A method and system for early warning of marine red tide disaster based on unmanned aerial vehicle

Abstract

The application discloses a kind of marine red tide disaster early warning method and system based on unmanned plane, mainly related to marine red tide disaster early warning technical field, to solve the problem of insufficient multi-source data fusion, high maintenance cost, hyperspectral sensor processing slow of existing scheme.There is:by being equipped with thermal infrared sensor and chlorophyll concentration collector unmanned plane, determine initial red tide abnormal area;Based on the biological fluorescence band involved in red tide, configure corresponding fluorescence sensor;Through the unmanned plane equipped with fluorescence sensor, detect whether there is fluorescence reaction in initial red tide abnormal area;When there is fluorescence reaction, determine that initial red tide abnormal area is red tide abnormal area;By collecting reflected fluorescence intensity, determine the corresponding red tide marker concentration;Obtain the biological fluorescence band of current fluorescence sensor, determine the red tide marker type of current red tide abnormal area.

Description

Technical Field

[0001] This application relates to the field of marine red tide disaster early warning technology, and in particular to a method and system for marine red tide disaster early warning based on unmanned aerial vehicles (UAVs). Background Technology

[0002] Currently, marine red tide monitoring mainly relies on three technologies: satellite remote sensing, manned vessel sampling, and fixed buoys. Satellite remote sensing (such as MODIS and Sentinel-3) retrieves chlorophyll a concentration from multispectral data, but it is affected by cloud interference and has low spatial resolution (typically >300 meters), making it difficult to capture small-scale outbreaks in the early stages of red tides. Manned vessel sampling can obtain accurate data, but it is costly (over 100,000 yuan per voyage), slow in response (24-48 hours from sampling to laboratory analysis), and cannot cover complex nearshore waters. Fixed buoy networks (such as the GOOS system) can achieve continuous monitoring, but deployment density is limited (typical spacing of 50 kilometers), and maintenance costs are high. In recent years, some studies have attempted to use UAVs equipped with hyperspectral sensors for red tide screening, but this requires specialized hyperspectral sensors and machine learning models, as well as preprocessing redundant bands and addressing signal-to-noise ratio issues.

[0003] The above technologies have the following problems: (1) High maintenance costs: satellite remote sensing, manned ship sampling and fixed buoys have high maintenance costs; (2) Slow processing of hyperspectral sensors: the area of ​​spectral detection is large, and redundant bands need to be preprocessed and the signal-to-noise ratio problem needs to be solved. Summary of the Invention

[0004] This application provides a method and system for early warning of marine red tide disasters based on unmanned aerial vehicles (UAVs), in order to solve the problems of high maintenance costs and slow processing of hyperspectral sensors in existing solutions.

[0005] Firstly, this application provides a method for early warning of marine red tide disasters based on unmanned aerial vehicles (UAVs), the method comprising:

[0006] A drone equipped with a thermal infrared sensor and a chlorophyll concentration collector collects water temperature and chlorophyll concentration data at various preset detection points in the area to be monitored. Based on the water temperature and chlorophyll concentration at the preset detection points, the initial red tide anomaly area is determined. Based on the biofluorescence bands involved in the red tide, corresponding fluorescence sensors are configured. The drone equipped with the fluorescence sensor detects whether there is a fluorescence reaction in the initial red tide anomaly area. When a fluorescence reaction is present, the initial red tide anomaly area is determined to be a red tide anomaly area. The concentration of the corresponding red tide marker is determined by collecting the reflected fluorescence intensity. The biofluorescence band of the current fluorescence sensor is obtained to determine the type of red tide marker in the current red tide anomaly area. The red tide marker concentration and red tide marker type are uploaded to the corresponding preset red tide monitoring terminal.

[0007] In one implementation of this application, the initial red tide anomaly area is determined based on the water temperature and chlorophyll concentration at preset detection points, specifically including:

[0008] Based on the correspondence between chlorophyll concentration and abnormal temperature rise threshold, determine the abnormal temperature rise threshold corresponding to the current chlorophyll concentration;

[0009] Using a preset detection point as the center and a preset distance as the radius, a background area is obtained, and the water temperature at several locations on the edge of the background area is randomly collected; the location water temperature is input into a clustering algorithm, and the average location water temperature of the cluster center with the largest number of clusters is determined as the background water temperature.

[0010] Calculate the difference between the background water temperature and the water body temperature. When it exceeds the abnormal temperature rise threshold, determine the area with the preset detection point as the center and the preset distance as the radius as the preset ratio as the initial red tide abnormal area; where the preset ratio is greater than 0 and less than 1.

[0011] In one implementation of this application, a drone equipped with a fluorescence sensor is used to detect whether there is a fluorescent reaction in the initial red tide anomaly area, specifically including:

[0012] By using a drone equipped with a fluorescence sensor to switch between different biofluorescence bands in the initial red tide anomaly area, the presence of fluorescence reaction in the initial red tide anomaly area can be detected.

[0013] In one implementation of this application, the concentration of the corresponding red tide marker is determined by collecting the intensity of the reflected fluorescence, specifically including:

[0014] Through the formula:

[0015] Determine the corresponding red tide marker concentrations;

[0016] in, and This is a preset value.

[0017] In one implementation of this application, the biofluorescence band of the current fluorescence sensor is obtained to determine the type of red tide marker in the current red tide anomaly area, specifically including:

[0018] Obtain the number of red tide marker types corresponding to the current biofluorescence band of the fluorescence sensor. When the number of types is unique, determine the unique red tide marker type as the red tide marker type of the current red tide anomaly area.

[0019] When the number of types is not unique, the type of red tide markers in the current red tide anomaly area can be determined by using a drone equipped with an image recognition device.

[0020] In one implementation of this application, after acquiring the biofluorescence band of the current fluorescence sensor and determining the type of red tide marker in the current red tide anomaly area, the method further includes:

[0021] Obtain the number of preset detection points corresponding to the red tide anomaly area. When the number of preset detection points is less than the preset value, add a new preset detection point in the red tide anomaly area.

[0022] Secondly, this application provides a marine red tide disaster early warning system based on unmanned aerial vehicles (UAVs), the system comprising:

[0023] The determination module is used to collect water temperature and chlorophyll concentration at each preset detection point in the area to be detected by a drone equipped with a thermal infrared sensor and a chlorophyll concentration collector; and to determine the initial red tide anomaly area based on the water temperature and chlorophyll concentration at the preset detection points.

[0024] The detection module is used to configure corresponding fluorescence sensors based on the biofluorescence bands involved in red tides;

[0025] A drone equipped with a fluorescence sensor was used to detect whether there was a fluorescent reaction in the initial red tide anomaly area;

[0026] The upload module is used to identify the initial red tide anomaly area as a red tide anomaly area when a fluorescence reaction is present; to determine the corresponding red tide marker concentration by collecting the reflected fluorescence intensity; to obtain the biofluorescence band of the current fluorescence sensor to determine the red tide marker type of the current red tide anomaly area; and to upload the red tide marker concentration and red tide marker type to the corresponding preset red tide monitoring terminal.

[0027] In one implementation of this application, the determining module includes a determining unit.

[0028] It is used to determine the abnormal temperature threshold corresponding to the current chlorophyll concentration based on the correspondence between chlorophyll concentration and abnormal temperature threshold.

[0029] Using a preset detection point as the center and a preset distance as the radius, a background area is obtained, and the water temperature at several locations on the edge of the background area is randomly collected; the location water temperature is input into a clustering algorithm, and the average location water temperature of the cluster center with the largest number of clusters is determined as the background water temperature.

[0030] Calculate the difference between the background water temperature and the water body temperature. When it exceeds the abnormal temperature rise threshold, determine the area with the preset detection point as the center and the preset distance as the radius as the preset ratio as the initial red tide abnormal area; where the preset ratio is greater than 0 and less than 1.

[0031] In one implementation of this application, the upload module includes a concentration determination unit.

[0032] Used in the formula:

[0033] Determine the corresponding red tide marker concentrations;

[0034] in, and This is a preset value.

[0035] In one implementation of this application, the upload module includes a marker determination unit.

[0036] The number of red tide marker types corresponding to the current biofluorescence band of the fluorescence sensor is used to obtain the number of red tide marker types. When the number of types is unique, the unique red tide marker type is determined as the red tide marker type of the current red tide anomaly area.

[0037] When the number of types is not unique, the type of red tide markers in the current red tide anomaly area can be determined by using a drone equipped with an image recognition device.

[0038] As can be seen from the above technical solutions, this application has the following advantages:

[0039] Reduced operation and maintenance costs:

[0040] Compared to satellite remote sensing and manned ships, the cost of a single inspection by an unmanned aerial vehicle (UAV) system is lower. In addition, the equipment used in this application (thermal infrared sensor, chlorophyll concentration collector, fluorescence sensor) is relatively conventional, readily available and inexpensive.

[0041] Breakthrough in hyperspectral processing efficiency:

[0042] By employing band pre-screening technology, the number of bands to be processed is reduced from 256 in conventional hyperspectral instruments to a few key biological fluorescence bands, eliminating the need to preprocess redundant bands and solve the signal-to-noise ratio problem. Attached Figure Description

[0043] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a flowchart of a method for early warning of marine red tide disasters based on unmanned aerial vehicles (UAVs) provided in an embodiment of this application.

[0045] Figure 2 This is a schematic diagram of the internal structure of a marine red tide disaster early warning system based on unmanned aerial vehicles (UAVs) provided in an embodiment of this application. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Those skilled in the art should understand that the embodiments described below are merely preferred embodiments of this disclosure and do not imply that this disclosure can only be implemented through these preferred embodiments. These preferred embodiments are merely used to explain the technical principles of this disclosure and are not intended to limit the scope of protection of this disclosure. Based on the preferred embodiments provided by this disclosure, all other embodiments obtained by those skilled in the art without creative effort should still fall within the scope of protection of this disclosure.

[0048] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0049] The technical solutions proposed in the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0050] The embodiment provides a method for early warning of marine red tide disasters based on unmanned aerial vehicles (UAVs), such as Figure 1 As shown in the embodiments of this application, the method mainly includes the following steps:

[0051] Step 110: Using a drone equipped with a thermal infrared sensor and a chlorophyll concentration collector, collect water temperature and chlorophyll concentration at each preset detection point in the area to be detected; based on the water temperature and chlorophyll concentration at the preset detection points, determine the initial red tide anomaly area.

[0052] In some embodiments, the initial red tide anomaly area is determined based on the water temperature and chlorophyll concentration at preset detection points, specifically including:

[0053] Based on the correspondence between chlorophyll concentration and abnormal temperature rise threshold, determine the abnormal temperature rise threshold corresponding to the current chlorophyll concentration;

[0054] Using a preset detection point as the center and a preset distance as the radius, a background area is obtained, and the water temperature at several locations on the edge of the background area is randomly collected; the location water temperature is input into a clustering algorithm, and the average location water temperature of the cluster center with the largest number of clusters is determined as the background water temperature.

[0055] Calculate the difference between the background water temperature and the water body temperature. When it exceeds the abnormal temperature rise threshold, determine the area with the preset detection point as the center and the preset distance as the radius as the preset ratio as the initial red tide abnormal area; where the preset ratio is greater than 0 and less than 1.

[0056] Those skilled in the art will understand that this step can establish a dynamic correspondence between chlorophyll concentration and abnormal temperature rise threshold. For example, when the chlorophyll concentration reaches 10 μg / L, the system automatically matches an abnormal temperature rise threshold of 3.2℃. Additionally, the K-means clustering algorithm can be used to process water temperature data from 32 sampling points within a 500-meter radius, automatically removing deviations (such as interference points from ship wakes). Furthermore, this step delineates the abnormal region using a preset proportional radius (typically 30% of the detection radius), avoiding boundary errors generated by the traditional rectangular grid method.

[0057] Step 120: Based on the biofluorescence bands involved in the red tide, configure the corresponding fluorescence sensors; use a drone equipped with a fluorescence sensor to detect whether there is a fluorescence reaction in the initial red tide abnormal area.

[0058] Among these methods, drones equipped with fluorescence sensors are used to detect whether there is a fluorescent reaction in the initial red tide anomaly area. Specifically, this can be achieved by:

[0059] By using a drone equipped with a fluorescence sensor to switch between different biofluorescence bands in the initial red tide anomaly area, the presence of fluorescence reaction in the initial red tide anomaly area can be detected.

[0060] Those skilled in the art will understand that this step can achieve species-specific detection of red tides by configuring sensors with characteristic bands such as dinoflagellates (685nm) and cyanobacteria (650nm). The UAV automatically switches sensor bands during flight (e.g., the 0.2-second band switching speed of the Turner Cyclops-7), allowing for the simultaneous acquisition of data on multiple algae in a single aerial survey. At a flight altitude of 50m, the UAV can achieve a pixel resolution of 0.1m², providing more accurate monitoring of micro-red tide patches than satellite remote sensing (100m² level). Furthermore, this application can utilize airborne edge computing devices (such as the NVIDIA Jetson TX2) to implement real-time threshold alarms for fluorescence intensity (RFU), resulting in a shorter time from data acquisition to warning issuance.

[0061] Step 130: When a fluorescence reaction is present, the initial red tide anomaly area is identified as a red tide anomaly area; the concentration of the corresponding red tide marker is determined by collecting the reflected fluorescence intensity; the biofluorescence band of the current fluorescence sensor is obtained to determine the red tide marker type of the current red tide anomaly area; the red tide marker concentration and red tide marker type are uploaded to the corresponding preset red tide monitoring terminal.

[0062] Those skilled in the art will understand that this step can determine the concentration of the corresponding red tide marker by setting a fluorescence intensity threshold (e.g., chlorophyll a > 50 RFU); based on a characteristic band database (containing several types of red tide biological spectral fingerprints), species identification can be quickly completed through band matching.

[0063] In some embodiments, the concentration of the corresponding red tide marker is determined by collecting the intensity of the reflected fluorescence, specifically including:

[0064] Through the formula:

[0065] Determine the corresponding red tide marker concentrations;

[0066] in, and This is a preset value.

[0067] This includes acquiring the biofluorescence band of the current fluorescence sensor and determining the type of red tide markers in the current red tide anomaly area, specifically including:

[0068] Obtain the number of red tide marker types corresponding to the current biofluorescence band of the fluorescence sensor. When the number of types is unique, determine the unique red tide marker type as the red tide marker type of the current red tide anomaly area.

[0069] When the number of types is not unique, the type of red tide markers in the current red tide anomaly area can be determined by using a drone equipped with an image recognition device.

[0070] It should be noted that this step utilizes a fluorescence sensor to acquire biofluorescence bands, directly matching them against a pre-set database (containing several types of red tide biological spectral fingerprints). When the number of types is unique, the red tide marker type is immediately determined, reducing manual identification time. When a fluorescence band corresponds to multiple potential species, a drone equipped with an image recognition device is automatically invoked for secondary confirmation, avoiding the risk of misjudgment from single-spectrum detection. The determination method (direct confirmation or image verification) can be autonomously selected based on the band matching results, requiring no manual intervention and improving red tide monitoring efficiency. This step can combine a dual verification mechanism of fluorescence and morphological characteristics to improve the accuracy of red tide biological type identification.

[0071] After acquiring the biofluorescence band of the current fluorescence sensor and determining the type of red tide markers in the current red tide anomaly area, the method further includes:

[0072] Obtain the number of preset detection points corresponding to the red tide anomaly area. When the number of preset detection points is less than the preset value, add a new preset detection point in the red tide anomaly area.

[0073] It should be noted that this step automatically obtains the number of preset monitoring points within the red tide anomaly area. When the number of monitoring points is insufficient (less than the preset number), new monitoring points are automatically added to ensure that the coverage density of the monitoring network meets the requirements. Through monitoring point quantity assessment, the system can autonomously determine whether additional monitoring points are needed, enabling dynamic adjustment of the monitoring network without manual intervention. This mechanism for adding monitoring points ensures continuous and comprehensive monitoring of the red tide anomaly area, avoiding monitoring blind spots caused by insufficient monitoring points and improving the reliability of data collection. Furthermore, the entire process can be fully automated, from monitoring point quantity assessment to the decision-making and execution of adding new monitoring points, achieving automated management of the monitoring process.

[0074] Based on the above description, those skilled in the art will understand that this embodiment can achieve simultaneous acquisition of two parameters (water temperature + chlorophyll concentration) through the coordinated operation of a thermal infrared sensor and a chlorophyll concentration collector. An initial screening mechanism is established based on the correspondence between abnormal temperature rise thresholds and chlorophyll concentration. Combined with a clustering algorithm for background water temperature (taking the mean of the most frequent cluster centers), the initial abnormal area can be quickly delineated with a preset detection point as the center (the radius is reduced by a preset ratio), reducing the workload of subsequent detection.

[0075] This embodiment can be configured with a fluorescence sensor that can switch biological fluorescence bands, and the initial abnormal area can be verified by multi-band scanning. When a fluorescence reaction is detected, the presence of red tide can be confirmed, and the concentration of markers can be quantified by characteristic fluorescence intensity (such as chlorophyll a > 50 RFU threshold), avoiding misjudgment caused by a single sensor.

[0076] This embodiment can rely on a database containing the spectral fingerprints of red tide organisms to achieve rapid species identification through band matching. When the biofluorescence band corresponds to multiple species, an image recognition device is activated for secondary confirmation, forming a dual verification mechanism of "spectral screening - image verification" to improve the accuracy of red tide organism type identification.

[0077] This embodiment can automatically trigger the addition of new detection points based on the number of preset detection points within the red tide anomaly area (when the number is less than the preset value), thus achieving adaptive expansion of the monitoring network. This dynamic adjustment mechanism ensures continuous tracking capability for spreading red tides.

[0078] This embodiment can calculate the concentration of red tide markers using a formula, and automatically associate the concentration data with species type before uploading it to the monitoring terminal, thus meeting the data standard requirements for environmental supervision.

[0079] In addition, this application Figure 2 This application provides an embodiment of a marine red tide disaster early warning system based on unmanned aerial vehicles (UAVs). Figure 2 As shown in the embodiments of this application, the system mainly includes:

[0080] The determination module 210 is used to collect water temperature and chlorophyll concentration at each preset detection point in the area to be detected by a drone equipped with a thermal infrared sensor and a chlorophyll concentration collector; and to determine the initial red tide anomaly area based on the water temperature and chlorophyll concentration at the preset detection points.

[0081] The determining module 210 includes a determining unit.

[0082] It is used to determine the abnormal temperature threshold corresponding to the current chlorophyll concentration based on the correspondence between chlorophyll concentration and abnormal temperature threshold.

[0083] Using a preset detection point as the center and a preset distance as the radius, a background area is obtained, and the water temperature at several locations on the edge of the background area is randomly collected; the location water temperature is input into a clustering algorithm, and the average location water temperature of the cluster center with the largest number of clusters is determined as the background water temperature.

[0084] Calculate the difference between the background water temperature and the water body temperature. When it exceeds the abnormal temperature rise threshold, determine the area with the preset detection point as the center and the preset distance as the radius as the preset ratio as the initial red tide abnormal area; where the preset ratio is greater than 0 and less than 1.

[0085] Those skilled in the art will understand that the determining module 210 can establish a dynamic correspondence between chlorophyll concentration and abnormal temperature rise threshold. For example, when the chlorophyll concentration reaches 10 μg / L, the system automatically matches a temperature difference threshold of 3.2℃. Additionally, the K-means clustering algorithm can be used to process the water temperature data from 32 sampling points within a 500-meter radius, automatically removing deviations (such as interference points from ship wakes). Furthermore, this step delineates the abnormal area using a preset proportional radius (usually 30% of the detection radius), avoiding boundary errors generated by the traditional rectangular grid method.

[0086] The detection module 220 is used to configure corresponding fluorescence sensors based on the biofluorescence bands involved in the red tide; and to detect whether there is a fluorescence reaction in the initial red tide abnormal area by using a drone equipped with a fluorescence sensor.

[0087] Those skilled in the art will understand that the detection module 220 can achieve species-specific detection of red tides by configuring sensors with characteristic bands such as dinoflagellates (685nm) and cyanobacteria (650nm). The UAV automatically switches sensor bands during flight (e.g., the 0.2-second band switching speed of the Turner Cyclops-7), allowing simultaneous acquisition of data on multiple algae in a single aerial survey. At a flight altitude of 50m, the UAV can achieve a pixel resolution of 0.1m², providing more accurate monitoring of micro-red tide patches than satellite remote sensing (100m² level). Furthermore, this application can utilize an airborne edge computing device (such as the NVIDIA Jetson TX2) to implement real-time threshold alarms for fluorescence intensity (RFU), resulting in a shorter time from data acquisition to warning issuance.

[0088] The upload module 230 is used to identify the initial red tide anomaly area as a red tide anomaly area when a fluorescence reaction is present; to determine the corresponding red tide marker concentration by collecting the reflected fluorescence intensity; to obtain the biofluorescence band of the current fluorescence sensor to determine the red tide marker type of the current red tide anomaly area; and to upload the red tide marker concentration and red tide marker type to the corresponding preset red tide monitoring terminal.

[0089] Those skilled in the art will understand that the upload module 230 can determine the concentration of the corresponding red tide marker by setting a fluorescence intensity threshold (e.g., chlorophyll a > 50 RFU); based on the characteristic band database (containing several types of red tide biological spectral fingerprints), species identification can be quickly completed through band matching.

[0090] Upload module 230 includes a concentration determination unit.

[0091] Used in the formula:

[0092] Determine the corresponding red tide marker concentrations;

[0093] in, and This is a preset value.

[0094] Upload module 230 includes a marker determination unit.

[0095] The number of red tide marker types corresponding to the current biofluorescence band of the fluorescence sensor is used to obtain the number of red tide marker types. When the number of types is unique, the unique red tide marker type is determined as the red tide marker type of the current red tide anomaly area.

[0096] When the number of types is not unique, the type of red tide markers in the current red tide anomaly area can be determined by using a drone equipped with an image recognition device.

[0097] Based on the above description, those skilled in the art will understand that this embodiment can establish a dynamic matching relationship between chlorophyll concentration (e.g., 10 μg / L) and abnormal temperature rise threshold (e.g., 3.2℃) through the coordinated operation of a thermal infrared sensor and a chlorophyll concentration collector. The background water temperature data from 32 sampling points within a 500-meter radius are processed using a K-means clustering algorithm, automatically removing interfering values ​​(e.g., ship wakes), and using the mean of the most cluster centers as the background benchmark. Anomaly areas are delineated by a preset proportion (e.g., 30% of the detection radius), reducing boundary errors compared to the traditional rectangular grid method.

[0098] This embodiment can be configured with fluorescence sensors targeting characteristic bands such as dinoflagellates (685nm) and cyanobacteria (650nm). The UAV completes multi-band scanning with a switching speed of 0.2 seconds (e.g., Turner Cyclops-7). A resolution of 0.1m² pixels is achieved at a flight altitude of 50m, providing more accurate identification of micro-red tide patches than satellite remote sensing (100m² level). An airborne edge computing device (e.g., NVIDIA Jetson TX2) triggers real-time fluorescence intensity threshold alarms (e.g., chlorophyll a > 50 RFU), reducing warning delays.

[0099] This embodiment utilizes a database containing spectral fingerprints of several types of red tide organisms to quickly identify species types through band matching. When a band corresponds to multiple species, an image recognition device is invoked for secondary confirmation. Fluorescence intensity is converted into biomarker concentration through formulaic calculation, and the data is automatically associated with species types and uploaded to the monitoring terminal to generate a structured report.

[0100] This embodiment can dynamically trigger the addition of new detection points based on the number of preset detection points in the abnormal area (if the number is insufficient), ensuring the ability to continuously track the red tide spread path.

[0101] This embodiment can automate the entire process from data collection and calculation to species identification, thereby improving monitoring efficiency and accuracy.

[0102] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily 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 invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for early warning of marine red tide disasters based on unmanned aerial vehicles (UAVs), characterized in that, The method includes: Using a drone equipped with a thermal infrared sensor and a chlorophyll concentration collector, water temperature and chlorophyll concentration are collected at various preset detection points in the area to be detected. Based on the water temperature and chlorophyll concentration at the preset detection points, the initial red tide anomaly area is determined; specifically including: Based on the correlation between chlorophyll concentration and abnormal temperature rise threshold, the abnormal temperature rise threshold corresponding to the current chlorophyll concentration is determined. A background area is obtained with a preset detection point as the center and a preset distance as the radius. Water temperatures at several locations along the edge of the background area are randomly collected. These location water temperatures are input into a clustering algorithm, and the average location water temperature at the center of the cluster with the largest number of clusters is determined as the background water temperature. The difference between the background water temperature and the water body temperature is calculated. When the difference exceeds the abnormal temperature rise threshold, an area with a preset detection point as the center and a preset distance as the radius is determined as the initial red tide abnormal area; where the preset proportion is greater than 0 and less than 1. Based on the biofluorescence bands involved in red tides, corresponding fluorescence sensors are configured; drones equipped with fluorescence sensors are used to detect whether there is a fluorescence reaction in the initial red tide anomaly area; When a fluorescence reaction is present, the initial red tide anomaly area is identified as a red tide anomaly area; the concentration of the corresponding red tide marker is determined by collecting the reflected fluorescence intensity; the biofluorescence band of the current fluorescence sensor is obtained to determine the type of red tide marker in the current red tide anomaly area; and the red tide marker concentration and red tide marker type are uploaded to the corresponding preset red tide monitoring terminal.

2. The method for early warning of marine red tide disasters based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, Using drones equipped with fluorescence sensors, the presence of fluorescence reactions in areas of initial red tide anomalies was detected, specifically including: By using a drone equipped with a fluorescence sensor to switch between different biofluorescence bands in the initial red tide anomaly area, the presence of fluorescence reaction in the initial red tide anomaly area can be detected.

3. The method for early warning of marine red tide disasters based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The concentration of the corresponding red tide marker is determined by collecting the intensity of the reflected fluorescence, specifically including: Through the formula: Determine the corresponding red tide marker concentrations; in, and This is a preset value.

4. The method for early warning of marine red tide disasters based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, Acquire the biofluorescence band of the current fluorescence sensor to determine the type of red tide markers in the current red tide anomaly area, specifically including: Obtain the number of red tide marker types corresponding to the current biofluorescence band of the fluorescence sensor. When the number of types is unique, determine the unique red tide marker type as the red tide marker type of the current red tide anomaly area. When the number of types is not unique, the type of red tide markers in the current red tide anomaly area can be determined by using a drone equipped with an image recognition device.

5. The method for early warning of marine red tide disasters based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, After acquiring the biofluorescence band of the current fluorescence sensor and determining the type of red tide marker in the current red tide anomaly area, the method further includes: Obtain the number of preset detection points corresponding to the red tide anomaly area. When the number of preset detection points is less than the preset value, add a new preset detection point in the red tide anomaly area.

6. A marine red tide disaster early warning system based on unmanned aerial vehicles (UAVs), characterized in that, The system includes: The determination module is used to collect water temperature and chlorophyll concentration at each preset detection point in the area to be detected by a drone equipped with a thermal infrared sensor and a chlorophyll concentration collector; and to determine the initial red tide anomaly area based on the water temperature and chlorophyll concentration at the preset detection points. The determination module includes a determination unit, used to determine the abnormal temperature rise threshold corresponding to the current chlorophyll concentration based on the correspondence between chlorophyll concentration and abnormal temperature rise threshold; to obtain a background area with a preset detection point as the center and a preset distance as the radius, and to randomly collect the location water temperature at several locations on the edge of the background area; to input the location water temperature into a clustering algorithm, and to determine the average location water temperature of the cluster center with the most clusters as the background water temperature; to calculate the difference between the background water temperature and the water body temperature, and when it is greater than the abnormal temperature rise threshold, to determine the area with the preset detection point as the center and a preset distance as the radius as a preset proportion as the initial red tide abnormal area; wherein, the preset proportion is greater than 0 and less than 1; The detection module is used to configure corresponding fluorescence sensors based on the biofluorescence bands involved in red tides; A drone equipped with a fluorescence sensor was used to detect whether there was a fluorescent reaction in the initial red tide anomaly area; The upload module is used to identify the initial red tide anomaly area as a red tide anomaly area when a fluorescence reaction is present; to determine the corresponding red tide marker concentration by collecting the reflected fluorescence intensity; to obtain the biofluorescence band of the current fluorescence sensor to determine the red tide marker type of the current red tide anomaly area; and to upload the red tide marker concentration and red tide marker type to the corresponding preset red tide monitoring terminal.

7. The marine red tide disaster early warning system based on unmanned aerial vehicles (UAVs) according to claim 6, characterized in that, The upload module includes a concentration determination unit. Used in the formula: Determine the corresponding red tide marker concentrations; in, and This is a preset value.

8. The marine red tide disaster early warning system based on unmanned aerial vehicles (UAVs) according to claim 6, characterized in that, The upload module includes a marker determination unit. The number of red tide marker types corresponding to the current biofluorescence band of the fluorescence sensor is used to obtain the number of red tide marker types. When the number of types is unique, the unique red tide marker type is determined as the red tide marker type of the current red tide anomaly area. When the number of types is not unique, the type of red tide markers in the current red tide anomaly area can be determined by using a drone equipped with an image recognition device.

Images