Red tide occurrence high-risk area discrimination method and device based on water color satellite
By constructing a variety of red tide discrimination algorithms and generating deterministic indexes, combined with actual measured data evaluation, the problem of uncertainty of a single algorithm in red tide monitoring is solved, and the accuracy and efficiency of the judgment of high-risk areas of red tides are improved.
Patent Information
- Application Number
- CN202411092169.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-08-09
AI Technical Summary
Under the influence of the complex biooptical characteristics of nearshore water bodies, the existing red tide monitoring technology has differences in the location and range of red tide algorithms, resulting in high monitoring uncertainty and it is difficult for users to choose an accurate and reliable algorithm.
By downloading the aqua satellite data for preprocessing, a variety of red tide discrimination algorithms are constructed to generate a red tide determinism index, and the index is evaluated using actual measured data to determine the high-risk zone of red tide, reducing the uncertainty of a single algorithm.
It provides a more accurate location and range of red tide occurrence, reduces monitoring uncertainty, improves the accuracy and efficiency of the judgment of high-risk areas of red tides, and reduces the harm of red tides.
Smart Images

Figure CN119091314A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of satellite remote sensing image processing, and in particular to a method and device for distinguishing high-risk areas for red tide occurrence based on water color satellites. Background Art
[0002] Red tide has become one of the major marine disasters in my country. Using water color remote sensing technology to carry out effective, real-time and accurate red tide monitoring is of great significance to the scientific understanding of red tides, responding to red tide disasters, and avoiding various negative effects caused by red tides. When a red tide occurs in a water body, the red tide water body will undergo significant changes in the reflectance spectrum compared to the surrounding non-red tide water bodies. Therefore, an algorithm can be established using the spectral differences of the red tide water body itself to distinguish between red tide water bodies and non-red tide water bodies. However, due to the complex bio-optical properties of nearshore water bodies and the influence of the algorithm itself, there are certain differences in the location and range of each red tide algorithm. These differences will cause certain troubles in business applications, and users do not know which algorithm has more accurate and reliable judgment results. Summary of the invention
[0003] In view of the above problems, the present invention proposes a method and device for identifying high-risk areas for red tides based on water color satellites. After downloading water color data and preprocessing it, remote sensing reflectivity is used to construct a variety of red tide identification algorithms, and then a red tide certainty index is constructed based on these red tide identification algorithms. Then, the value of the red tide certainty index is evaluated using the measured red tide data. When the red tide certainty index is greater than a certain threshold, it means that the certainty of the occurrence of red tides at the identified location is the highest, and it is a high-risk area for red tides, thereby reducing the uncertainty of a single algorithm in red tide monitoring.
[0004] On the one hand, a method for identifying high-risk areas for red tides based on water color satellites, the specific steps are as follows:
[0005] S1, obtain multi-temporal satellite remote sensing images during the red tide period, pre-process the acquired satellite remote sensing images, and obtain the remote sensing reflectivity of each band of the satellite remote sensing images;
[0006] S2, based on the remote sensing reflectivity, using several red tide discrimination algorithms to respectively perform red tide discrimination, and respectively obtain algorithm discrimination results of whether each pixel of the satellite remote sensing image is a red tide point;
[0007] S3, obtaining the red tide certainty index according to the judgment results of each algorithm, and constructing the algorithm judgment result matrix corresponding to the red tide certainty index based on the algorithm judgment results of all pixels of the satellite remote sensing image;
[0008] S4, obtaining measured red tide data in the same area, comparing the measured red tide data with the corresponding pixel values in the discrimination result matrix of each algorithm one by one, and classifying and counting the comparison results to obtain the corresponding confusion matrix;
[0009] Among them, the comparison results include that the algorithm discrimination result and the red tide measured data are both red tide points, the algorithm discrimination result is a red tide point but the red tide measured data is a non-red tide point, the algorithm discrimination result is a non-red tide point but the red tide measured data is a red tide point, and the algorithm discrimination result and the red tide measured data are both non-red tide points; the total number of red tide points where the algorithm discrimination result and the red tide measured data are both red tide points is TP, the total number of red tide points where the algorithm discrimination result is a red tide point but the red tide measured data is non-red tide points is FP, the total number of red tide points where the algorithm discrimination result is non-red tide points but the red tide measured data is red tide points is FN, and the total number of non-red tide points where the algorithm discrimination result and the red tide measured data are both non-red tide points is TN; TP, FP, FN and TN corresponding to a red tide certainty index value constitute the confusion matrix corresponding to the value;
[0010] S5, evaluate the confusion matrix based on the F-measure algorithm, and obtain the comprehensive evaluation index FM corresponding to each value of the red tide certainty index. The red tide certainty index value with the highest FM value is used as the judgment threshold of the high-risk area of red tide;
[0011] S6, determining the range consisting of all pixels whose red tide certainty index values are greater than or equal to the judgment threshold as a high-risk area for red tide.
[0012] Preferably, in S1, the preprocessing is projection transformation and Mosaic splicing.
[0013] Preferably, in S2, the several red tide discrimination algorithms include a red tide index method, a red tide detection index method, an improved fluorescence height method and a normalized difference chlorophyll index method;
[0014] The results of the algorithm for judging whether each pixel of a satellite remote sensing image is a red tide point using the red tide index method are as follows:
[0015]
[0016] Among them, RI represents the calculated value of the red tide index method of the current pixel; R rs (560) represents the remote sensing reflectance of the current pixel in the 560nm band; R rs (443) represents the remote sensing reflectance of the current pixel in the 443nm band; R rs (490) represents the remote sensing reflectance of the current pixel in the 490nm band;
[0017] RI>4.0 was used as the criterion for the occurrence of red tide in each pixel;
[0018] The results of the algorithm for judging whether each pixel of the satellite remote sensing image is a red tide point by the red tide detection index method are as follows:
[0019]
[0020] Among them, RDI represents the calculated value of the red tide detection index method of the current pixel; R rs (λ1) represents the remote sensing reflectance of the current pixel in the λ1 band, λ1=665nm; R rs (λ2) represents the remote sensing reflectance of the current pixel in the λ2 band, λ2 = 560nm; R rs (λ3) represents the remote sensing reflectance of the λ3 band of the current pixel, λ3=753nm;
[0021] RDI>0.16 was used as the criterion for the occurrence of red tide in each pixel;
[0022] The improved fluorescence height method is used to determine whether each pixel in the satellite remote sensing image is a red tide point. The results are as follows:
[0023]
[0024] Where MFLH represents the calculated value of the improved fluorescence height method of the current pixel; nL w (681) represents the water-leaving radiance of the current pixel in the 681nm band; nL w (665) represents the water-leaving radiance of the current pixel in the 665nm band; nL w (754) represents the water-leaving radiance of the current pixel in the 754nm band; nL represents the water-leaving radiance of the λ band. W (λ)=R rs (λ)×F0,R rs (λ) represents the remote sensing reflectance in the λ band, and F0 is the solar irradiance outside the atmosphere;
[0025] Using MFLH>0.4mW cm -2 μm -1 sr -1 And b bp / MFLH<0.2mW cm -2 μm -1 sr -1 As the criterion for the occurrence of red tide in each pixel, b bp is the backscattering coefficient;
[0026] The results of the algorithm for judging whether each pixel of satellite remote sensing images is a red tide point by using the normalized difference chlorophyll index method are as follows:
[0027]
[0028] Among them, NDCI represents the calculated value of the normalized difference chlorophyll index method of the current pixel; R rs (708) represents the remote sensing reflectance of the current pixel in the 708nm band; R rs(665) represents the remote sensing reflectance of the current pixel in the 665nm band;
[0029] NDCI>0 is used as the criterion for judging the occurrence of red tide in each pixel.
[0030] Preferably, in S3, the red tide certainty index is obtained according to the discrimination results of each algorithm, specifically as follows:
[0031] When a pixel in the satellite remote sensing image is judged as a red tide by one of the red tide index method, the red tide detection index method, the improved fluorescence height method and the normalized difference chlorophyll index method, the red tide deterministic index RTCI is set to 1; if a pixel in the image is judged as a red tide by two of the four red tide remote sensing discrimination algorithms, the RTCI is set to 2; if a pixel in the image is judged as a red tide by three of the four red tide remote sensing discrimination algorithms, the RTCI is set to 2 or 3; if a pixel in the image is judged as a red tide by four of the four red tide remote sensing discrimination algorithms, the RTCI is set to 2, 3 or 4.
[0032] Preferably, in S5, the comprehensive evaluation index FM is expressed as follows:
[0033]
[0034] On the other hand, a device for identifying high-risk areas of red tide based on water color satellites includes the following:
[0035] The satellite remote sensing image acquisition and preprocessing module is used to acquire multi-temporal satellite remote sensing images during the red tide period, preprocess the acquired satellite remote sensing images, and obtain the remote sensing reflectivity of each band of the satellite remote sensing images;
[0036] A red tide discrimination algorithm module is used to discriminate red tides using several red tide discrimination algorithms based on the remote sensing reflectivity, and obtain algorithm discrimination results of whether each pixel of the satellite remote sensing image is a red tide point;
[0037] The red tide deterministic index module is used to obtain the red tide deterministic index according to the judgment results of each algorithm, and to construct the algorithm judgment result matrix corresponding to the red tide deterministic index based on the algorithm judgment results of all pixels of the satellite remote sensing image;
[0038] The measured data comparison module is used to obtain the measured data of red tide in the same area, compare the measured data of red tide with the corresponding pixel values in the discrimination result matrix of each algorithm one by one, and classify and count the comparison results to obtain the corresponding confusion matrix;
[0039] Among them, the comparison results include that the algorithm discrimination result and the red tide measured data are both red tide points, the algorithm discrimination result is a red tide point but the red tide measured data is a non-red tide point, the algorithm discrimination result is a non-red tide point but the red tide measured data is a red tide point, and the algorithm discrimination result and the red tide measured data are both non-red tide points; the total number of red tide points where the algorithm discrimination result and the red tide measured data are both red tide points is TP, the total number of red tide points where the algorithm discrimination result is a red tide point but the red tide measured data is non-red tide points is FP, the total number of red tide points where the algorithm discrimination result is non-red tide points but the red tide measured data is red tide points is FN, and the total number of non-red tide points where the algorithm discrimination result and the red tide measured data are both non-red tide points is TN; TP, FP, FN and TN corresponding to a red tide certainty index value constitute the confusion matrix corresponding to the value;
[0040] The algorithm discrimination result evaluation module is used to evaluate the confusion matrix based on the F-measure algorithm, obtain the comprehensive evaluation index FM corresponding to each value of the red tide certainty index, and take the red tide certainty index value with the highest FM value as the judgment threshold of the high-risk area of red tide;
[0041] The red tide high risk area determination module is used to determine the range composed of all pixels whose red tide certainty index values are greater than or equal to the determination threshold as the red tide high risk area.
[0042] Compared with the prior art, the present invention has the following beneficial effects:
[0043] (1) The present invention comprehensively utilizes a variety of red tide discrimination algorithms to generate a red tide certainty index, which can provide a relatively accurate location and range of red tide occurrence in business applications. The threshold of the certainty index is selected after evaluation of field measured data, and the location of the high-risk area of red tide occurrence is further clarified, thereby providing a reference basis for red tide emergency decision-making and management;
[0044] (2) The present invention constructs a variety of red tide discrimination algorithms and constructs a red tide certainty index based on the multiple red tide discrimination algorithms, thereby reducing the uncertainty of a single algorithm in red tide monitoring and providing a more stable and reliable red tide monitoring service;
[0045] (3) The present invention constructs a business-oriented and automated identification process for high-risk areas for red tide occurrence, which can improve the accuracy and efficiency of identifying high-risk areas for red tide occurrence, thereby reducing the harm of red tide. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] The present invention is further described in detail below in conjunction with the accompanying drawings;
[0047] Figure 1 This is a flow chart of a method for identifying high-risk areas for red tide occurrence based on water color satellites according to an embodiment of the present invention;
[0048] Figure 2A technical process for distinguishing high-risk areas for red tide occurrences of a method for distinguishing high-risk areas for red tide occurrences based on a water color satellite in an embodiment of the present invention;
[0049] Figure 3 The present invention provides a method for distinguishing high-risk areas for red tide occurrence based on water color satellites, and shows spatial distribution maps of red tide certainty indexes in different phases; (a) shows the spatial distribution map of red tide certainty index on May 21, 2019; (b) shows the spatial distribution map of red tide certainty index on May 23, 2019; (c) shows the spatial distribution map of red tide certainty index on April 29, 2020; (d) shows the spatial distribution map of red tide certainty index on August 14, 2020; (f) shows the spatial distribution map of red tide certainty index on August 15, 2020; each pixel only shows the highest red tide certainty index value;
[0050] Figure 4 This is a module diagram of a device for identifying high-risk areas for red tide occurrence based on water color satellites according to an embodiment of the present invention. DETAILED DESCRIPTION
[0051] The present invention is further described below through specific implementation modes.
[0052] like Figure 1 and Figure 2 As shown, a method for identifying high-risk areas for red tides based on water color satellites is shown, and the specific steps are as follows.
[0053] S1, obtain multi-temporal satellite remote sensing images during the red tide period, pre-process the obtained satellite remote sensing images, and obtain the remote sensing reflectivity of each band of the satellite remote sensing images.
[0054] During the high incidence period of red tide (March to June each year), the information products of the OLCI (Ocean and Land Colour Instrument) water colour sensor inversion carried by the Sentinel-3 (Sentinel-3) water colour satellite were downloaded. The data set format is OLCILevel 2Ocean Colour Full Resolution in NTC-Sentinel-3. The spatial resolution of the OLCI product can reach 300×300m. The downloaded L2 files were preprocessed in the Sentinel Application Platform (SNAP) software, that is, the GPT command line was used to perform batch projection transformation and Mosaic splicing, and the projection method used was Equidistant Cylindrical, so as to obtain the remote sensing reflectance of each band required for the red tide discrimination algorithm.
[0055] S2, based on the remote sensing reflectivity, using several red tide discrimination algorithms to respectively perform red tide discrimination, and obtain algorithm discrimination results of whether each pixel of the satellite remote sensing image is a red tide point.
[0056] Specifically, the red tide identification algorithms used include the red tide index method, the red tide detection index method, the improved fluorescence height method and the normalized difference chlorophyll index method. The expressions of each red tide identification algorithm are as follows:
[0057] The Red Tide Index method uses remote sensing data of water color, such as the remote sensing reflectance (R rs ) is constructed, where the blue and green bands of the RI calculation formula are both subtracted from R rs (443) is to eliminate the influence of turbid water. The expression of red tide index is as follows:
[0058]
[0059] Among them, RI represents the calculated value of the red tide index method of the current pixel; R rs (560) represents the remote sensing reflectance of the current pixel in the 560nm band; R rs (443) represents the remote sensing reflectance of the current pixel in the 443nm band; R rs (490) represents the remote sensing reflectance of the current pixel in the 490nm band.
[0060] RI>4.0 is used as the criterion for judging the occurrence of red tide in each pixel.
[0061] The expression of Red Tide Detection Index is as follows:
[0062]
[0063] Among them, RDI represents the calculated value of the red tide detection index method of the current pixel; R rs (λ1) represents the remote sensing reflectance of the current pixel in the λ1 band, λ1=665nm; R rs (λ2) represents the remote sensing reflectance of the current pixel in the λ2 band, λ2 = 560nm; R rs (λ3) represents the remote sensing reflectance of the λ3 band of the current pixel, λ3 = 753nm.
[0064] The modified fluorescence line height method is a spectral difference method that uses the water-free radiance (nL) of three bands: 665nm, 754nm and 681nm. w), whose expression is shown as follows:
[0065]
[0066] Where MFLH represents the calculated value of the improved fluorescence height method of the current pixel; nL w (681) represents the water-leaving radiance of the current pixel in the 681nm band; nL w (665) represents the water-leaving radiance of the current pixel in the 665nm band; nL w (754) represents the water-leaving radiance of the current pixel in the 754nm band; nL represents the water-leaving radiance of the λ band. W (λ)=R rs (λ)×F0,R rs (λ) represents the remote sensing reflectivity in the λ band, and F0 is the solar irradiance outside the atmosphere.
[0067] After experimental optimization using field observation data, MFLH>0.4mW cm -2 μm -1 sr -1 And b bp / MFLH<0.2mW cm -2 μm -1 sr -1 As the criterion for judging the occurrence of red tide in each pixel, b bp is the backscattering coefficient.
[0068] The Normalized Difference Chlorophyll Index is as follows:
[0069]
[0070] Among them, NDCI represents the calculated value of the normalized difference chlorophyll index method of the current pixel; R rs (708) represents the remote sensing reflectance of the current pixel in the 708nm band; R rs (665) represents the remote sensing reflectance of the current pixel in the 665nm band.
[0071] NDCI>0 is used as the criterion for judging the occurrence of red tide in each pixel.
[0072] S3, obtaining the red tide deterministic index according to the judgment results of each algorithm, and constructing the algorithm judgment result matrix corresponding to the red tide deterministic index based on the algorithm judgment results of all pixels of the satellite remote sensing image.
[0073] Based on the above four red tide discrimination algorithms, a comprehensive discrimination index, namely the Red Tide Certainty Index (RTCI), is proposed. The definition of this index is as follows: when a certain pixel in the satellite remote sensing image is discriminated as a red tide by one of the red tide index method, red tide detection index method, improved fluorescence height method and normalized difference chlorophyll index method, the RTCI is set to 1; if a certain pixel in the image is discriminated as a red tide by two of the four red tide remote sensing discrimination algorithms, the RTCI is set to 2; if a certain pixel in the image is discriminated as a red tide by three of the four red tide remote sensing discrimination algorithms, the RTCI is set to 2 or 3; if a certain pixel in the image is discriminated as a red tide by four of the four red tide remote sensing discrimination algorithms, the RTCI is set to 2, 3 or 4.
[0074] S4, obtaining measured red tide data in the same area, comparing the measured red tide data with the corresponding pixel values in the discrimination result matrix of each algorithm one by one, and classifying and counting the comparison results to obtain the corresponding confusion matrix;
[0075] Specifically, the steps to obtain actual red tide observation data are as follows:
[0076] From the red tide information released by the China Marine Disaster Bulletin or various places, the recorded red tide events are collected, and the information such as the time, location and affected area of each red tide event is extracted. However, the red tide locations recorded in the general marine disaster bulletin do not have accurate longitude and latitude, and the area provided is generally statistical information during the entire red tide event. Simply using this information to evaluate the true value of the algorithm effectiveness will obviously have a certain deviation. This embodiment uses the red tide information provided by the Marine Disaster Bulletin, combined with the remote sensing FRGB map (pseudo-color map) to jointly judge and select the red tide observation data used to evaluate the effectiveness of the algorithm.
[0077] The specific method is as follows: First, select the corresponding remote sensing image according to the date and location recorded in the bulletin, and try to select images without clouds or with less cloud impact; then use the three bands of OLCI: R (865nm), G (560nm) and B (443nm) to generate a false color (False RGB) image. Through observation and comparison, it is found that the points in the FRGB image that appear red or reddish brown and have significant differences from the surrounding pixels are consistent with the red tide positions reported in the bulletin. Therefore, using the SNPP software, by drawing a polygonal ROI (Region Of Interest) area, the corresponding points in the red or reddish brown abnormal area on the FRGB image are selected as the true value of the red tide (Red Tide), and the points around it where the water color is not abnormal are selected as the true value of the non-red tide point (Non-Red Tide).
[0078] Specifically, the measured red tide data are compared with the corresponding pixel values in the discrimination result matrix of each algorithm one by one, and the comparison results are classified and counted to obtain the corresponding confusion matrix as follows:
[0079] The true value of red tide, that is, the actual observed red tide, is recorded as RT; the true value of non-red tide point, that is, the actual observed non-red tide, is recorded as NRT; the results of each red tide discrimination algorithm are also divided into two types: algorithm-discriminated red tide (red tide, recorded as rt) or algorithm-discriminated non-red tide (non-red tide, recorded as nrt). The binary classification confusion matrix formed in this way is shown in the following table.
[0080] Table 1 Confusion matrix of red tide discrimination results
[0081]
[0082] As shown in Table 1, there are four situations when comparing the algorithm judgment results with the actual observation values: 1) both the algorithm judgment and the actual observation are judged as red tide points (rt-RT); 2) the algorithm judgment result is a red tide point, but the actual red tide does not occur (rt-NRT); 3) the algorithm judgment result is a non-red tide point, while the actual observation value is a red tide point (nrt-RT); 4) both the algorithm judgment result and the actual observation value are non-red tide points (nrt-NRT).
[0083] The number of all comparison situations in the comparison results is accumulated respectively, and the total number of red tide points where both the algorithm judgment results and the red tide measured data are red tide points is TP, the total number of red tide points where the algorithm judgment results are red tide points but the red tide measured data are non-red tide points is FP, the total number of red tide points where the algorithm judgment results are non-red tide points but the red tide measured data are red tide points is FN, and the total number of non-red tide points where both the algorithm judgment results and the red tide measured data are non-red tide points is TN.
[0084] S5, evaluate the confusion matrix based on the F-measure algorithm, and obtain the comprehensive evaluation index FM corresponding to each value of the red tide certainty index. The red tide certainty index value with the highest FM value is used as the judgment threshold for the high-risk area of red tide.
[0085] Specifically, the method of evaluating the confusion matrix of each value of the red tide certainty index based on the F-measure algorithm is as follows:
[0086] Sensitivity refers to the correct proportion of the red tide identification algorithm in identifying red tide points, which measures the algorithm's ability to identify actual red tide seawater. The sensitivity formula is as follows:
[0087]
[0088] Wherein, Sensitivity represents the calculated sensitivity value.
[0089] Specificity refers to the correct proportion of red tide discrimination algorithms in identifying non-red tide points, which measures the algorithm's ability to identify seawater without red tide. The specificity formula is as follows:
[0090]
[0091] Wherein, Specificity represents the calculated value of specificity.
[0092] Accuracy refers to the proportion of red tide points to the total number of red tide points and non-red tide points that are correctly distinguished. The accuracy formula is as follows:
[0093]
[0094] Among them, Specificity represents the accuracy calculation value.
[0095] The comprehensive evaluation index is the harmonic average of sensitivity and accuracy, and its calculation formula is as follows:
[0096]
[0097] Among them, FM (i.e. F-measure) represents a comprehensive evaluation index. Usually, the closer the FM value is to 1, the better the algorithm effect is.
[0098] The false negative rate is the ratio of red tide points that the algorithm mistakenly identifies as non-red tide points; the false positive rate is the ratio of non-red tide points that the algorithm mistakenly identifies as red tide points. The formulas for the false negative rate and false positive rate are as follows:
[0099]
[0100] Among them, False neg.(%) represents the percentage value of the missed alarm rate; False pos.(%) represents the percentage value of the false alarm rate.
[0101] This example takes the East China Sea as an example, and selects OLCI remote sensing images of five phases with relatively less cloud cover to carry out evaluations when RTCI takes different values. These five remote sensing images recorded the red tide of Prorocentrum scintillans / Noctiluca scintillans in the Xiangshan-Yushan waters of Ningbo on May 21 and May 23, 2019, the red tide of Prorocentrum scintillans in the East China Sea in the Nanji-Shipu-Yushan waters of Wenzhou on April 29, 2020, and the red tide of diatoms of Rhizoctonia fasciatus in the waters east of Zhujiajian, Zhoushan on August 14 and August 15, 2020.
[0102] Figure 3The spatial distribution diagram of the deterministic index of red tide in the above five phases is shown in Table 2. The evaluation results of the F-measure algorithm when RTCI takes different values are given. Generally speaking, when RTCI is 2, its FM value is the highest (0.94), and the underreporting rate and false alarm rate are relatively low; as the RTCI value increases, the FM value does not increase, but decreases. This is because as the RTCI value increases, there are higher requirements for algorithm consistency. Only when the consistency of the four algorithms is relatively good, the FM value will increase with the increase of the RTCI value. When RTCI is 1, the use of a single algorithm may overestimate the impact range of red tides in complex nearshore waters, resulting in a lower FM value. It can be seen that the reasonable value of RTCI (as shown in Table 2, the judgment threshold of this embodiment takes RTCI = 2) can effectively reduce misjudgment, thereby improving the accuracy of red tide discrimination (FM value).
[0103] Table 2 F-measure statistical results for different RTCI values
[0104]
[0105] Among them, Sen. is the abbreviation of Sensibility, and Spe. is the abbreviation of Specificity.
[0106] S6, determining the range consisting of all pixels whose red tide certainty index values are greater than or equal to the judgment threshold as a high-risk area for red tide.
[0107] In this embodiment, it is found through evaluation that when RTCI ≥ 2, the uncertainty of a single algorithm in red tide monitoring can be reduced to a certain extent. The certainty of the occurrence of red tide in these pixels is the highest, and they can be identified as high-risk areas for red tide occurrence.
[0108] This embodiment comprehensively utilizes a variety of red tide discrimination algorithms to generate a red tide certainty index, which can provide a relatively accurate location and range of red tide occurrence in business applications. The threshold of the red tide certainty index is selected after evaluation through on-site measured data, and the location of the high-risk area for red tide occurrence is further clarified, thereby providing a reference basis for red tide emergency decision-making management; this embodiment constructs a business-based and automated discrimination process for high-risk areas for red tide occurrence, which can improve the accuracy and efficiency of the discrimination of high-risk areas for red tide occurrence, thereby reducing the harm of red tides.
[0109] like Figure 4 As shown, the present invention also discloses a device for identifying high-risk areas for red tide occurrence based on water color satellites, comprising:
[0110] The satellite remote sensing image acquisition and preprocessing module 401 is used to acquire multi-temporal satellite remote sensing images during the red tide period, preprocess the acquired satellite remote sensing images, and obtain the remote sensing reflectivity of each band of the satellite remote sensing images.
[0111] The red tide identification algorithm module 402 is used to use several red tide identification algorithms to perform red tide identification based on the remote sensing reflectivity, and obtain algorithm identification results of whether each pixel of the satellite remote sensing image is a red tide point.
[0112] The red tide certainty index module 403 is used to obtain the red tide certainty index according to the judgment results of each algorithm, and to construct an algorithm judgment result matrix corresponding to the red tide certainty index based on the algorithm judgment results of all pixels of the satellite remote sensing image.
[0113] The measured data comparison module 404 is used to obtain the measured data of red tide in the same area, compare the measured data of red tide with the corresponding pixel values in the discrimination result matrix of each algorithm one by one, and classify and count the comparison results to obtain the corresponding confusion matrix.
[0114] Among them, the comparison results include that both the algorithm judgment result and the red tide measured data are red tide points, the algorithm judgment result is a red tide point but the red tide measured data is a non-red tide point, the algorithm judgment result is a non-red tide point but the red tide measured data is a red tide point, and the algorithm judgment result and the red tide measured data are both non-red tide points; the total number of red tide points where both the algorithm judgment result and the red tide measured data are red tide points is TP, the total number of red tide points where the algorithm judgment result is a red tide point but the red tide measured data are non-red tide points is FP, the total number of non-red tide points where the algorithm judgment result is a red tide point but the red tide measured data are red tide points is FN, and the total number of non-red tide points where both the algorithm judgment result and the red tide measured data are non-red tide points is TN; TP, FP, FN and TN corresponding to a red tide certainty index value constitute the confusion matrix corresponding to the value.
[0115] The algorithm discrimination result evaluation module 405 is used to evaluate the confusion matrix based on the F-measure algorithm, obtain the comprehensive evaluation index FM corresponding to each value of the red tide certainty index, and take the red tide certainty index value with the highest FM value as the judgment threshold for the high risk area of red tide.
[0116] The red tide high risk area determination module 406 is used to determine the range formed by all pixels whose red tide certainty index values are greater than or equal to the determination threshold as a red tide high risk area.
[0117] A specific implementation of a device for distinguishing high-risk areas for red tide occurrence based on water color satellites. The same method for distinguishing high-risk areas for red tide occurrence based on water color satellites will not be repeated in this embodiment.
[0118] The above is only a specific implementation of the present invention, but the design concept of the present invention is not limited thereto. Any non-substantial changes to the present invention using this concept shall be deemed as an infringement of the protection scope of the present invention.
Claims
1. A method for identifying high-risk areas for red tides based on water color satellites, characterized in that: The specific steps are as follows: S1, obtain multi-temporal satellite remote sensing images during the red tide period, pre-process the acquired satellite remote sensing images, and obtain the remote sensing reflectivity of each band of the satellite remote sensing images; S2, based on the remote sensing reflectivity, using several red tide discrimination algorithms to respectively perform red tide discrimination, and respectively obtain algorithm discrimination results of whether each pixel of the satellite remote sensing image is a red tide point; S3, obtaining the red tide certainty index according to the judgment results of each algorithm, and constructing the algorithm judgment result matrix corresponding to the red tide certainty index based on the algorithm judgment results of all pixels of the satellite remote sensing image; S4, obtaining measured red tide data in the same area, comparing the measured red tide data with the corresponding pixel values in the discrimination result matrix of each algorithm one by one, and classifying and counting the comparison results to obtain the corresponding confusion matrix; Among them, the comparison results include that the algorithm discrimination result and the red tide measured data are both red tide points, the algorithm discrimination result is a red tide point but the red tide measured data is a non-red tide point, the algorithm discrimination result is a non-red tide point but the red tide measured data is a red tide point, and the algorithm discrimination result and the red tide measured data are both non-red tide points; the total number of red tide points where the algorithm discrimination result and the red tide measured data are both red tide points is TP, the total number of red tide points where the algorithm discrimination result is a red tide point but the red tide measured data is non-red tide points is FP, the total number of red tide points where the algorithm discrimination result is non-red tide points but the red tide measured data is red tide points is FN, and the total number of non-red tide points where the algorithm discrimination result and the red tide measured data are both non-red tide points is TN; TP, FP, FN and TN corresponding to a red tide certainty index value constitute the confusion matrix corresponding to the value; S5, evaluate the confusion matrix based on the F-measure algorithm, and obtain the comprehensive evaluation index FM corresponding to each value of the red tide certainty index. The red tide certainty index value with the highest FM value is used as the judgment threshold of the high-risk area of red tide; S6, determining the range consisting of all pixels whose red tide certainty index values are greater than or equal to the judgment threshold as a high-risk area for red tide.
2. The method for identifying high-risk areas for red tide occurrence based on water color satellite according to claim 1 is characterized in that: In S1, the preprocessing includes projection transformation and Mosaic splicing.
3. The method for identifying high-risk areas for red tide occurrence based on water color satellite according to claim 1 is characterized in that: In S2, the several red tide identification algorithms include a red tide index method, a red tide detection index method, an improved fluorescence height method, and a normalized difference chlorophyll index method; The results of the algorithm for judging whether each pixel of a satellite remote sensing image is a red tide point using the red tide index method are as follows: Among them, RI represents the calculated value of the red tide index method of the current pixel; R rs (560) represents the remote sensing reflectance of the current pixel in the 560nm band; R rs (443) represents the remote sensing reflectance of the current pixel in the 443nm band; R rs (490) represents the remote sensing reflectance of the current pixel in the 490nm band; RI>4.0 was used as the criterion for the occurrence of red tide in each pixel; The results of the algorithm for judging whether each pixel of the satellite remote sensing image is a red tide point by the red tide detection index method are as follows: Among them, RDI represents the calculated value of the red tide detection index method of the current pixel; R rs (λ1) represents the remote sensing reflectance of the current pixel in the λ1 band, λ1=665nm; R rs (λ2) represents the remote sensing reflectance of the current pixel in the λ2 band, λ2 = 560nm; R rs (λ3) represents the remote sensing reflectance of the λ3 band of the current pixel, λ3=753nm; RDI>0.16 was used as the criterion for the occurrence of red tide in each pixel; The improved fluorescence height method is used to determine whether each pixel in the satellite remote sensing image is a red tide point. The results are as follows: Where MFLH represents the calculated value of the improved fluorescence height method of the current pixel; nL w (681) represents the water-leaving radiance of the current pixel in the 681nm band; nL w (665) represents the water-leaving radiance of the current pixel at 665 nm; nL w (754) represents the water-leaving radiance of the current pixel in the 754nm band; nL represents the water-leaving radiance of the λ band. W (λ)=R rs (λ)×F0,R rs (λ) represents the remote sensing reflectance in the λ band, and F0 is the solar irradiance outside the atmosphere; Using MFLH>0.4mW cm -2 μm -1 sr -1 And b bp / MFLH<0.2mW cm -2 μm -1 sr -1 As the criterion for judging the occurrence of red tide in each pixel, b bp is the backscattering coefficient; The results of the algorithm for judging whether each pixel of satellite remote sensing images is a red tide point by using the normalized difference chlorophyll index method are as follows: Among them, NDCI represents the calculated value of the normalized difference chlorophyll index method of the current pixel; R rs (708) represents the remote sensing reflectance of the current pixel in the 708nm band; R rs (665) represents the remote sensing reflectance of the current pixel in the 665nm band; NDCI>0 is used as the criterion for judging the occurrence of red tide in each pixel.
4. The method for identifying high-risk areas for red tide occurrence based on water color satellite according to claim 3 is characterized in that: In S3, the red tide certainty index is obtained according to the judgment results of each algorithm, which is as follows: When a pixel in the satellite remote sensing image is judged as a red tide by one of the red tide index method, the red tide detection index method, the improved fluorescence height method and the normalized difference chlorophyll index method, the red tide deterministic index RTCI is set to 1; if a pixel in the image is judged as a red tide by two of the four red tide remote sensing discrimination algorithms, the RTCI is set to 2; if a pixel in the image is judged as a red tide by three of the four red tide remote sensing discrimination algorithms, the RTCI is set to 2 or 3; if a pixel in the image is judged as a red tide by four of the four red tide remote sensing discrimination algorithms, the RTCI is set to 2, 3 or 4.
5. The method for identifying high-risk areas for red tide occurrence based on water color satellite according to claim 4 is characterized in that: In S5, the comprehensive evaluation index FM is expressed as follows:
6. A device for identifying high-risk areas of red tide based on water color satellites, comprising the following: The satellite remote sensing image acquisition and preprocessing module is used to acquire multi-temporal satellite remote sensing images during the red tide period, preprocess the acquired satellite remote sensing images, and obtain the remote sensing reflectivity of each band of the satellite remote sensing images; A red tide discrimination algorithm module is used to discriminate red tides using several red tide discrimination algorithms based on the remote sensing reflectivity, and obtain algorithm discrimination results of whether each pixel of the satellite remote sensing image is a red tide point; The red tide deterministic index module is used to obtain the red tide deterministic index according to the judgment results of each algorithm, and to construct the algorithm judgment result matrix corresponding to the red tide deterministic index based on the algorithm judgment results of all pixels of the satellite remote sensing image; The measured data comparison module is used to obtain the measured data of red tide in the same area, compare the measured data of red tide with the corresponding pixel values in the discrimination result matrix of each algorithm one by one, and classify and count the comparison results to obtain the corresponding confusion matrix; Among them, the comparison results include that the algorithm discrimination result and the red tide measured data are both red tide points, the algorithm discrimination result is a red tide point but the red tide measured data is a non-red tide point, the algorithm discrimination result is a non-red tide point but the red tide measured data is a red tide point, and the algorithm discrimination result and the red tide measured data are both non-red tide points; the total number of red tide points where the algorithm discrimination result and the red tide measured data are both red tide points is TP, the total number of red tide points where the algorithm discrimination result is a red tide point but the red tide measured data is non-red tide points is FP, the total number of red tide points where the algorithm discrimination result is non-red tide points but the red tide measured data is red tide points is FN, and the total number of non-red tide points where the algorithm discrimination result and the red tide measured data are both non-red tide points is TN; TP, FP, FN and TN corresponding to a red tide certainty index value constitute the confusion matrix corresponding to the value; The algorithm discrimination result evaluation module is used to evaluate the confusion matrix based on the F-measure algorithm, obtain the comprehensive evaluation index FM corresponding to each value of the red tide certainty index, and take the red tide certainty index value with the highest FM value as the judgment threshold of the high-risk area of red tide; The red tide high risk area determination module is used to determine the range composed of all pixels whose red tide certainty index values are greater than or equal to the determination threshold as the red tide high risk area.
Citation Information
Patent Citations
Offshore ocean red tide identification method
CN113092383A
Red tide identification method based on remote sensing image, medium and system
CN118072182A
