Global chlorophyll remote sensing inversion optimization method and device covering high latitude area

Abstract

The invention discloses a global chlorophyll a remote sensing inversion optimization method and device covering a high latitude area, and the method comprises the steps: carrying out the space-time matching of sea surface temperature based on an actual measurement data set of the actual measurement remote sensing reflectivity and the actual measurement chlorophyll a concentration, and carrying out the integrated construction of a training data set, the measured remote sensing reflectivity (412 nm, 443 nm, 490 nm, 555 nm and 670 nm), the sea surface temperature, the longitude, the latitude and the measured chlorophyll a concentration of five wave bands are included; a BP neural network is applied, a global chlorophyll a remote sensing inversion model covering a high latitude area is constructed, the training data set is adopted for training, an initial weight and a threshold value of the BP neural network are obtained through iterative solution optimization of a genetic algorithm, and a trained global chlorophyll a remote sensing inversion model is obtained; and inputting the satellite remote sensing reflectivity to be inverted and the matched sea surface temperature, longitude and latitude into the trained global chlorophyll a remote sensing inversion model, and performing inversion to obtain the chlorophyll a concentration. The method can effectively improve the global remote sensing inversion precision of chlorophyll a.

Description

Technical Field

[0001] The present invention relates to the field of ocean color remote sensing, and in particular to a global chlorophyll a remote sensing inversion optimization method and device covering high latitudes. Background Art

[0002] The North and South Poles are sensitive areas to climate change, especially the Arctic, where the warming rate is more than twice the global average. With the melting of the ice sheet surface and the increase in extreme weather events, the marine ecosystems of the North and South Poles have undergone great changes. Chlorophyll a (Chla) is a common indicator of phytoplankton biomass in water bodies. Long-term, continuous and large-scale observation of Chla in water bodies through satellite remote sensing is an important means to study the response of marine ecosystems in high-latitude areas to climate change. The existing remote sensing Chla standard algorithm is an empirical algorithm based on the blue-green band ratio of remote sensing reflectance (Rrs). According to the radiation transfer theory, Rrs is determined by water absorption and backscattering signals. In the vast ocean water body, phytoplankton is the main water color component, and phytoplankton absorption dominates the changes in water optical signals. Moreover, the functional relationship between phytoplankton light absorption coefficient and Chla is relatively stable, and the standard empirical algorithm has good inversion performance. However, the water environment in high-latitude areas is special. Due to long-term adaptation to low temperature and weak light conditions, the concentration of pigments in phytoplankton cells is high, and the light absorption capacity corresponding to the same magnitude of Chla is significantly lower than that in other sea areas. At the same time, similar to other offshore areas, some areas in high latitudes are affected by runoff input and have high CDOM content. Affected by the special optical properties of water bodies, the standard remote sensing chlorophyll Chlaa products in high latitudes have serious overestimation in low-value areas and underestimation in high-value areas.

[0003] Based on the field observation of optical data of water bodies in high latitudes, researchers have developed different Chla optimization algorithms for high latitudes. One of the optimization algorithms is to revise the parameters of the standard empirical algorithm by using the matching blue-green band Rrs and Chla, and propose the OC4L algorithm suitable for the Arctic Ocean waters. Another optimization algorithm is to further subdivide the Arctic Ocean into two areas, interior shelves and inflow / outflow shelves, based on a large amount of field observation data, to debug the parameters of the standard empirical algorithm in different areas and establish the AO.Reg.emp algorithm; in addition, genetic algorithms are applied based on multi-band Rrs, Chla, absorption, and scattering observation data to adjust the parameters of the semi-analytical algorithm GSM regionally and optimize the establishment of the AO.GSM algorithm. These high-latitude Chla inversion algorithms are basically based on the regional optimization of algorithm parameters. Although they solve the particularity of the optical properties of water bodies in high latitudes and improve the inversion accuracy of Chla in high latitudes, the algorithms all have regional application limitations. To date, there is still a lack of algorithms that can cover high latitudes and be applicable to other regions of the world.

[0004] The use of multi-band spectra instead of blue-green band spectra as algorithm input has been widely used in complex water bodies such as nearshore and lakes. Combining the advantages of machine learning in nonlinear fitting, the introduction of multi-band remote sensing reflectance in addition to the blue and green bands, especially the ultraviolet band Rrs412, can better separate the contribution of non-pigment components such as CDOM or suspended sediment and phytoplankton to remote sensing signals, and reduce the inversion uncertainty of Chla to a certain extent. However, the inversion of Chla using only multi-band spectra cannot eliminate the inversion uncertainty caused by the temporal and spatial changes in the optical characteristics of water bodies (such as the difference in the absorption capacity of phytoplankton in high latitudes and non-high latitudes). Sea surface temperature (SST) is an important factor affecting the population structure, pigment composition and growth rate of phytoplankton. At the same time, SST is affected by solar radiation, wind, ocean currents and other factors, has significant temporal and spatial distribution characteristics, and is an important indicator of water mass distribution. Some studies have used the correlation between remote sensing SST and Chla, using remote sensing SST as an auxiliary parameter and applying the spatiotemporal Bayesian maximum entropy characteristics of Chla to establish an algorithm, which is mainly used to improve the spatial coverage of remote sensing Chla. However, there has been no attempt to directly combine multi-band spectral signals with SST and longitude and latitude information to optimize the inversion of remote sensing Chla. Summary of the invention

[0005] The purpose of this application is to propose a global chlorophyll a remote sensing inversion optimization method and device covering high latitudes in response to the above-mentioned technical problems.

[0006] In a first aspect, the present invention provides a global chlorophyll a remote sensing inversion optimization method covering high latitudes, comprising the following steps:

[0007] A measured data set based on measured remote sensing reflectance and measured chlorophyll a concentration was constructed. The measured data set was temporally and spatially matched with the sea surface temperature to extract the sea surface temperature corresponding to each sample in the measured data set, and a training data set was constructed by integration. The training data set contained measured remote sensing reflectance, sea surface temperature, longitude, latitude and measured chlorophyll a concentration. The measured remote sensing reflectance included five central bands of 412, 443, 490, 555 and 670 nm.

[0008] The BP neural network is used to construct a global chlorophyll a remote sensing inversion model covering high latitudes. The initial weights and thresholds of the BP neural network are obtained through iterative optimization of the genetic algorithm. The global chlorophyll a remote sensing inversion model is trained using a training data set to obtain a trained global chlorophyll a remote sensing inversion model.

[0009] The satellite remote sensing reflectivity to be inverted and its matching sea surface temperature, longitude and latitude are input into the trained global chlorophyll a remote sensing inversion model to invert the chlorophyll a concentration.

[0010] Preferably, the measured remote sensing reflectivity is selected from the field observation data within the range of + / -6nm of the five central bands of 412, 443, 490, 555 and 670nm in the SeaWiFS sensor; the measured chlorophyll a concentration adopts the HPLC measurement result or the fluorescence measurement result.

[0011] Preferably, performing spatiotemporal matching of the sea surface temperature on the measured data set to extract the sea surface temperature corresponding to each sample in the measured data set specifically includes:

[0012] The sampling time, longitude and latitude of each sample in the measured data set are temporally and spatially matched in the 4km resolution Aqua-MODIS multi-year climatological monthly average sea surface temperature data to obtain the sea surface temperature corresponding to each sample in the measured data set.

[0013] Preferably, the BP neural network includes an input layer, a first hidden layer, a second hidden layer and an output layer. The number of nodes in the input layer is 8, and the 8 nodes in the input layer correspond to the remote sensing reflectance of 412, 443, 490, 555, and 670 nm and their temporal and spatial matching sea surface temperature, longitude and latitude. The number of nodes in the output layer is 1, corresponding to the chlorophyll a concentration; the number of nodes in the first hidden layer and the second hidden layer are both 4.

[0014] In a second aspect, the present invention provides a global chlorophyll a remote sensing inversion optimization device covering high latitudes, comprising:

[0015] The training data construction module is configured to construct a measured data set based on the measured remote sensing reflectance and the measured chlorophyll a concentration, perform spatiotemporal matching of the sea surface temperature on the measured data set to extract the sea surface temperature corresponding to each sample in the measured data set, and integrate and construct a training data set, the training data set includes the measured remote sensing reflectance, sea surface temperature, longitude, latitude and measured chlorophyll a concentration, and the measured remote sensing reflectance includes five band values ​​of 412, 443, 490, 555 and 670nm;

[0016] The model building module is configured to apply the BP neural network to build a global chlorophyll a remote sensing inversion model covering high latitudes. The initial weights and thresholds of the BP neural network are obtained by iterative optimization of the genetic algorithm. The global chlorophyll a remote sensing inversion model is trained using the training data set to obtain a trained global chlorophyll a remote sensing inversion model.

[0017] The inversion module is configured to input the satellite remote sensing reflectivity to be inverted and its matching sea surface temperature, longitude and latitude into a trained global chlorophyll a remote sensing inversion model to invert the chlorophyll a concentration.

[0018] In a third aspect, the present invention provides an electronic device comprising one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0019] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any implementation manner in the first aspect.

[0020] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method described in any implementation manner in the first aspect.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] (1) The global chlorophyll a remote sensing inversion optimization method covering high latitudes proposed in the present invention fully considers the interference of CDOM and suspended particles in the blue-green band on chlorophyll a inversion and the temporal and spatial differences in the optical characteristics of water bodies (absorption and backscattering of each component), especially the optical particularity of water bodies in high latitudes. The multi-band spectrum is combined with SST and longitude and latitude to invert chlorophyll a concentration, and the advantages of BP neural network in nonlinear fitting are comprehensively utilized to effectively improve the accuracy of global remote sensing inversion of chlorophyll a, among which the inversion effect in high latitudes is particularly significantly improved. The present invention can provide strong technical support for the study of the response of global marine ecosystems to climate change, including high latitudes.

[0023] (2) The global chlorophyll a remote sensing inversion optimization method covering high latitudes proposed in the present invention is based on an open field data set recognized in the international water color remote sensing field. The data are distributed in various oceans around the world and cover different types of water bodies. The global inversion adopts a unified algorithm model, avoiding the inversion uncertainty problem caused by algorithm models in different regions.

[0024] (3) The remote sensing reflectance in the training data set of the global chlorophyll a remote sensing inversion optimization method covering high latitudes proposed in the present invention is selected from the field observation data within the range of + / -6nm of the five central bands of 412, 443, 490, 555 and 670nm in the SeaWiFS sensor, that is, it is assumed that there is no significant difference in the remote sensing reflectance of the central band difference of + / -6nm. Therefore, for other water color satellite sensors, such as MODIS, MERIS, VIIRS and other sensors, if they have five central bands of 412, 443, 490, 555, 670nm+ / -6nm, the model is also applicable, which is conducive to the integration and product application of chlorophyll a concentration inversion products from different water color satellites. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 A schematic diagram of a flow chart of a global chlorophyll a remote sensing inversion optimization method covering high latitudes according to an embodiment of the present application;

[0027] Figure 2 A flowchart of a model construction process of a global chlorophyll a remote sensing inversion optimization method covering high latitudes according to an embodiment of the present application;

[0028] Figure 3 A schematic diagram of a BP neural network of a global chlorophyll a remote sensing inversion optimization method covering high latitudes according to an embodiment of the present application;

[0029] Figure 4 A diagram showing the chlorophyll a inversion accuracy evaluation results of the trained global chlorophyll a remote sensing inversion model and the NASA standard OCI algorithm of the global chlorophyll a remote sensing inversion optimization method covering high latitudes of an embodiment of the present application (gray: low and middle latitudes; black: high latitudes), wherein (a) the NASA standard OCI algorithm, (b) the trained global chlorophyll a remote sensing inversion model (GBP) of the present invention;

[0030] Figure 5 An inversion result diagram of a trained global chlorophyll a remote sensing inversion model covering high latitudes in the global chlorophyll a remote sensing inversion optimization method of an embodiment of the present application;

[0031] Figure 6 A schematic diagram of a global chlorophyll a remote sensing inversion optimization device covering high latitudes according to an embodiment of the present application;

[0032] Figure 7 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] Figure 1The embodiment of the present application provides a global chlorophyll a remote sensing inversion optimization method covering high latitudes, comprising the following steps:

[0035] S1, construct a measured data set based on the measured remote sensing reflectance and the measured chlorophyll a concentration, perform spatiotemporal matching of the sea surface temperature on the measured data set to extract the sea surface temperature corresponding to each sample in the measured data set, and integrate to construct a training data set. The training data set contains the measured remote sensing reflectance, sea surface temperature, longitude, latitude and measured chlorophyll a concentration. The measured remote sensing reflectance includes five central bands of 412, 443, 490, 555 and 670nm.

[0036] In a specific embodiment, the measured remote sensing reflectivity is selected from the field observation data within the range of + / -6nm of the five central bands of 412, 443, 490, 555 and 670nm in the SeaWiFS sensor; the measured chlorophyll a concentration adopts the HPLC measurement result or the fluorescence measurement result.

[0037] In a specific embodiment, performing spatiotemporal matching of sea surface temperature on a measured data set to extract the sea surface temperature corresponding to each sample in the measured data set specifically includes:

[0038] The sampling time, longitude and latitude of each sample in the measured data set are temporally and spatially matched in the 4km resolution Aqua-MODIS multi-year climatological monthly average sea surface temperature data to obtain the sea surface temperature corresponding to each sample in the measured data set.

[0039] Specifically, refer to Figure 2 , the embodiment of the present application establishes a measured data set that matches the synchronously sampled measured remote sensing reflectance (Rrs) and the measured chlorophyll a concentration based on an open field data set recognized in the international water color remote sensing field (same as NASA Chla standard algorithm). Among them, the measured remote sensing reflectance selects the field observation data within the range of + / -6nm of the five central bands of 412, 443, 490, 555, and 670nm in the SeaWiFS sensor; the measurement results of the measured chlorophyll a concentration can choose HPLC measurement results or fluorescence measurement results. When there are both HPLC measurement results and fluorescence measurement results, HPLC measurement results are preferred, and fluorescence measurement results are second. In addition to the measured remote sensing reflectance and the measured chlorophyll a concentration, the measured data set also includes the sampling time, longitude, and latitude of each sample.

[0040] The measured data set was matched with sea surface temperature (SST) in time and space. The Aqua-MODIS multi-year climatological monthly average sea surface temperature data (spatial resolution of 4km) was downloaded from the NASA ocean color website. According to the sampling time, longitude and latitude of each sample in the above measured data set, the sea surface temperature corresponding to each sample was matched by month. The training data set of the global chlorophyll a remote sensing inversion model was constructed based on the measured data set and the sea surface temperature obtained by time and space matching, which includes the measured remote sensing reflectance, sea surface temperature, longitude, latitude, and measured chlorophyll a concentration in five bands (412, 443, 490, 555 and 670nm).

[0041] Establish a test data set. The satellite remote sensing reflectance and measured chlorophyll a concentration in the test data set used in the embodiments of the present application directly adopt NASA SeaBASS for detecting the validation data set of SeaWiFS satellite water color products. According to the longitude and latitude information and sampling time in the validation data set, the sea surface temperature corresponding to each sample is matched in the same way as above. The matching data of integrated satellite remote sensing reflectance (412, 443, 490, 555, 670nm), sea surface temperature (SST), longitude, latitude and measured chlorophyll a concentration are used as the test data set of the global chlorophyll a remote sensing inversion model.

[0042] S2, using BP neural network, constructs a global chlorophyll a remote sensing inversion model covering high latitudes. The initial weights and thresholds of the BP neural network are obtained by iterative optimization of genetic algorithm. The global chlorophyll a remote sensing inversion model is trained using the training data set to obtain a trained global chlorophyll a remote sensing inversion model.

[0043] In a specific embodiment, the BP neural network includes an input layer, a first hidden layer, a second hidden layer and an output layer. The number of nodes in the input layer is 8, and the 8 nodes in the input layer correspond to the remote sensing reflectance of 412, 443, 490, 555, and 670 nm and their temporal and spatial matching sea surface temperature, longitude and latitude. The number of nodes in the output layer is 1, corresponding to the chlorophyll a concentration; the number of nodes in the first hidden layer and the second hidden layer are both 4.

[0044] Specifically, the global chlorophyll a remote sensing inversion model covering high latitudes mentioned in the embodiments of the present application adopts a BP neural network, and the initial weights and thresholds of the BP neural network are obtained by iterative optimization through genetic algorithm. In the training process of the global chlorophyll a remote sensing inversion model, the global chlorophyll a remote sensing inversion model is trained using a training data set to obtain a trained global chlorophyll a remote sensing inversion model. The performance of the trained global chlorophyll a remote sensing inversion model is evaluated using a test data set to evaluate the inversion accuracy of the global chlorophyll a remote sensing inversion model. The evaluation indicators are mainly MAPD (Mean absolute percentage difference), RMSE, and the linear regression determination coefficient R between the inverted chlorophyll a concentration and the measured chlorophyll a concentration. 2 The specific formulas of MAPD and RMSE are as follows:

[0045]

[0046] Among them, n represents the total number of samples, C_satellite represents the inverted chlorophyll a concentration, and C_in situ represents the measured chlorophyll a concentration.

[0047] The inversion accuracy evaluation results of the trained global chlorophyll a remote sensing inversion model and the inversion accuracy evaluation results of the NASA standard OCI algorithm are shown in Figure 4 By comparison, the trained global chlorophyll a remote sensing inversion model proposed in the embodiment of the present application has higher inversion accuracy, especially in high latitudes.

[0048] S3, inputting the satellite remote sensing reflectivity to be inverted and its matching sea surface temperature, longitude and latitude into the trained global chlorophyll a remote sensing inversion model to invert the chlorophyll a concentration.

[0049] Specifically, the trained global chlorophyll a remote sensing inversion model is deployed, the water color remote sensing file is read as needed, and the satellite remote sensing reflectance, longitude, latitude and sampling date of the corresponding bands (412, 443, 490, 555 and 670nm) are exported; the sea surface temperature of each pixel is matched according to the longitude, latitude and sampling date information by the time and space matching method. The satellite remote sensing reflectance, sea surface temperature, longitude and latitude of the corresponding bands (412, 443, 490, 555, 670nm) are input into the trained global chlorophyll a remote sensing inversion model. The inversion result output after the model is run is the chlorophyll a concentration, and it is mapped according to the longitude and latitude. Taking the inversion of the monthly average remote sensing chlorophyll a concentration of VIIRS_SNPP in April 2020 as an example, the inversion results obtained by the trained global chlorophyll a remote sensing inversion model are as follows Figure 5 shown.

[0050] Further references Figure 6 As an implementation of the methods shown in the above figures, the present application provides an embodiment of a global chlorophyll a remote sensing inversion optimization device covering high latitudes. Figure 1 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0051] The embodiment of the present application provides a global chlorophyll a remote sensing inversion optimization device covering high latitudes, comprising:

[0052] The training data construction module 1 is configured to construct a measured data set based on the measured remote sensing reflectance and the measured chlorophyll a concentration, perform spatiotemporal matching of the sea surface temperature on the measured data set to extract the sea surface temperature corresponding to each sample in the measured data set, and integrate and construct a training data set, wherein the training data set includes the measured remote sensing reflectance, sea surface temperature, longitude, latitude and measured chlorophyll a concentration, and the measured remote sensing reflectance includes five band values ​​of 412, 443, 490, 555 and 670 nm;

[0053] The model building module 2 is configured to apply a BP neural network to build a global chlorophyll a remote sensing inversion model covering high latitudes. The initial weights and thresholds of the BP neural network are obtained by iteratively solving and optimizing the genetic algorithm. The global chlorophyll a remote sensing inversion model is trained using a training data set to obtain a trained global chlorophyll a remote sensing inversion model.

[0054] The inversion module 3 is configured to input the satellite remote sensing reflectivity to be inverted and its matching sea surface temperature, longitude and latitude into the trained global chlorophyll a remote sensing inversion model to invert and obtain the chlorophyll a concentration.

[0055] Figure 7 Schematic diagram of the hardware structure of the electronic device provided by the embodiment of the present invention. Figure 7 As shown, the electronic device of this embodiment includes: a processor 701 and a memory 702; wherein the memory 702 is used to store computer-executable instructions; the processor 701 is used to execute the computer-executable instructions stored in the memory to implement the various steps performed by the electronic device in the above embodiment. For details, please refer to the relevant description in the above method embodiment.

[0056] Optionally, the memory 702 may be independent or integrated with the processor 701 .

[0057] When the memory 702 is independently provided, the electronic device further includes a bus 703 for connecting the memory 702 and the processor 701 .

[0058] The embodiment of the present invention further provides a computer storage medium, in which computer execution instructions are stored. When the processor 701 executes the computer execution instructions, the above method is implemented.

[0059] The embodiment of the present invention further provides a computer program product, including a computer program. When the computer program is executed by the processor 701, the above method is implemented.

[0060] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0061] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to implement the solution of this embodiment.

[0062] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each module may exist physically separately, or two or more modules may be integrated into one unit. The unit formed by the above modules may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0063] The above-mentioned integrated module implemented in the form of a software function module can be stored in a computer-readable storage medium. The above-mentioned software function module is stored in a storage medium, including a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor 701 to perform some steps of the methods of various embodiments of the present application.

[0064] It should be understood that the processor 701 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or the processor 701 may be any conventional processor 701, etc. The steps of the method disclosed in the invention may be directly embodied in the execution of the hardware processor 701, or may be executed by a combination of hardware and software modules in the processor 701.

[0065] The memory 702 may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk memory, and may also be a USB flash drive, a mobile hard disk, a read-only memory, a magnetic disk, or an optical disk.

[0066] The bus 703 may be an Industry Standard Architecture (ISA), a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus 703 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus 703 in the drawings of the present application is not limited to only one bus 703 or one type of bus 703.

[0067] The above storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The storage medium can be any available medium that can be accessed by a general or special purpose computer.

[0068] An exemplary storage medium is coupled to the processor 701, so that the processor 701 can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor 701. The processor 701 and the storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as ASIC). Of course, the processor 701 and the storage medium can also exist as discrete components in an electronic device or a main control device.

[0069] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0070] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A global chlorophyll a remote sensing inversion optimization method covering high latitudes, characterized in that: The following steps are involved: Constructing a measured data set based on measured remote sensing reflectance and measured chlorophyll a concentration, performing spatiotemporal matching of sea surface temperature on the measured data set to extract the sea surface temperature corresponding to each sample in the measured data set, and integrating to construct a training data set, wherein the training data set includes measured remote sensing reflectance, sea surface temperature, longitude, latitude and measured chlorophyll a concentration, wherein the measured remote sensing reflectance includes five central bands of 412, 443, 490, 555 and 670 nm; A BP neural network is applied to construct a global chlorophyll a remote sensing inversion model covering high latitudes, wherein the initial weights and thresholds of the BP neural network are obtained by iterative optimization of a genetic algorithm, and the global chlorophyll a remote sensing inversion model is trained using a training data set to obtain a trained global chlorophyll a remote sensing inversion model; The satellite remote sensing reflectivity to be inverted and its matching sea surface temperature, longitude and latitude are input into the trained global chlorophyll a remote sensing inversion model to invert and obtain the chlorophyll a concentration.

2. The global chlorophyll a remote sensing inversion optimization method covering high latitudes according to claim 1, characterized in that: The measured remote sensing reflectivity uses field observation data within the range of + / -6nm of five central bands of 412, 443, 490, 555 and 670nm in the SeaWiFS sensor; the measured chlorophyll a concentration uses HPLC measurement results or fluorescence measurement results.

3. The global chlorophyll a remote sensing inversion optimization method covering high latitudes according to claim 1, characterized in that: Performing spatiotemporal matching of the sea surface temperature on the measured data set to extract the sea surface temperature corresponding to each sample in the measured data set specifically includes: The sampling time, longitude and latitude of each sample in the measured data set are temporally and spatially matched in the 4km resolution Aqua-MODIS multi-year climatological monthly average sea surface temperature data to obtain the sea surface temperature corresponding to each sample in the measured data set.

4. The global chlorophyll a remote sensing inversion optimization method covering high latitudes according to claim 1, characterized in that: The BP neural network includes an input layer, a first hidden layer, a second hidden layer and an output layer. The number of nodes in the input layer is 8, and the 8 nodes in the input layer correspond to 412, 443, 490, 555, 670nm remote sensing reflectance and their spatiotemporal matching sea surface temperature, longitude and latitude respectively. The number of nodes in the output layer is 1, corresponding to chlorophyll a concentration; the number of nodes in the first hidden layer and the second hidden layer are both 4.

5. A global chlorophyll a remote sensing inversion optimization device covering high latitudes, characterized in that: include: A training data construction module is configured to construct a measured data set based on measured remote sensing reflectance and measured chlorophyll a concentration, perform spatiotemporal matching of sea surface temperature on the measured data set to extract the sea surface temperature corresponding to each sample in the measured data set, and integrate and construct a training data set, wherein the training data set includes measured remote sensing reflectance, sea surface temperature, longitude, latitude and measured chlorophyll a concentration, and the measured remote sensing reflectance includes five band values ​​of 412, 443, 490, 555 and 670 nm; The model building module is configured to apply a BP neural network to build a global chlorophyll a remote sensing inversion model covering high latitudes, wherein the initial weights and thresholds of the BP neural network are obtained by iteratively solving and optimizing the genetic algorithm, and the global chlorophyll a remote sensing inversion model is trained using a training data set to obtain a trained global chlorophyll a remote sensing inversion model; The inversion module is configured to input the satellite remote sensing reflectivity to be inverted and its matching sea surface temperature, longitude and latitude into the trained global chlorophyll a remote sensing inversion model to invert and obtain the chlorophyll a concentration.

6. An electronic device comprising: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images