status characteristic analysis results, wherein the multiple seagrass status characteristic analysis results include a number of seagrass abnormal status warning information.

[0018] According to the designated multiple seagrass bed monitoring areas, image sensors such as high-resolution cameras, drones, seabed robots or underwater cameras are deployed to conduct regular or real-time image acquisition in each monitoring area, and the collected image data is input into the image feature detection channel for feature extraction. The image feature detection channel can identify and extract the seagrass features in the image, such as the morphological characteristics of seagrass, the color changes of seagrass, the coverage of seagrass, the growth pattern of seagrass, etc.

[0019] The seagrass features are compared with the seagrass feature recognition library, and those that do not match the normal seagrass features in the comparison results are marked as abnormal seagrass status, including similar and non-similar abnormalities. Similar abnormalities indicate abnormalities in the health status of seagrass, such as malnutrition, disease, withering, etc. For example, if the color change of seagrass leaves exceeds a certain threshold, it may indicate health problems; if the coverage is lower than the set threshold, it may be a signal of degradation or environmental deterioration; non-similar abnormalities indicate the presence of invasion, such as floating large seaweed, which not only competes with seagrass for nutrients, but also blocks light, resulting in a decrease in the light intensity reaching the seagrass canopy, thereby causing the degradation of the seagrass bed. According to the abnormal status of seagrass, several corresponding seagrass abnormal status warning information are generated.

[0020] According to the light sensors in the multiple sets of monitoring equipment groups, combined with the light feature analysis channel, light detection feature analysis is performed on the multiple seagrass bed monitoring areas to generate multiple light detection feature analysis results, wherein the multiple light detection feature analysis results include a number of light abnormality point warning information.

[0021] Light sensors are deployed in multiple seagrass bed monitoring areas to collect light data at different depths in real time. Commonly used light sensors include light intensity sensors, spectrum sensors, and radiometers. Based on historical data and ecological research, a normal light intensity range is set. For example, multiple light detection thresholds are preset based on parameters such as light intensity and wavelength. By comparing with the preset thresholds, abnormal light points are determined. That is, when the light value collected by the light sensor exceeds the set normal range, the corresponding collection point is automatically marked as a light abnormal point, and the geographical location of the abnormal point and the time when the abnormality occurred are recorded, and a number of abnormal light point warning information is generated.

[0022] Based on the several warning information on abnormal seagrass states and the several warning information on abnormal lighting points, abnormal management of the corresponding seagrass bed monitoring area is performed. At the same time, abnormal linkage analysis is performed in combination with the several warning information on abnormal seagrass states and the several warning information on abnormal lighting points. According to the linkage analysis results, abnormal management of the corresponding seagrass bed monitoring area is performed.

[0023] The seagrass abnormal status warning information reflects the abnormalities in the growth status and health status of the seagrass bed, such as seagrass withering, yellowing, disease, etc., as well as biological invasions of the seagrass bed, such as the invasion of large floating seaweed. The warning information includes the abnormal type, abnormal area and location, and the severity of the abnormality. The abnormal light point warning information reflects that light problems directly affect the health status of the seagrass bed, such as excessive light, insufficient light, light fluctuations, etc. The warning information includes abnormal light type, light intensity deviation value, abnormal point location, and abnormal time period.

[0024] A preliminary diagnosis is conducted on each warning information to evaluate the severity of the anomaly, the frequency of occurrence and the potential threat to seagrass beds. The warning points are prioritized based on factors such as the severity and frequency of the anomaly information. Severe anomalies, such as large-scale insufficient light or seagrass withering, should be dealt with as a priority. Depending on the type of anomaly, appropriate response measures are taken, such as adjusting light intensity, removing invasive species, strengthening water quality management, and releasing beneficial substances.

[0025] Abnormal linkage analysis reveals the potential correlation between different types of abnormalities through the integration of cross-data sources, helping to determine whether certain abnormal phenomena are causally related to each other, or whether one abnormality may aggravate the occurrence of another abnormality. Specifically, by analyzing the relationship between the abnormal state of seagrass and the abnormal light points, it is determined whether light changes are the root cause of abnormal seagrass health. For example, if excessive light appears in a certain area and accompanied by yellowing of seagrass, this may mean that excessive light directly affects the photosynthesis of seagrass. The seagrass abnormal state data and light abnormality data are integrated for joint analysis. The analysis methods include multivariate regression analysis, neural network models, support vector machines, etc., to identify and classify different types of abnormal patterns. For example, when light abnormalities continue to exceed a certain threshold, it may trigger a seagrass withering warning; excessive light fluctuations may cause unstable seagrass growth.

[0026] Combine historical data with real-time monitoring results to predict future development trends of abnormal situations. For example, linkage analysis can be used to predict that insufficient light in a certain area may lead to long-term seagrass health problems. Based on the linkage analysis results, more precise abnormal management strategies can be formulated. For example, if the linkage analysis shows that excessive or weak light causes seagrass health problems, the light can be adjusted, such as by adjusting underwater lights, installing shading nets, etc.

[0027] Through intelligent data analysis and decision-making, timely and accurate management of seagrass bed abnormalities can be achieved, the negative impact of environmental disturbances on seagrass health can be minimized, and effective restoration measures can be taken to ensure the stability and sustainability of ecological restoration areas.

[0028] Furthermore, if Figure 2As shown, the image sensors in the plurality of monitoring equipment groups are combined with image feature detection channels to perform seagrass status feature analysis on the plurality of seagrass bed monitoring areas to generate a plurality of seagrass status feature analysis results, wherein the plurality of seagrass status feature analysis results include a plurality of seagrass abnormal status warning information, and the method includes:

[0029] Through the image sensor in the first monitoring equipment group, continuous depth image acquisition is performed in the first seagrass bed monitoring area to obtain a first continuous image data set, wherein the first seagrass bed monitoring area is any area of ​​the multiple seagrass bed monitoring areas, and the first continuous image data set has an image depth identifier; the seagrass type and seagrass growth status of the first seagrass bed monitoring area are obtained, and a seagrass feature recognition library is established; an initial pre-anchor frame is set according to the seagrass growth status, and the initial pre-anchor frame is adjusted based on the image depth identifier to establish a pyramid anchor frame; the first continuous image data set is framed and compared step by step with the pyramid anchor frame, and based on the frame selection and comparison results, the seagrass feature recognition library is called to perform feature recognition, and the plurality of seagrass abnormal status warning information is generated based on the feature recognition results.

[0030] Each monitoring equipment group is used to monitor a designated area in real time and is composed of multiple sensors. An analysis object is randomly selected from multiple seagrass bed monitoring areas as the first seagrass bed monitoring area, and the first monitoring equipment group corresponding to the first seagrass bed monitoring area is obtained. The image sensor in the first monitoring equipment group, which can be a high-resolution camera, a drone, an underwater robot or an underwater camera, is used to perform continuous depth photography of the target area to capture image changes in the seagrass growth area. The first continuous image data set obtained includes, in addition to conventional two-dimensional images, depth information of each pixel point, and the depth mark provides the underwater depth of the photography.

[0031] Obtain the type of seagrass, for example, directly interact with the ecological restoration system to obtain the type of seagrass in the first seagrass bed monitoring area, or match the seagrass in the image with the existing species library to determine the type of seagrass in the image. Common seagrass types include brittle grass, ocean grass, and cow grass; obtain the growth status of seagrass, for example, identify the different stages of seagrass growth based on its morphology, color and other characteristics, such as seedling stage, growth stage, aging stage, etc.

[0032] The seagrass feature recognition library is a database established based on a large amount of annotated data. It contains typical image features and corresponding growth status information for each seagrass type. Each seagrass type or growth status corresponds to an entry, which includes image features, type information and growth status annotations. The seagrass feature recognition library is used to compare new image data, quickly identify seagrass types and determine their growth status.

[0033] An anchor frame is a rectangular frame used to predict the target position in the target detection algorithm. According to the growth status of seaweed, such as healthy, damaged, overcrowded, lacking, etc., the initial pre-anchor frame is set in the image. For example, healthy seaweed usually has dense leaves and complete morphology, and may require a larger anchor frame; while damaged or diseased seaweed may require a smaller or irregular frame. The seaweed in the image is at different depths, and the depth information can be used to correct the spatial position of the anchor frame. For example, seaweed far away from the camera will appear smaller, while seaweed closer to the camera will appear larger. According to the depth mark, the size of the initial pre-anchor frame is adjusted to make it more consistent with the size and position of the actual seaweed. Through the calculation of the depth mark, the real scale of each anchor frame relative to the actual seaweed can be obtained, so as to adjust the aspect ratio of the initial pre-anchor frame. According to the adjustment result, a pyramid anchor frame is established to better cover the seaweed area. The pyramid anchor frame is used to process targets of different scales in the image. By constructing multiple anchor frame levels of different scales, seaweed targets of different sizes and depths can be processed more effectively.

[0034] Using the pyramid anchor frame, the first continuous image dataset is compared in a step-by-step frame selection manner. The anchor frame of each layer is matched with the seagrass area in the image dataset to find the seagrass area that best meets the characteristics. Specifically, starting from the bottom layer of the pyramid, the image is initially framed by a larger anchor frame. As it is passed upward step by step, the scale of the frame selection is gradually reduced to ensure accurate frame selection from global to local.

[0035] The established seagrass feature recognition library is called to match the selected area through the feature recognition library to identify the type, health status and growth status of the seagrass. Based on the comparison results, the seagrass areas that may have abnormalities are identified. For example, seaweed invasion, seagrass lesions, missing, over-reproduction, etc. are all marked as abnormal states. According to the identified abnormal states, several seagrass abnormal state warning messages are generated. Each abnormal state warning message includes the specific location of the seagrass area, the type of abnormality detected, the severity of the abnormality, etc.

[0036] Furthermore, the method of adjusting the initial pre-anchor frame based on the image depth identifier to establish a pyramid anchor frame includes:

[0037] Acquire sensor control parameters of the image sensor, wherein the sensor control parameters include focal length, resolution, and field of view angle; adjust the size and position of the initial pre-anchor frame according to the focal length, resolution, and field of view angle in combination with the image depth identifier, and establish the pyramid anchor frame according to the adjustment result.

[0038] Get the sensor control parameters, including focal length, resolution, and field of view. The focal length is a parameter that affects the field of view of the image sensor. It determines the camera's viewing angle and the projection size of the object in the image. The focal length directly affects the size and near-far distribution of the seagrass area captured by the sensor. When the focal length is longer, the distant seagrass in the seagrass bed will appear smaller in the image, and the size of the pre-anchor frame needs to be increased accordingly. Conversely, when the focal length is shorter, the distant seagrass area will be smaller and the nearby area will appear larger, and the pre-anchor frame needs to be adjusted to be smaller.

[0039] Resolution refers to the clarity of an image, usually expressed in pixels. When the resolution is higher, the position and morphological features of seaweed can be identified and located more accurately. Among the control parameters of the image sensor, the resolution determines the small range of changes that can be distinguished. The size of the anchor frame needs to be adjusted according to the resolution to adapt to the clarity of the image.

[0040] The field of view angle determines the image range captured by the sensor and affects the distribution of objects in the image and their relative positions. When the field of view angle is larger, the pre-anchor frame needs to cover more seagrass areas; when the field of view angle is smaller, the pre-anchor frame will be more concentrated in a small area.

[0041] By calculating the focal length and the field of view of the image, the mapping relationship of the anchor frame in space is determined, and its size is adjusted as needed. For example, the anchor frame needs to be larger for seaweed in the distance, while a smaller frame can be used for seaweed nearby. The resolution adaptability of the anchor frame is adjusted according to the resolution. If the image resolution is high, the size of the anchor frame can be smaller and the target can be located more accurately. If the resolution is low, the anchor frame needs to be enlarged to ensure the visibility of the target. The field of view angle determines the range of images that the camera can capture, and therefore affects the spatial positioning of the anchor frame. If the field of view angle is large, the seaweed targets appearing in the image will be more scattered, and the anchor frame needs to be enlarged to adapt to a wider area. If the field of view angle is small, the anchor frame can be relatively small to cover the target area intensively.

[0042] The image depth identifier is the actual distance from each pixel to the camera. It provides depth data for each pixel. The combination of the depth identifier and the sensor control parameters can further adjust the position and size of the anchor frame in the image. For example, the depth information can help determine the position of the target object. If the target seaweed is farther away, the depth information will provide additional adjustment basis, so that the anchor frame can adapt to seaweed targets at different depths.

[0043] According to different focal lengths, field of view angles, resolutions and depth identification information, the size and position of the anchor frame are adjusted layer by layer, and finally a multi-level anchor frame structure that adapts to different scales is formed to obtain a pyramid anchor frame. The creation of the pyramid anchor frame ensures that the system can simultaneously process seagrass targets of different distances and sizes. Through multi-level frame selection, all targets from large-scale seagrass beds to small-scale details can be captured.

[0044] Furthermore, the method for generating the plurality of seaweed abnormal state warning information based on the feature recognition result includes:

[0045] Establish seaweed similarity constraints based on the seaweed type; for the first seaweed abnormal status warning information, perform warning evaluation according to the similarity constraints; when the warning evaluation result is a similar abnormality, perform seaweed abnormality warning feedback; when the warning evaluation result is not a similar abnormality, perform invasion warning feedback.

[0046] Similar constraints refer to a set of rules set according to the characteristics of the seagrass type. These rules are used to constrain the growth or state change patterns that the same type of seagrass should follow during the monitoring process. For example, the normal growth rate and health status of the same type of seagrass should be within a certain range, and growth or degradation beyond this range can be regarded as abnormal.

[0047] When an abnormal seaweed status warning is generated, the warning information is evaluated according to the same type of constraints to determine whether it is outside the normal growth variation range of the same type of seaweed. For example, for the abnormal seaweed detected, it can be determined by comparing the historical data of the abnormal seaweed to determine whether it is an abnormality of the same type of seaweed. If so, it is an abnormality of the same type. If not, there may be an invasion.

[0048] When the warning evaluation result is the same type of abnormality, it means that the abnormality is an abnormal change of the seagrass type within its ecological range, such as seagrass lesions, absence, over-reproduction, etc., so as to provide abnormal warning feedback for seagrass.

[0049] When the early warning evaluation result is not a similar abnormality, it means that the detected seaweed does not belong to the same type of seaweed in the area, that is, there is an invasion of alien species, such as floating large seaweed. Floating large seaweed not only competes with seaweed for nutrients, but also blocks light, resulting in a decrease in the light intensity reaching the seagrass canopy, which may lead to the degradation of the seagrass bed. This is used to provide early warning feedback for the invasion of alien species. For example, further inspection of the floating large seaweed in the seagrass bed. When the distribution amount and distribution area increase to affect the survival of the seagrass, it should be manually removed in time. If necessary, a protective net should be added around the seagrass bed, and the large seaweed on the protective net should be removed regularly.

[0050] Through the above process, it is possible to effectively distinguish between abnormal changes caused by abnormal seagrass growth status and those caused by external invasion, and provide appropriate early warning feedback according to different situations.

[0051] Furthermore, when the warning evaluation result is a similar abnormality, a seaweed abnormality warning feedback is performed, and the method includes:

[0052] When the warning evaluation result is the same type of abnormality, the proportion of abnormal individuals is obtained to generate a first abnormality coefficient; the abnormal individual states of the same type of abnormalities are collected to obtain an abnormal individual state data set; based on a preset abnormal individual state evaluation function, the abnormal individual state data set is analyzed to generate an abnormal individual state evaluation coefficient and a second abnormality coefficient; and according to the first abnormality coefficient and the second abnormality coefficient, seaweed abnormality warning feedback is performed.

[0053] If the abnormal state of the seagrass bed is assessed as a similar abnormality, the proportion of abnormal seagrass individuals in the monitoring area is analyzed. By analyzing the collected image data, it can be determined which seagrass individuals show abnormal characteristics. By calculating the proportion of these individuals, the proportion of abnormal individuals is obtained. For example, if 20 of the 100 seagrass samples in the monitoring area are abnormal, then the proportion of abnormal individuals is 20%. Based on the proportion of abnormal individuals, the first abnormality coefficient is generated. This coefficient can reflect the prevalence of the current seagrass bed abnormality. If the proportion of abnormal individuals is high, it means that the impact of the abnormality is more widespread, and higher intervention is required.

[0054] Detailed status information of these abnormal individuals is further collected. This status information can reflect the health status of the seagrass, such as shrinkage, yellowing or decay. All the collected abnormal individual status data form an abnormal individual status data set, which includes information in multiple dimensions.

[0055] In order to further evaluate the status of abnormal individuals, an abnormal individual status evaluation function is introduced. This function scores the various states of each abnormal individual based on the growth characteristics of seaweed, outputs a comprehensive abnormal individual status evaluation coefficient, and generates a second abnormal coefficient. The second abnormal coefficient reflects the severity of the abnormal individual. This coefficient can be a standardized value between 0 and 1, which is used to measure the status of each abnormal individual. The higher the value, the more serious the abnormality. Further, the status evaluation results of all abnormal individuals are summarized to obtain the comprehensive second abnormal coefficient of the area. For example, if most abnormal individuals show severe slow growth or withering, the second abnormal coefficient will be higher.

[0056] The first anomaly coefficient and the second anomaly coefficient are combined, for example, weighted summed up, to generate a comprehensive anomaly feedback result. This comprehensive coefficient can reflect the extensiveness and severity of the anomaly. If the comprehensive anomaly coefficient is high, it means that the anomaly is serious, which may be caused by pollution, invasion or major environmental changes. In this case, an emergency feedback is issued, requiring rapid intervention measures, such as clearing pollution and repairing the environment.

[0057] Further, according to the light sensors in the plurality of monitoring equipment groups, combined with the light feature analysis channel, light detection feature analysis is performed on the plurality of seagrass bed monitoring areas to generate a plurality of light detection feature analysis results, wherein the plurality of light detection feature analysis results include a plurality of light abnormality point warning information, and the method includes:

[0058] Through the light sensor in the first monitoring equipment group, continuous depth light monitoring is performed in the first seagrass bed monitoring area to obtain a first continuous light data set, wherein the first continuous light data set has a light depth identifier; based on the light depth identifier, a predetermined light intensity data set of the first seagrass bed monitoring area is loaded; according to the predetermined light intensity data set, a light intensity deviation analysis is performed on the first continuous light data set, and points whose light intensity deviation values ​​exceed the preset light intensity deviation values ​​are marked as light abnormality points, and first light abnormality point warning information is generated; and so on, the multiple seagrass bed monitoring areas are traversed to generate the several light abnormality point warning information.

[0059] In the first seagrass bed monitoring area, continuous depth light monitoring is carried out through light sensors to collect light intensity data. This process records the light changes in the area in real time or periodically. Continuous depth monitoring means that data collection will cover different depth positions and capture the changing trends of light. The collected light data is marked according to the collection time and depth to form a data set with light depth identifiers. These identifiers help distinguish light values ​​at different times and depths.

[0060] Based on the light depth identifier in the data, the predetermined light intensity dataset corresponding to each depth position is loaded. This predetermined dataset is preset according to the light intensity range under normal seagrass growth conditions, including the ideal light intensity values ​​of seagrass beds in different depth areas. These data are usually derived from historical monitoring, scientific research or expert settings and are used to represent ideal or normal light conditions.

[0061] Compare the first continuous light data set with the predetermined light intensity data set to perform light intensity deviation analysis, mainly analyzing the difference between the actual value of light intensity and the predetermined value, and mark the points where the deviation exceeds the preset deviation threshold. The light intensity deviation threshold can be adjusted according to the growth requirements and environmental conditions of seaweed. For example, seaweed growth has different adaptability to light, and may be more sensitive to light deviation in some areas, while in other areas, it has a higher tolerance. If the deviation value of a point is greater than the preset threshold, for example, more than ±20%, the point is marked as a light abnormal point, and the corresponding first light abnormal point warning information is generated to notify relevant personnel to further analyze the causes of these abnormalities and take corresponding intervention measures.

[0062] After completing the analysis of abnormal light points in the first seagrass bed monitoring area, continue to traverse other seagrass bed monitoring areas and perform the same light deviation analysis to obtain abnormal light points, and finally generate the aforementioned abnormal light point warning information. Through this traversal, the abnormal light points in all seagrass bed monitoring areas can be fully identified.

[0063] Furthermore, the method of combining the plurality of seaweed abnormal state warning information and the plurality of light abnormal point warning information to perform abnormal linkage analysis includes:

[0064] When the second seaweed abnormal status warning information is triggered, a continuous tracking window is established; continuous light tracking is performed within the continuous tracking window to obtain the second light abnormality point warning information; when the second light abnormality point warning information is not empty, the abnormal linkage analysis of the second seaweed bed monitoring area is triggered.

[0065] When the second seagrass abnormal status warning information in the second seagrass bed monitoring area is triggered, a continuous tracking window is started. The duration of the window can be set according to actual needs. For example, the tracking time length is determined according to the nature of the abnormality. It can be set to ten minutes, thirty minutes, etc. This window is used for subsequent abnormal monitoring to ensure that the abnormal status can be continuously tracked.

[0066] In the continuous tracking window, continuous light tracking is performed, data is obtained in real time through the light sensor, and the light intensity is analyzed to identify abnormal light points and generate related second abnormal light point warning information. The generation process of the abnormal light point warning information has been described in detail in the previous description and will not be repeated here for the sake of brevity.

[0067] When the second abnormal light point warning information is not empty, it means that the abnormal light point has been detected within the continuous tracking window, which further triggers the abnormal linkage analysis, that is, comprehensively analyzing the relationship between the abnormal seagrass state and the abnormal light point, evaluating whether there is a correlation, and taking appropriate management measures. For example, abnormal light may be one of the direct causes of the abnormal seagrass state, or it may aggravate the existing abnormal state. Through linkage analysis, the most reasonable management or response strategy can be comprehensively judged.

[0068] Furthermore, the method further comprises:

[0069] According to the abnormal linkage analysis results, an abnormal situation analysis is performed to obtain abnormal situation information, wherein the abnormal situation information has an abnormal development direction; according to the abnormal development direction, an abnormal reminder is issued to an adjacent area of ​​the second seagrass bed monitoring area.

[0070] Abnormal situation refers to the possible evolution trend of abnormal events within a specific time, including the expansion, slowdown, and aggravation of the abnormality. Specifically, by deeply mining abnormal data and environmental information, the potential development direction of the abnormality can be identified, including whether the abnormal event may further expand, and the direction of the abnormal spread after expansion, which may spread to adjacent areas.

[0071] If the abnormal development direction indicates that the seagrass beds in a certain adjacent area may be affected, an abnormal alert will be automatically issued. This alert can cover multiple adjacent areas to ensure that environmental managers in adjacent areas can take necessary actions in a timely manner, such as strengthening monitoring, adjusting environmental conditions, carrying out restoration work, etc., to ensure that the health of the entire seagrass bed is not affected.

[0072] In summary, the method for monitoring the health status of seagrass beds in an ecological restoration area provided in the embodiment of the present application has the following technical effects:

[0073] According to the ecological characteristics of seagrass beds, the target ecological restoration area is divided into multiple seagrass bed monitoring areas, and a corresponding monitoring equipment group is configured for each area. This process ensures that the health status of different areas can be accurately monitored separately, and each monitoring area is equipped with appropriate monitoring equipment, thereby improving the accuracy and comprehensiveness of monitoring; through image sensors combined with image feature detection channels, seagrass status characteristics of multiple seagrass bed monitoring areas are analyzed to generate multiple seagrass status characteristic analysis results, including seagrass abnormal status warning information. This process combines visual data and image processing technology, and can detect the health status of seagrass beds in real time. If seagrass abnormalities are found, early warnings will be issued, thereby improving the sensitivity and accuracy of abnormality detection; through light sensors combined with light feature analysis functions, the light conditions of the seagrass bed area are monitored. Light is a key factor affecting seagrass growth, so accurate monitoring of light is essential. The light detection feature analysis results generated in this step include warning information on abnormal light points, which can issue a warning when the light conditions do not meet the health requirements of the seagrass bed, preventing poor light conditions from having an adverse effect on the growth of seagrass. By combining the abnormal seagrass status warning information and the abnormal light point warning information, an abnormal linkage analysis is performed, which can not only handle a single type of abnormality, but also integrate the mutual influence of multiple factors to conduct more comprehensive abnormal management. The linkage analysis results provide more accurate and comprehensive abnormal management measures for the seagrass bed monitoring area. This linkage analysis can integrate a variety of monitoring data, identify the relationships and potential chain reactions between different factors, and enhance the comprehensiveness of abnormality detection and management. Overall, this method effectively improves the monitoring efficiency, prediction accuracy and emergency management capabilities of seagrass beds, and provides strong technical support for marine ecological restoration and seagrass bed protection.

[0074] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for monitoring the health status of seagrass beds in an ecological restoration area, characterized in that: The method comprises: Obtaining the seagrass bed ecological area of ​​the target ecological restoration area, dividing the area based on the ecological characteristics of the seagrass bed, and determining multiple seagrass bed monitoring areas, wherein the multiple seagrass bed monitoring areas correspond to multiple sets of monitoring equipment groups; According to the image sensors in the plurality of monitoring equipment groups, combined with the image feature detection channels, seagrass status feature analysis is performed on the plurality of seagrass bed monitoring areas to generate a plurality of seagrass status feature analysis results, wherein the plurality of seagrass status feature analysis results include a plurality of seagrass abnormal status warning information; According to the light sensors in the plurality of monitoring equipment groups, combined with the light feature analysis channel, light detection feature analysis is performed on the plurality of seagrass bed monitoring areas to generate a plurality of light detection feature analysis results, wherein the plurality of light detection feature analysis results include a plurality of light abnormality point warning information; Based on the several warning information on abnormal seagrass states and the several warning information on abnormal lighting points, abnormal management of the corresponding seagrass bed monitoring area is performed. At the same time, abnormal linkage analysis is performed in combination with the several warning information on abnormal seagrass states and the several warning information on abnormal lighting points. According to the linkage analysis results, abnormal management of the corresponding seagrass bed monitoring area is performed.

2. A method for monitoring the health status of seagrass beds in an ecological restoration area as claimed in claim 1, characterized in that: The method comprises: performing seagrass status characteristic analysis on the plurality of seagrass bed monitoring areas according to the image sensors in the plurality of monitoring equipment groups in combination with the image feature detection channel, and generating a plurality of seagrass status characteristic analysis results, wherein the plurality of seagrass status characteristic analysis results include a plurality of seagrass abnormal status warning information, and comprising: By using an image sensor in the first monitoring equipment group, continuous depth image acquisition is performed in a first seagrass bed monitoring area to obtain a first continuous image data set, wherein the first seagrass bed monitoring area is any area of ​​the multiple seagrass bed monitoring areas, and the first continuous image data set has an image depth identifier; Acquire the seagrass type and seagrass growth status in the first seagrass bed monitoring area, and establish a seagrass feature recognition library; An initial pre-anchor frame is set according to the growth state of the seaweed, and the initial pre-anchor frame is adjusted based on the image depth identifier to establish a pyramid anchor frame; The first continuous image data set is framed and compared step by step using the pyramid anchor frame, and based on the framed and compared results, the seaweed feature recognition library is called to perform feature recognition, and the plurality of seaweed abnormal status warning information is generated based on the feature recognition results.

3. A method for monitoring the health status of seagrass beds in an ecological restoration area as claimed in claim 2, characterized in that: The method of adjusting the initial pre-anchor frame based on the image depth identifier to establish a pyramid anchor frame includes: Acquire sensor control parameters of the image sensor, wherein the sensor control parameters include focal length, resolution, and field of view angle; According to the focal length, resolution, and field of view, combined with the image depth identifier, the size and position of the initial pre-anchor frame are adjusted, and the pyramid anchor frame is established according to the adjustment result.

4. A method for monitoring the health status of seagrass beds in an ecological restoration area as claimed in claim 2, characterized in that: The method of generating the plurality of seaweed abnormal state warning information based on the feature recognition result comprises: establishing a seagrass class constraint based on the seagrass type; For the first seaweed abnormal state warning information, a warning evaluation is performed according to the same type of constraints; When the warning evaluation result is the same type of abnormality, the seagrass abnormality warning feedback is carried out; When the warning evaluation result is not the same type of abnormality, an intrusion warning feedback is performed.

5. A method for monitoring the health status of seagrass beds in an ecological restoration area as claimed in claim 4, characterized in that: When the warning evaluation result is the same type of abnormality, seaweed abnormality warning feedback is performed, and the method includes: When the early warning evaluation result is the same type of anomaly, the proportion of abnormal individuals is obtained to generate the first anomaly coefficient; Collect the abnormal individual states of the same abnormality to obtain the abnormal individual state data set; Based on a preset abnormal individual state evaluation function, the abnormal individual state data set is analyzed to generate an abnormal individual state evaluation coefficient and a second abnormal coefficient; According to the first abnormality coefficient and the second abnormality coefficient, seaweed abnormality warning feedback is performed.

6. A method for monitoring the health status of seagrass beds in an ecological restoration area as claimed in claim 1, characterized in that: According to the light sensors in the plurality of monitoring equipment groups, combined with the light feature analysis channel, light detection feature analysis is performed on the plurality of seagrass bed monitoring areas to generate a plurality of light detection feature analysis results, wherein the plurality of light detection feature analysis results include a plurality of light abnormality point warning information, and the method includes: Performing continuous depth light monitoring in the first seagrass bed monitoring area through the light sensor in the first monitoring equipment group to obtain a first continuous light data set, wherein the first continuous light data set has a light depth identifier; Based on the light depth identifier, loading a predetermined light intensity data set of the first seagrass bed monitoring area; According to the predetermined light intensity data set, performing light intensity deviation analysis on the first continuous light data set, marking points where the light intensity deviation value exceeds the preset light intensity deviation value as light abnormality points, and generating first light abnormality point warning information; By analogy, the plurality of seagrass bed monitoring areas are traversed to generate warning information of the plurality of abnormal light points.

7. A method for monitoring the health status of seagrass beds in an ecological restoration area as claimed in claim 1, characterized in that: The method of combining the plurality of seaweed abnormal state warning information and the plurality of light abnormal point warning information to perform abnormal linkage analysis includes: When the second seagrass abnormal state warning information is triggered, a continuous tracking window is established; Executing continuous illumination tracking within the continuous tracking window to obtain warning information of a second illumination abnormality point; When the second abnormal light point warning information is not empty, the abnormal linkage analysis of the second seagrass bed monitoring area is triggered.

8. A method for monitoring the health status of seagrass beds in an ecological restoration area as claimed in claim 7, characterized in that: The method further comprises: According to the abnormal linkage analysis result, abnormal situation analysis is performed to obtain abnormal situation information, wherein the abnormal situation information has an abnormal development direction; According to the abnormal development direction, abnormal reminders are issued to the adjacent areas of the second seagrass bed monitoring area.

Images