CN112418506A - Coastal zone wetland ecological safety pattern optimization method and device based on machine learning - Google Patents
Coastal zone wetland ecological safety pattern optimization method and device based on machine learning Download PDFInfo
- Publication number
- CN112418506A CN112418506A CN202011294365.1A CN202011294365A CN112418506A CN 112418506 A CN112418506 A CN 112418506A CN 202011294365 A CN202011294365 A CN 202011294365A CN 112418506 A CN112418506 A CN 112418506A
- Authority
- CN
- China
- Prior art keywords
- wetland
- coastal zone
- ecological
- shape
- plaque
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000005457 optimization Methods 0.000 title claims abstract description 33
- 238000010801 machine learning Methods 0.000 title claims abstract description 30
- 238000000605 extraction Methods 0.000 claims abstract description 51
- 238000004458 analytical method Methods 0.000 claims abstract description 31
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 14
- 230000002285 radioactive effect Effects 0.000 claims description 18
- 230000003595 spectral effect Effects 0.000 claims description 12
- 238000012360 testing method Methods 0.000 claims description 12
- 230000015572 biosynthetic process Effects 0.000 claims description 9
- 230000005855 radiation Effects 0.000 claims description 9
- 238000003786 synthesis reaction Methods 0.000 claims description 9
- 238000000540 analysis of variance Methods 0.000 claims description 7
- 238000002310 reflectometry Methods 0.000 claims description 7
- 238000000692 Student's t-test Methods 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 6
- 238000012353 t test Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000013519 translation Methods 0.000 claims description 4
- 238000011049 filling Methods 0.000 claims description 3
- 238000011835 investigation Methods 0.000 claims description 3
- 238000012805 post-processing Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000002194 synthesizing effect Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000005562 fading Methods 0.000 claims description 2
- 230000000630 rising effect Effects 0.000 claims description 2
- 238000007637 random forest analysis Methods 0.000 abstract description 3
- 238000012706 support-vector machine Methods 0.000 abstract description 2
- 240000002044 Rhizophora apiculata Species 0.000 description 20
- 230000009545 invasion Effects 0.000 description 17
- 241001149258 Sporobolus alterniflorus Species 0.000 description 15
- 241000196324 Embryophyta Species 0.000 description 6
- 241000894007 species Species 0.000 description 6
- 238000013461 design Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 230000003993 interaction Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000009792 diffusion process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 241000120622 Rhizophoraceae Species 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 230000004083 survival effect Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 241000272517 Anseriformes Species 0.000 description 1
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- FAPWRFPIFSIZLT-UHFFFAOYSA-M Sodium chloride Chemical compound [Na+].[Cl-] FAPWRFPIFSIZLT-UHFFFAOYSA-M 0.000 description 1
- 238000004173 biogeochemical cycle Methods 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000001125 extrusion Methods 0.000 description 1
- 238000013467 fragmentation Methods 0.000 description 1
- 238000006062 fragmentation reaction Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000009364 mariculture Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009919 sequestration Effects 0.000 description 1
- 239000011780 sodium chloride Substances 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 239000002352 surface water Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Software Systems (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Multimedia (AREA)
- Remote Sensing (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Astronomy & Astrophysics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a coastal zone wetland ecological safety pattern optimization method and device based on machine learning, wherein the method comprises the following steps: carrying out pixel scale supervision and classification extraction on the coastal zone wetland of the target coast by adopting a support vector machine and a random forest method to obtain an extraction result of the coastal zone wetland; carrying out shape recognition on each wetland plaque in the coastal zone wetland extraction result by adopting a plaque shape recognition algorithm based on an information theory; and comparing the ecological disturbance degree differences of the wetland patches with different shapes by using variance analysis to obtain the optimal coastal zone wetland landscape type with statistical significance, thereby obtaining a landscape ecological safety pattern mode capable of maximizing ecological toughness. The method can assist in guiding the optimization of the landscape pattern of the coastal zone wetland and provide a landscape planning idea and a theoretical basis.
Description
Technical Field
The invention relates to the field of marine ecology, in particular to a coastal zone wetland ecological safety pattern optimization method and device based on machine learning.
Background
The coastal zone wetland has important ecological functions and values as a unique ecological system formed by the interaction of sea and land, and can provide various ecological system service values such as carbon sequestration, biological diversity protection, wind prevention, wave fixation, leisure recreation and the like. The safety of the land ecosystem and the marine ecosystem connected with the coastal zone wetland is directly influenced by the ecological safety condition of the coastal zone wetland, and the coastal zone wetland ecological safety control method is an important basis for realizing regional sustainable development. In recent decades, due to the influence of human social and economic activities, the population of cities is increased, the cities are rapidly expanded, wetland resources are damaged or degraded in quantity and quality to different degrees, the fragmentation of the internal habitat of the wetland is intensified, the landscape diversity is reduced, and the structure and the function of the wetland ecosystem are remarkably changed.
As the most intense region of human activities, large-scale sea reclamation activities are carried out in the coastal zone in the last decades, and the sea level is raised under the climate change, so that double extrusion is caused to the wetland habitat of the coastal zone, and the ecological safety of the wetland landscape of the coastal zone is seriously threatened. The landscape ecological safety pattern optimization is used as an important basic work for ecological protection and management utilization, can help managers to master the survival state and the health condition of an ecological system in time, and can provide data support for the specification of ecological system protection policies. Therefore, the ecological safety pattern optimization of the coastal zone wetland landscape in the coastal zone area is an urgent need for ecological environment protection.
In the face of wide-range ecological environment degradation, landscape ecological safety pattern design must be carried out on a large scale. The existing landscape ecological safety pattern optimization method can perform ecological optimization according to local conditions on a small scale based on a planning design visual angle, but is difficult to perform efficient and automatic planning on a large scale. In addition, limited by the scale and the planning level, the existing method is difficult to give full play to the resistance and resilience of the ecosystem to external interference, and has defects even if applied to large-scale work.
In summary, the existing coastal zone wetland landscape ecological safety pattern optimization method mainly has the following defects:
1. the existing coastal zone wetland extraction method has high labor and time cost and is difficult to carry out large-scale extraction work. The existing coastal zone wetland extraction algorithm usually eliminates land surface ground objects in advance through artificially outlined coastline boundaries so as to reduce extraction errors. However, it is difficult to perform large-scale coastal zone wetland extraction work using this method because manual extraction of the coastline is time-consuming and labor-consuming, and the extraction accuracy is difficult to unify due to subjective factors.
2. The landscape ecological safety pattern optimization method is difficult to avoid artificial subjective factors in the process and cannot perform full-automatic optimization. For example, existing methods often utilize a least cumulative resistance model that requires artificial setting of resistance surface weights when performing landscape corridor identification; in addition, the existing method usually uses buffer area analysis in the defined research area or the species diffusion area, the setting of the buffer radius usually depends on the subjective judgment of the qualitative of workers, and the quantitative basis is lacked.
3. Ecological toughness is not considered. In the prior art, when landscape pattern optimization is carried out, horizontal migration and diffusion of species are mostly considered, an ecological toughness concept is less introduced to enter landscape pattern optimization, and the resistance and resilience of an ecological system to external disturbance cannot be fully known and exerted.
Disclosure of Invention
In view of the above, the invention aims to provide a coastal zone wetland ecological safety pattern optimization method and device based on machine learning, which can assist in guiding the landscape pattern optimization of the coastal zone wetland and provide a landscape planning idea and a theoretical basis.
The embodiment of the invention provides a coastal zone wetland ecological safety pattern optimization method based on machine learning, which comprises the following steps:
performing pixel-scale supervision and classification extraction on the coastal zone wetland of the target coast by utilizing the coastal zone tide level information to obtain a coastal zone wetland extraction result;
carrying out shape recognition on each wetland plaque in the coastal zone wetland extraction result by adopting a plaque shape recognition algorithm based on an information theory;
and comparing the ecological disturbance degree differences of the wetland patches with different shapes by using variance analysis to obtain the optimal coastal zone wetland landscape type with statistical significance, thereby obtaining a landscape ecological safety pattern mode capable of maximizing ecological toughness.
Preferably, the step of performing pixel-scale supervised classification extraction on the coastal zone wetland of the target coast by using the coastal zone tide level information to obtain the coastal zone wetland extraction result specifically comprises the following steps:
carrying out radiometric calibration, geometric correction and atmospheric correction on the acquired original multispectral remote sensing image of the target coast to obtain surface reflectivity data;
obtaining a buffer area of a preset distance of a coastline of a target coastline to obtain a coastline range;
carrying out cloud removal processing on the earth surface reflectivity data to generate corresponding SR images, calculating NDVI and NDWI spectral indexes of each SR image, and then respectively obtaining a tide fading image, a tide midebbing image and a tide rising image by utilizing maximum NDVI synthesis, median synthesis and maximum NDWI synthesis;
superposing the ebb tide image, the middle tide image and the flood tide image to generate a superposed image; the number of spectral bands contained in the superposed image is the sum of the number of spectral bands of the three images;
inputting all spectral bands of a training sample point and a verification sample point obtained by utilizing a high-definition Google Earth image and field investigation and the superposed image into a machine learning classifier trained in advance to obtain a coastal zone wetland classification result;
repeatedly classifying the data of each year before and after the target year, and finally taking the result obtained by synthesizing the mode of the three-year classification results as the coastal zone wetland extraction result; wherein the coastal zone wetland extraction result comprises a plurality of wetland patches.
Preferably, the method further comprises the following steps:
carrying out post-treatment on the coastal zone wetland extraction result; the post-processing comprises eliminating fragmentary patches, filling holes and deleting obvious classification errors so as to ensure that the overall precision reaches more than 85 percent.
Preferably, the method for identifying the shape of each wetland plaque in the coastal zone wetland extraction result by using a plaque shape identification algorithm based on an information theory specifically comprises the following steps:
extracting the geometric center of each wetland patch, and performing equiangular radiation to the periphery by taking the geometric center as a diffusion center; wherein the number of the radioactive rays is N;
measuring the length of the radioactive rays to the boundary of each wetland plaque, taking the radioactive rays in the due north direction as a first ray, and sequencing N radioactive rays in turn in the clockwise direction;
plotting the length of the N sequenced radioactive rays, and regarding the length as a group of signals;
and judging the shape of each wetland plaque by comparing the similarity of the signal with the signal of a standard shape predefined by a user. Wherein the signal similarity is measured by Jensen-Shannon divergence; the calculation formula is as follows:
wherein D represents relative entropy; p and q are two sets of discrete signal values; sMIs a coastal zone wetland plaque radioactive ray distance signal; sSIs a predefined standard shape of the radiation distance signal.
Preferably, the method further comprises the following steps:
translating the radioactive ray distance signal of the wetland plaque to the left for M times, calculating Jensen-Shannon divergence degrees of each standard shape according to the translated result, and recording the minimum value and the corresponding standard plaque;
and after the M times of translation are finished, selecting the minimum value in the results, and taking the corresponding standard shape as the shape judgment result of the wetland patch of the coastal zone.
Preferably, the variance analysis is used for comparing the ecological disturbance degree difference of the wetland patches with different shapes to obtain the optimal coastal zone wetland landscape type with statistical significance, and the method specifically comprises the following steps:
selecting indexes for quantifying ecological disturbance;
determining whether the ecological disturbance degrees of the wetland patches with the regular shape and the complex shape have significant difference by using a t test;
judging the difference of ecological disturbance degree of the wetland plaque communities in each shape by using a variance analysis method;
and after the analysis of variance is finished, selecting a statistical result with significance from the analysis results, and acquiring the shape type with the lowest ecological disturbance degree as the patch type with the highest ecological toughness, thereby obtaining a landscape ecological safety pattern mode capable of maximizing the ecological toughness.
Preferably, the method for judging the difference of the ecological disturbance degree of the wetland plaque communities in each shape by using the variance analysis method specifically comprises the following steps:
carrying out data normal distribution test and variance homogeneity test, and adopting a student t test method for the habitat patch data meeting the normal distribution and the variance homogeneity; if the data fails to satisfy both conditions, then a non-parametric t-test is used.
Preferably, the method for judging the difference of the ecological disturbance degree of the wetland plaque communities in each shape by using the variance analysis method specifically comprises the following steps:
carrying out normal distribution and homogeneity of variance test on the data, and adopting an ANOVA variance analysis method on the wetland patches meeting the test; and adopting a non-parameter variance analysis method for the unsatisfied data.
The embodiment of the invention also provides a coastal zone wetland ecological safety pattern optimization device based on machine learning, which comprises the following steps:
the extraction unit is used for performing pixel-scale supervision and classification extraction on the coastal zone wetland of the target coast by utilizing the coastal zone tide level information to obtain a coastal zone wetland extraction result;
the identification unit is used for carrying out shape identification on each wetland plaque in the coastal zone wetland extraction result by adopting a plaque shape identification algorithm based on an information theory;
and the disturbance analysis unit is used for comparing the ecological disturbance degree difference of the wetland patches with different shapes by using variance analysis to obtain the optimal coastal zone wetland landscape type with statistical significance, so that a landscape ecological safety pattern mode capable of maximizing ecological toughness is obtained.
The coastal zone wetland ecological safety pattern optimization method based on machine learning provided by the embodiment of the invention has the following advantages:
1. the tidal level information is fully utilized to automatically extract the coastal zone wetland. By using the tidal level information of the coastal zone wetland as the distinguishing characteristic of the coastal zone wetland and the land vegetation, the invention innovatively introduces the tidal level into the machine learning classifier as the classifying characteristic, and can quickly and accurately extract the vegetation distribution data of the coastal zone wetland. Because the classification noise of the land vegetation can be better filtered by utilizing the tide level information, the method does not need to remove the land part based on manually drawn coastline data in advance, and has the potential of carrying out the automatic extraction work of the coastal zone wetland on a large scale.
2. Based on the landscape shape recognition algorithm of unsupervised learning in machine learning, the landscape ecological safety pattern can be automatically recognized. After the user inputs the standard shape in advance, the shapes of all target coastal zone wetland landscapes can be automatically identified according to the shape feature similarity, and the landscape shape with the best interference resistance is identified by comparing the interference degree difference among different shapes, so that the space planning and the landscape design of the coastal zone are guided. Based on the algorithm, the landscape pattern type capable of maximizing ecological toughness can be obtained quickly and automatically without human intervention.
3. The method introduces a toughness thinking into landscape ecological safety pattern optimization of the coastal zone wetland, fully considers interaction between species and between landscape units (such as patches), and takes the improvement of the resistance of the coastal zone wetland to external interference as a target of a toughness method.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a schematic flow chart of a method for optimizing the ecological safety pattern of a coastal zone wetland based on machine learning according to a first embodiment of the present invention;
fig. 2 is another flow chart of the method for optimizing the ecological safety pattern of the coastal zone wetland based on machine learning according to the first embodiment of the present invention;
FIG. 3 is a flow chart of the coastal zone wetland supervision and classification;
FIG. 4 is a flow chart of shape interpretation of coastal zone wetlands;
FIG. 5 is an example of the result of the coastal zone wetland extraction;
fig. 6 is a coastal zone wetland shape recognition example;
FIG. 7 is an example of an Spartina alterniflora intrusion;
FIG. 8 is a mangrove landscape ecological security pattern identification;
fig. 9 is a schematic structural diagram of a coastal zone wetland ecological safety pattern optimization device based on machine learning according to a second embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings of the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
Referring to fig. 1 and 2, a first embodiment of the present invention provides a method for optimizing an ecological safety pattern of a coastal zone wetland based on machine learning, which at least includes the following steps:
s101, performing pixel-scale supervision and classification extraction on the coastal zone wetland of the target coast by utilizing the coastal zone tide level information to obtain a coastal zone wetland extraction result.
Specifically, as shown in fig. 3:
firstly, radiometric calibration, geometric correction and atmospheric correction are carried out on the collected original multispectral remote sensing image of the target coast to obtain earth surface reflectivity data.
Then, a buffer area of a predetermined distance from the coastline of the target coast is acquired, and the coastline range is obtained.
Generating a coastline 10km buffer area by using openstreet map global coastline data which can be obtained in a public manner, preliminarily obtaining a coastline range, and then removing permanent water body areas (oceans, lakes and rivers) according to global surface water covering products to generate the coastline range which is used as a boundary of subsequent coastline wetland extraction work;
and then, carrying out cloud removal processing on the surface reflectivity data to generate corresponding SR images, calculating NDVI and NDWI spectral indexes of each SR image, and then respectively obtaining a ebb tide image, a midtide image and a flood tide image by utilizing maximum NDVI synthesis, median synthesis and maximum NDWI synthesis.
Secondly, overlapping the ebb tide image, the middle tide image and the flood tide image to generate an overlapped image; the number of spectral bands contained in the superposed image is the sum of the number of spectral bands of the three images;
further, inputting all spectral bands of training sample points and verification sample points obtained by utilizing a high-definition Google Earth image and field investigation and the superposed image into a machine learning classifier trained in advance to obtain a coastal zone wetland classification result;
finally, repeatedly classifying the data of each year before and after the target year, and finally taking the result obtained by synthesizing the mode of the three-year classification results as the coastal zone wetland extraction result; wherein the coastal zone wetland extraction result comprises a plurality of wetland patches.
Wherein, after a preliminary coastal zone wetland extraction result is obtained, the coastal zone wetland extraction result can be subjected to post-treatment; the post-processing comprises eliminating fragmentary patches, filling holes and deleting obvious classification errors so as to ensure that the overall precision reaches more than 85 percent.
And S102, performing shape recognition on each wetland plaque in the coastal zone wetland extraction result by adopting a plaque shape recognition algorithm based on an information theory.
In this embodiment, wetland plaque shape is a very important aspect of landscape ecological safety landscape architecture. The shape of the plaque boundaries can affect the ecological flow from matrix to plaque or plaque to plaque. The more complex the shape of the plaque, the more interactions (whether useful or harmful) with the surrounding substrate, for which it is necessary to identify the shape of each wetland plaque.
Specifically, as shown in fig. 4:
firstly, extracting the geometric center of each wetland patch, and carrying out equiangular radioactive rays towards the periphery by taking the geometric center as a diffusion center; wherein the number of the radioactive rays is N.
For example, the value of N may be 32, and of course, the value of N may be set according to actual needs, and these schemes are all within the protection scope of the present invention.
Then, measuring the length of the radiation to the boundary of each wetland plaque, taking the radiation in the due north direction as a first line, and sequencing N radiation in turn in the clockwise direction;
then, the length of the N sequenced radioactive rays is plotted and is regarded as a group of signals;
finally, as shown in fig. 5, the shape of each wetland plaque is determined by comparing the signal similarity with the standard shape predefined by the user. Wherein the signal similarity is measured by Jensen-Shannon divergence; the calculation formula is as follows:
wherein D represents relative entropy; p and q are two sets of discrete signal values; sMIs a coastal zone wetland plaque radioactive ray distance signal; sSIs a predefined standard shape of the radiation distance signal.
In the present embodiment, a case where the similarity is low may occur even with similar signals due to the influence of the signal phase. Therefore, the radiation distance signal of the coastal wetland patches is translated for M times to the left, the Jensen-Shannon divergence degree of each standard shape is calculated for the result after each translation, and the minimum value and the corresponding standard patch are recorded. And after the M times of translation are finished, selecting the minimum value in the results, and taking the corresponding standard patch as the shape judgment result of the coastal zone wetland patch.
Wherein, M can also be 32, and of course, the value of M can also be set according to the actual requirement, and these schemes are all within the protection scope of the present invention.
S103, comparing the ecological disturbance degree differences of the wetland patches with different shapes by using variance analysis, and obtaining the optimal coastal zone wetland landscape type with statistical significance, thereby obtaining a landscape ecological safety pattern mode capable of maximizing ecological toughness.
In this embodiment, identifying the optimal landscape shape requires comparing the ecological disturbance degrees in different shapes, and finding out the type with the minimum disturbance degree. Because the shape complexity is one of the factors to be considered in landscape design, the standard shape is divided into two categories of a regular shape and a complex shape, the ecological disturbance conditions of the regular shape and the complex shape are directly compared, and then the specific shape is identified. By using the variance analysis method, the difference of the interference degree of each shape coastal zone wetland community can be compared, and the patch shape with statistical significance and obviously less interference degree is identified as the optimal patch shape.
Specifically, the method comprises the following steps:
firstly, an index for quantifying ecological disturbance is selected.
Wherein, the ecological disturbance (such as wild fire, drought, hurricane, biological invasion, etc.) is quantified by using a certain index, such as hurricane disturbance intensity is quantified by using hurricane grade, biological invasion disturbance intensity is quantified by using invasion biological range, etc.
Then, whether the ecological disturbance degree of the wetland plaque with the regular shape and the wetland plaque with the complex shape has a significant difference is determined by using a t test.
Firstly, performing data normal distribution test and variance homogeneity test, and adopting a student t test method for habitat patch data meeting the normal distribution and variance homogeneity; if the data fails to satisfy both conditions, then a non-parametric t-test is used
And then, judging the difference of the ecological disturbance degree of the wetland plaque communities in all shapes by using a variance analysis method.
Wherein, normal distribution and homogeneity of variance test are carried out on the data, and an ANOVA variance analysis method is adopted for the wetland patches meeting the test; and adopting a non-parameter variance analysis method for the unsatisfied data.
And finally, after the analysis of variance is finished, selecting a statistical result with significance (p is less than 0.05) in the analysis result, and acquiring the shape type with the lowest ecological disturbance degree as the plaque type with the highest ecological toughness, thereby obtaining the landscape ecological safety pattern mode capable of maximizing the ecological toughness.
In summary, the method provided by the embodiment of the present invention has the following advantages:
1. the tidal level information is fully utilized to automatically extract the coastal zone wetland. By using the tidal level information of the coastal zone wetland as the distinguishing characteristic of the coastal zone wetland and the land vegetation, the invention innovatively introduces the tidal level into the machine learning classifier as the classifying characteristic, and can quickly and accurately extract the vegetation distribution data of the coastal zone wetland. Because the classification noise of the land vegetation can be better filtered by utilizing the tide level information, the method does not need to remove the land part based on manually drawn coastline data in advance, and has the potential of carrying out the automatic extraction work of the coastal zone wetland on a large scale.
2. Based on the landscape shape recognition algorithm of unsupervised learning in machine learning, the landscape ecological safety pattern can be automatically recognized. After the user inputs the standard shape in advance, the shapes of all target coastal zone wetland landscapes can be automatically identified according to the shape feature similarity, and the landscape shape with the best interference resistance is identified by comparing the interference degree difference among different shapes, so that the space planning and the landscape design of the coastal zone are guided. Based on the algorithm, the landscape pattern type capable of maximizing ecological toughness can be obtained quickly and automatically without human intervention.
3. The method introduces a toughness thinking into landscape ecological safety pattern optimization of the coastal zone wetland, fully considers interaction between species and between landscape units (such as patches), and takes the improvement of the resistance of the coastal zone wetland to external interference as a target of a toughness method.
In order to facilitate an understanding of the invention, the following description will be given by way of a practical example.
Biological invasion is a typical interference expression form faced by coastal zone wetlands, and causes the reduction of global biodiversity. Plant invasion, an important type of biological invasion, has become a serious ecological problem, threatening local species, affecting the structure and function of the ecosystem. In coastal zones, foreign invasive plants have profound effects on biogeochemical cycles, with serious consequences on the environment. Therefore, special attention must be paid to invasive plants in coastal zone areas to ensure ecological safety and maintain ecosystem sustainability. Spartina alterniflora has been classified as one of the most serious invasive plants by the national environmental protection Bureau. This invasion of foreign species has had a tremendous negative impact, for example, threatening the native wetland plants and waterfowls, and has had a negative impact on the development of fisheries, water transport, mariculture activities, and tourism. Spartina alterniflora was first introduced in 1979 from the coast of the American Atlantic ocean into China for the purpose of tidal land reclamation, stabilization of the coast and improvement of saline and alkaline soil. However, after the application of the spartina alterniflora, the spartina alterniflora rapidly expands and occupies a large area of coastal zone beach area from Hebei to Guangxi, and the survival and propagation of mangrove forest in China are severely restricted. Therefore, the invention analyzes the application case of the new method by taking plant invasion as an interference type and taking the invasion of spartina alterniflora to mangrove forest as a practical case.
Step 1: the tide level information is utilized to extract the mangrove forest and spartina alterniflora distribution data in 2020 of China. According to the method, Sentinel-2 multispectral Earth surface reflectivity data stored in a database are called on a Google Earth Engine cloud computing platform, and low-tide level images, normal-tide level images and high-tide level images are synthesized by using the maximum NDVI, the median and the maximum NDWI respectively. And superposing the data into a data set, and inputting the data set into a random forest classifier to obtain high-precision mangrove forest and spartina alterniflora distribution data. The overall accuracy of the data was verified to be about 88%, an example of which is shown in fig. 6.
Step 2: by utilizing the habitat patch shape identification method provided by the invention, each invaded mangrove patch is divided into standard shapes which are defined by a user in advance, wherein the standard shapes are rectangular, triangular, trapezoidal, L-shaped, U-shaped and Z-shaped. The trapezoid and the L-shaped are not symmetrical figures, so that the trapezoid and the L-shaped are split into a left subclass and a right subclass during shape recognition, and the subclasses are unified into respective categories in the following steps. Rectangles, triangles and trapezoids are defined as regular figures since they are basic figures in geometry, have stable area calculation formulas, and the remaining figures are defined as irregular figures, for example, as shown in fig. 7.
And step 3: the invasion degree of the spartina alterniflora to the mangrove forest is quantified and taken as a representation of ecological disturbance. As the spartina alterniflora and the mangrove grow on intertidal zones, the occupation of the spartina alterniflora on the intertidal zones around the mangrove can be regarded as the invasion of the mangrove, because the reproduction and the diffusion of the mangrove are interfered, and the living environment is limited. The invasion degree of the spartina alterniflora is described by the ratio of the area of the spartina alterniflora around the mangrove forest to the tidal flat area of the intertidal zone around the mangrove forest, and the invasion interference conditions under different patch shapes are compared according to the invasion degree, as shown in fig. 8.
And 4, step 4: area grouping ANOVA variance analysis is carried out on the mangrove forest patches to find that the invasion degree of the complex shape and the large class is obviously less than that of the regular shape for the small and medium size mangrove forest patches; for large mangrove plaques, the invasion degree of the spartina alterniflora in the complex shape and the regular shape has no obvious difference. With respect to the specific shape, for small mangrove forest patches, mangroves with L-shaped boundaries are significantly less intrusive than mangroves with rectangular boundary shapes; the U-shape is identified for medium-sized mangrove plaques as the least invasive mangrove plaque shape.
In the above embodiment, it can be recognized that the resistance of the mangrove forest to the spartina alterniflora intrusion interference is significantly different under different boundary shapes, and the L shape and the U shape in the complex shape can be considered as the habitat patch type with the highest resistance. The result can assist in guiding the prevention, control and treatment of the biological invasion of the coastal zone, and provide a landscape planning idea and a theoretical basis.
Referring to fig. 9, a second embodiment of the present invention further provides a coastal zone wetland ecological safety pattern optimization device based on machine learning, including:
the extraction unit 210 is configured to perform pixel-scale supervised classification extraction on the coastal zone wetland of the target coast by using a support vector machine and a random forest method to obtain a coastal zone wetland extraction result;
the identification unit 220 is used for performing shape identification on each wetland plaque in the coastal zone wetland extraction result by adopting a plaque shape identification algorithm based on an information theory;
and the disturbance analysis unit 230 is used for comparing the ecological disturbance degree differences of the wetland patches with different shapes by using variance analysis to obtain the optimal coastal zone wetland landscape type with statistical significance, so as to obtain a landscape ecological safety pattern mode capable of maximizing ecological toughness.
It should be noted that the above-described device embodiments are merely illustrative, where the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. In addition, in the drawings of the embodiment of the apparatus provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, and may be specifically implemented as one or more communication buses or signal lines. One of ordinary skill in the art can understand and implement it without inventive effort.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (9)
1. A coastal zone wetland ecological safety pattern optimization method based on machine learning is characterized by comprising the following steps:
performing pixel-scale supervision and classification extraction on the coastal zone wetland of the target coast by utilizing the coastal zone tide level information to obtain a coastal zone wetland extraction result;
carrying out shape recognition on each wetland plaque in the coastal zone wetland extraction result by adopting a plaque shape recognition algorithm based on an information theory;
and comparing the ecological disturbance degree differences of the wetland patches with different shapes by using variance analysis to obtain the optimal coastal zone wetland landscape type with statistical significance, thereby obtaining a landscape ecological safety pattern mode capable of maximizing ecological toughness.
2. The coastal zone wetland ecological safety pattern optimization method based on machine learning according to claim 1, characterized in that the coastal zone wetland of the target coast is subjected to pixel-scale supervised classification extraction by utilizing coastal zone tide level information to obtain a coastal zone wetland extraction result, and specifically comprises:
carrying out radiometric calibration, geometric correction and atmospheric correction on the acquired original multispectral remote sensing image of the target coast to obtain surface reflectivity data;
obtaining a buffer area of a preset distance of a coastline of a target coastline to obtain a coastline range;
carrying out cloud removal processing on the earth surface reflectivity data to generate corresponding SR images, calculating NDVI and NDWI spectral indexes of each SR image, and then respectively obtaining a tide fading image, a tide midebbing image and a tide rising image by utilizing maximum NDVI synthesis, median synthesis and maximum NDWI synthesis;
superposing the ebb tide image, the middle tide image and the flood tide image to generate a superposed image; the number of spectral bands contained in the superposed image is the sum of the number of spectral bands of the three images;
inputting all spectral bands of a training sample point and a verification sample point obtained by utilizing a high-definition Google Earth image and field investigation and the superposed image into a machine learning classifier trained in advance to obtain a coastal zone wetland classification result;
repeatedly classifying the data of each year before and after the target year, and finally taking the result obtained by synthesizing the mode of the three-year classification results as the coastal zone wetland extraction result; wherein the coastal zone wetland extraction result comprises a plurality of wetland patches.
3. The coastal zone wetland ecological safety pattern optimization method based on machine learning according to claim 2, characterized in that the coastal zone wetland ecological safety pattern optimization method further comprises:
carrying out post-treatment on the coastal zone wetland extraction result; the post-processing comprises eliminating fragmentary patches, filling holes and deleting obvious classification errors so as to ensure that the overall precision reaches more than 85 percent.
4. The coastal zone wetland ecological safety pattern optimization method based on machine learning according to claim 1, characterized in that the patch shape recognition algorithm based on information theory is adopted to perform shape recognition on each wetland patch in the coastal zone wetland extraction result, and specifically comprises:
extracting the geometric center of each wetland patch, and performing equiangular radiation to the periphery by taking the geometric center as a diffusion center; wherein the number of the radioactive rays is N;
measuring the length of the radioactive rays to the boundary of each wetland plaque, taking the radioactive rays in the due north direction as a first ray, and sequencing N radioactive rays in turn in the clockwise direction;
plotting the length of the N sequenced radioactive rays, and regarding the length as a group of signals;
judging the shape of each wetland plaque by comparing the similarity of the signal with a standard shape predefined by a user; wherein the signal similarity is measured by Jensen-Shannon divergence; the calculation formula is as follows:
wherein D represents relative entropy; p and q are two sets of discrete signal values; sMIs a coastal zone wetland plaque radioactive ray distance signal; ssIs a predefined standard shape of the radiation distance signal.
5. The coastal zone wetland ecological safety pattern optimization method based on machine learning according to claim 4, characterized in that the coastal zone wetland ecological safety pattern optimization method based on machine learning further comprises:
translating the radioactive ray distance signal of the wetland plaque to the left for M times, calculating Jensen-Shannon divergence degrees of each standard shape according to the translated result, and recording the minimum value and the corresponding standard plaque;
and after the M times of translation are finished, selecting the minimum value in the results, and taking the corresponding standard shape as the shape judgment result of the wetland patch of the coastal zone.
6. The coastal zone wetland ecological safety pattern optimization method based on machine learning according to claim 1, characterized in that variance analysis is used to compare the ecological disturbance degree differences of wetland patches of different shapes to obtain the optimal coastal zone wetland landscape type with statistical significance, and specifically comprises:
selecting indexes for quantifying ecological disturbance;
determining whether the ecological disturbance degrees of the wetland patches with the regular shape and the complex shape have significant difference by using a t test;
judging the difference of ecological disturbance degree of the wetland plaque communities in each shape by using a variance analysis method;
and after the analysis of variance is finished, selecting a statistical result with significance from the analysis results, and acquiring the shape type with the lowest ecological disturbance degree as the patch type with the highest ecological toughness, thereby obtaining a landscape ecological safety pattern mode capable of maximizing the ecological toughness.
7. The coastal zone wetland ecological safety pattern optimization method based on machine learning according to claim 6, characterized in that the determination of the ecological disturbance degree difference of wetland patch communities in each shape by using the variance analysis method specifically comprises:
carrying out data normal distribution test and variance homogeneity test, and adopting a student t test method for the habitat patch data meeting the normal distribution and the variance homogeneity; if the data fails to satisfy both conditions, then a non-parametric t-test is used.
8. The coastal zone wetland ecological safety pattern optimization method based on machine learning according to claim 6, characterized in that the determination of the ecological disturbance degree difference of wetland patch communities in each shape by using the variance analysis method specifically comprises:
carrying out normal distribution and homogeneity of variance test on the data, and adopting an ANOVA variance analysis method on the wetland patches meeting the test; and adopting a non-parameter variance analysis method for the unsatisfied data.
9. The utility model provides a coastal zone wetland ecological safety pattern optimizing apparatus based on machine learning which characterized in that includes:
the extraction unit is used for performing pixel-scale supervision and classification extraction on the coastal zone wetland of the target coast by utilizing the coastal zone tide level information to obtain a coastal zone wetland extraction result;
the identification unit is used for carrying out shape identification on each wetland plaque in the coastal zone wetland extraction result by adopting a plaque shape identification algorithm based on an information theory;
and the disturbance analysis unit is used for comparing the ecological disturbance degree difference of the wetland patches with different shapes by using variance analysis to obtain the optimal coastal zone wetland landscape type with statistical significance, so that a landscape ecological safety pattern mode capable of maximizing ecological toughness is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011294365.1A CN112418506B (en) | 2020-11-18 | 2020-11-18 | Coastal zone wetland ecological safety pattern optimization method and device based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011294365.1A CN112418506B (en) | 2020-11-18 | 2020-11-18 | Coastal zone wetland ecological safety pattern optimization method and device based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112418506A true CN112418506A (en) | 2021-02-26 |
| CN112418506B CN112418506B (en) | 2022-07-26 |
Family
ID=74774006
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011294365.1A Active CN112418506B (en) | 2020-11-18 | 2020-11-18 | Coastal zone wetland ecological safety pattern optimization method and device based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112418506B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113111672A (en) * | 2021-04-13 | 2021-07-13 | 中国科学院东北地理与农业生态研究所 | Method for judging true wetland plants |
| CN113807409A (en) * | 2021-08-27 | 2021-12-17 | 华北电力大学 | A coastal zone classification method based on discriminant analysis |
| CN113971769A (en) * | 2021-12-09 | 2022-01-25 | 中国科学院地理科学与资源研究所 | Long-term time-series identification method of coastal zone regional functions based on multi-source big data |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110103656A1 (en) * | 2009-04-17 | 2011-05-05 | Gheorghe Iordanescu | Quantification of Plaques in Neuroimages |
| CN105740794A (en) * | 2016-01-27 | 2016-07-06 | 中国人民解放军92859部队 | Satellite image based coastline automatic extraction and classification method |
| CN107229919A (en) * | 2017-06-05 | 2017-10-03 | 深圳先进技术研究院 | It is a kind of to be used for the ecological key element processing method and system of complicated ecological littoral zone |
| CN108509702A (en) * | 2018-03-21 | 2018-09-07 | 武汉理工大学 | Soil erosion optimal spatial scale selection model and its computational methods |
-
2020
- 2020-11-18 CN CN202011294365.1A patent/CN112418506B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110103656A1 (en) * | 2009-04-17 | 2011-05-05 | Gheorghe Iordanescu | Quantification of Plaques in Neuroimages |
| CN105740794A (en) * | 2016-01-27 | 2016-07-06 | 中国人民解放军92859部队 | Satellite image based coastline automatic extraction and classification method |
| CN107229919A (en) * | 2017-06-05 | 2017-10-03 | 深圳先进技术研究院 | It is a kind of to be used for the ecological key element processing method and system of complicated ecological littoral zone |
| CN108509702A (en) * | 2018-03-21 | 2018-09-07 | 武汉理工大学 | Soil erosion optimal spatial scale selection model and its computational methods |
Non-Patent Citations (5)
| Title |
|---|
| ZHEN ZHANG ET AL: "Spatially discontinuous relationships between salt marsh invasion and mangrove forest fragmentation", 《FOREST ECOLOGY AND MANAGEMENT》 * |
| 池源 等: "庙岛群岛北五岛景观格局特征及其生态效应", 《生态学报》 * |
| 白立敏: "基于景观格局视角的长春市城市生态韧性评价与优化研究", 《中国博士学位论文全文数据库》 * |
| 罗伟芳: "基于内容的图像检索技术研究", 《中国优秀硕士学位论文全文数据库》 * |
| 金卫斌 等: "中尺度流域的景观格局分析――以湖北四湖流域为例", 《长江流域资源与环境》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113111672A (en) * | 2021-04-13 | 2021-07-13 | 中国科学院东北地理与农业生态研究所 | Method for judging true wetland plants |
| CN113807409A (en) * | 2021-08-27 | 2021-12-17 | 华北电力大学 | A coastal zone classification method based on discriminant analysis |
| CN113807409B (en) * | 2021-08-27 | 2024-01-23 | 华北电力大学 | Coastal zone classification method based on discriminant analysis |
| CN113971769A (en) * | 2021-12-09 | 2022-01-25 | 中国科学院地理科学与资源研究所 | Long-term time-series identification method of coastal zone regional functions based on multi-source big data |
| CN113971769B (en) * | 2021-12-09 | 2022-06-14 | 中国科学院地理科学与资源研究所 | Long-term time-series identification method of coastal zone regional functions based on multi-source big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112418506B (en) | 2022-07-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Global offshore wind turbine dataset | |
| Wang et al. | Graph-based block-level urban change detection using Sentinel-2 time series | |
| Li et al. | A 30-year (1984–2013) record of annual urban dynamics of Beijing City derived from Landsat data | |
| CN111680866B (en) | Evaluation method, application and device for marine ecological protection importance | |
| Eid et al. | Coastal wetland vegetation features and digital Change Detection Mapping based on remotely sensed imagery: El-Burullus Lake, Egypt | |
| Peng et al. | Urban flood mapping with bitemporal multispectral imagery via a self-supervised learning framework | |
| CN112418506A (en) | Coastal zone wetland ecological safety pattern optimization method and device based on machine learning | |
| CN111310681B (en) | A remote sensing extraction method of mangrove distribution incorporating geoscience knowledge | |
| CN114581784B (en) | A construction method of long-term year-by-year mangrove remote sensing monitoring products | |
| CN115063437B (en) | Mangrove canopy visible light image index feature analysis method and system | |
| CN117409313B (en) | A method for constructing a phenological decline index for Spartina alterniflora based on Sentinel-2 optical images | |
| CN116012723A (en) | Wetland type extraction method and device based on time sequence remote sensing image and electronic equipment | |
| CN109657598A (en) | Seashore wetland Classification in Remote Sensing Image method based on Stratified Strategy | |
| Xu et al. | Mapping and analyzing the annual dynamics of tidal flats in the conterminous United States from 1984 to 2020 using Google Earth Engine | |
| Peng et al. | Detailed wetland-type classification using Landsat-8 time-series images: A pixel-and object-based algorithm with knowledge (POK) | |
| Sun et al. | Mapping China’s coastal aquaculture ponds expansion with sentinel-2 images during 2017–2021 | |
| Wang et al. | Long-term human expansion and the environmental impacts on the coastal zone of China | |
| CN112085725B (en) | Residual film residual quantity detection method and early warning system based on heuristic iterative algorithm | |
| Lemenkova | Artificial intelligence for computational remote sensing: Quantifying patterns of land cover types around cheetham wetlands, port phillip bay, Australia | |
| CN114202551B (en) | A method and device for hierarchical mapping of karst rocky desertification | |
| CN114596300B (en) | Coastal culture remote sensing information processing method and device, electronic equipment and storage medium | |
| CN115953684A (en) | Disaster damage assessment method, system, electronic equipment and medium for aquaculture area | |
| Sun et al. | Annual change in the distribution and landscape health of mangrove ecosystems in China from 2016 to 2023 with Sentinel imagery | |
| CN112907567A (en) | SAR image ordered artificial structure extraction method based on spatial reasoning method | |
| CN118072182B (en) | Red tide identification method, medium and system based on remote sensing image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |