CN110517575B - Method and device for mapping surface water body - Google Patents
Method and device for mapping surface water body Download PDFInfo
- Publication number
- CN110517575B CN110517575B CN201910771960.0A CN201910771960A CN110517575B CN 110517575 B CN110517575 B CN 110517575B CN 201910771960 A CN201910771960 A CN 201910771960A CN 110517575 B CN110517575 B CN 110517575B
- Authority
- CN
- China
- Prior art keywords
- water body
- target area
- data
- probability
- sample
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B29/00—Maps; Plans; Charts; Diagrams, e.g. route diagram
- G09B29/003—Maps
- G09B29/006—Representation of non-cartographic information on maps, e.g. population distribution, wind direction, radiation levels, air and sea routes
- G09B29/007—Representation of non-cartographic information on maps, e.g. population distribution, wind direction, radiation levels, air and sea routes using computer methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Ecology (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- General Physics & Mathematics (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
A method and a device for a surface water body map relate to the technical field of data processing. The surface water body mapping method comprises the steps of firstly obtaining original data of a target area when a surface water body mapping result is generated, and then preprocessing the original data to obtain a candidate water body sample set and a target area image element set of the target area; then, processing the target area pixel set through a pre-constructed prediction neural network and the candidate water body sample set to obtain a first probability that each pixel in the target area pixel set is marked as a water body; then, calculating a second probability that each pixel in the target area pixel set belongs to the water body according to the first probability; and finally, generating a surface water body mapping result of the target area according to the second probability. The method can avoid the interference of obstacles (such as a built-up area, shadows and the like), improve the water body mapping precision, reduce the artificial interference and participation and improve the automation degree of remote sensing water body mapping.
Description
Technical Field
The application relates to the technical field of data processing, in particular to a method and a device for a surface water body map.
Background
Surface water bodies such as rivers, lakes, reservoirs and the like play an important role in maintaining the social and economic development and the ecological system balance. However, the spatial distribution and physicochemical components of the surface water body are greatly changed under the influence of climate change and human activities, so that the monitoring of the surface water dynamic state in time is of great significance for water resource management, water damage prevention and control, water environment protection and other water-related research and planning. The existing water body mapping method is usually a water body index method, and the identification and extraction of water body information are realized by directly performing band operation on original data to enhance water body signals and inhibit non-water body signals. However, in practice, it is found that the existing water body mapping method is easily interfered by obstacles (such as clouds, cloud shadows, building shadows, terrain shadows, and the like) in original data when generating the water body mapping, and thus the obtained water body mapping has large error and low accuracy.
Disclosure of Invention
The embodiment of the application aims to provide a method and a device for mapping a surface water body, which can avoid the interference of obstacles and further improve the accuracy of water body mapping.
The embodiment of the application provides a method for mapping a surface water body in a first aspect, which comprises the following steps:
acquiring original data for surface water body mapping of a target area, and preprocessing the original data to obtain a candidate water body sample set and a target area image element set of the target area;
processing the target area pixel set through a pre-constructed prediction neural network and the candidate water body sample set to obtain a first probability that each pixel in the target area pixel set is marked as a water body;
calculating a second probability that each pixel in the target area pixel set belongs to the water body according to the first probability;
and generating a surface water body mapping result of the target area according to the second probability.
In the implementation process, when an earth surface water body drawing result is generated, original data of a target area is obtained, then the original data is preprocessed to obtain a candidate water body sample set and a target area image element set of the target area, then a verification sample subset is determined according to the candidate water body sample set and the target area image element set of the target area, a first probability that each image element in the target area image element set is marked as a water body is calculated, further, a second probability that each image element in the target area image element set belongs to the water body is calculated according to the first probability, and finally, an earth surface water body drawing result is generated according to the second probability.
Further, the original data comprises satellite remote sensing data of the target area and crowd-sourced map data of the target area;
preprocessing the original data to obtain a candidate water body sample set and a target area image element set of the target area, wherein the steps of:
acquiring a target area image element set of the target area according to the satellite remote sensing data, and preprocessing the satellite remote sensing data to obtain a target water body index;
and preprocessing the crowdsourcing map data according to the target water body index to obtain a candidate water body sample set.
In the implementation process, the satellite remote sensing data and the crowdsourcing map data are combined, so that the accuracy of ground feature identification and extraction is improved, meanwhile, the satellite remote sensing data and the crowdsourcing map data are preprocessed respectively, the interference of the atmosphere on the satellite remote sensing image and the error information in the crowdsourcing map data are further reduced, and the accuracy of the generated water body drawing result is improved.
Further, preprocessing the satellite remote sensing data to obtain a target water body index, and the method comprises the following steps:
carrying out radiometric calibration processing on the satellite remote sensing data to obtain radiance data;
performing atmospheric correction processing on the radiance data to obtain the earth surface reflectivity;
and calculating a target water body index according to the surface reflectivity.
In the implementation process, before the target water body index is calculated, radiometric calibration and atmospheric correction processing are carried out on the satellite remote sensing data to obtain the earth surface reflectivity, and finally the target water body index is calculated according to the earth surface reflectivity, so that the interference of atmospheric absorption and scattering can be reduced, the calculated target water body index is more reliable, and the error is small.
Further, preprocessing the crowd-sourced map data according to the target water body index to obtain a candidate water body sample set, including:
extracting element information related to a water body from the crowdsourcing map data, wherein the element information comprises planar element information and linear element information;
establishing a buffer of the linear element information in the crowd-sourced map data;
merging the planar element information and the linear element information to obtain merged data;
rasterizing the merged data to obtain crowd-sourced map water mask data;
and filtering the crowdsourcing map water mask data according to the target water body index to obtain a candidate water body sample set.
In the implementation process, the crowd-sourced map data includes rich water-related information such as rivers, streams and river banks, and also includes other non-water body information. By establishing a buffer area of the linear element information, more comprehensive water body information can be obtained. On the other hand, false water pixels in the crowdsourcing map water mask data can be removed by filtering the crowdsourcing map water mask data, so that a high-precision candidate water sample set is obtained, and the precision of generating a water body drawing result is improved.
Further, the step of processing the target area pixel set through a pre-constructed prediction neural network and the candidate water body sample set to obtain a first probability that each pixel in the target area pixel set is marked as a water body comprises the following steps:
selecting a preset first amount of background sample data from the satellite remote sensing data, and merging the background sample data and the candidate water body sample set to obtain a merged sample set;
determining a training sample subset and a validation sample subset from the merged sample set;
acquiring a pre-constructed prediction neural network, and processing the prediction neural network through the training sample subset to obtain processing data;
performing parameter correction on the prediction neural network according to the processing data and the verification sample subset to obtain a trained prediction neural network;
and processing the target area pixel set through the trained predictive neural network to obtain a first probability that each pixel in the target area pixel set is marked as a water body.
In the implementation process, when the first probability that each pixel in each target area pixel set is marked as the water body is calculated, a verification sample subset and a training sample subset are determined according to a candidate water body sample set and the target area pixel set of the target area, then the pre-constructed prediction neural network is trained through the training sample subset to obtain the trained prediction neural network, and then the first probability that each pixel in the target area pixel set is marked as the water body is calculated through the trained prediction neural network.
Further, calculating a second probability that each pixel in the target area pixel set belongs to the body of water according to the first probability, including:
processing the water body samples in the verification sample subset through the trained predictive neural network to obtain a third probability of each water body sample in the verification sample subset;
calculating the probability mean value of all water body samples in the verification sample subset according to the third probability;
and calculating a second probability that each pixel in the target area pixel set belongs to the water body according to the probability mean value and the first probability.
In the implementation process, a third probability of each water body sample in the verification sample subset is calculated through the trained predictive neural network; and then calculating the probability mean value of all water body samples in the verification sample subset according to the third probability, and finally calculating the second probability that each pixel in the target area pixel set belongs to the water body according to the probability mean value and the first probability.
Further, generating a surface water body mapping result of the target region according to the second probability, including:
obtaining a segmentation threshold interval, and constructing a positive sample-background scene confusion matrix according to the merged sample set;
determining a target segmentation threshold according to the positive sample-background scene confusion matrix and the segmentation threshold interval;
and generating a surface water body mapping result of the target area according to the target segmentation threshold and the second probability.
In the implementation process, before the surface water body mapping result is generated, a target segmentation threshold needs to be determined, and according to the positive sample-background scene confusion matrix and the segmentation threshold interval, the most appropriate segmentation threshold (namely the target segmentation threshold) can be determined, so that the optimal surface water body mapping result is generated.
Further, constructing a positive sample-background scene confusion matrix from the merged sample set, comprising:
determining a segmentation sample subset from the merged sample set, and determining a plurality of segmentation threshold values to be selected from the segmentation threshold value interval;
and constructing a positive sample-background scene confusion matrix corresponding to each to-be-selected segmentation threshold according to the segmentation sample subset.
In the implementation process, after the segmentation threshold to be selected is determined, the segmentation sample subset is determined from the merged sample set, and then the positive sample-background scene confusion matrix corresponding to each segmentation threshold to be selected is constructed according to the segmentation sample subset.
Further, determining a target segmentation threshold according to the positive sample-background scene confusion matrix and the segmentation threshold interval, including:
calculating the surface water body drawing precision corresponding to each to-be-selected segmentation threshold according to the positive sample-background scene confusion matrix corresponding to the to-be-selected segmentation threshold;
and determining the segmentation threshold to be selected corresponding to the maximum surface water body drawing precision as a target segmentation threshold.
In the implementation process, the to-be-selected segmentation threshold corresponding to the maximum surface water body drawing precision is determined as the target segmentation threshold, namely the optimal segmentation threshold, the surface water body drawing precision corresponding to the optimal segmentation threshold is maximum, and the optimal surface water body drawing is obtained according to the optimal segmentation threshold.
A second aspect of the embodiments of the present application provides a surface water body mapping apparatus, including:
the data acquisition module is used for acquiring original data for surface water body mapping of the target area;
the preprocessing module is used for preprocessing the original data to obtain a candidate water body sample set and a target area image element set of the target area;
the first probability calculation module is used for processing the target area pixel set through a pre-constructed prediction neural network and the candidate water body sample set to obtain a first probability that each pixel in the target area pixel set is marked as a water body;
the second probability calculation module is used for calculating a second probability that each pixel in the target area pixel set belongs to the water body according to the first probability;
and the mapping module is used for generating a surface water body mapping result of the target area according to the second probability.
In the implementation process, when an earth surface water body drawing result is generated, a data acquisition module firstly acquires original data of a target area, and preprocesses the original data to obtain a candidate water body sample set and a target area image element set of the target area, then a first probability calculation module determines a verification sample subset according to the candidate water body sample set and the target area image element set of the target area, and calculates a first probability that each image element in the target area image element set is marked as a water body, further a second probability calculation module calculates a second probability that each image element in the target area image element set belongs to the water body according to the first probability, and finally, a drawing module generates the earth surface water body drawing result according to the second probability.
A third aspect of embodiments of the present application provides a computer device, comprising a memory for storing a computer program and a processor for executing the computer program to make the computer device execute part or all of the surface water mapping method disclosed in the first aspect.
A fourth aspect of embodiments of the present application provides a computer-readable storage medium storing the computer program used in the computer device according to the third aspect.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments of the present application will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and that those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.
FIG. 1 is a schematic flow chart of a method for mapping a surface water system provided in an embodiment of the present application;
FIG. 2 is a schematic flow chart of a method for mapping a surface water system provided in the second embodiment of the present application;
FIG. 3 is a graph of a segmentation threshold versus a graph accuracy provided in a second embodiment of the present application;
fig. 4 is a schematic structural diagram of a surface water body mapping apparatus according to a third embodiment of the present application;
fig. 5 is a schematic structural diagram of another surface water body mapping apparatus provided in the third embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Example 1
Referring to fig. 1, fig. 1 is a schematic block diagram of a flow of a method for producing a map of a surface water according to an embodiment of the present disclosure. As shown in fig. 1, the surface water body mapping method includes:
s101, obtaining original data used for surface water body mapping of a target area, and preprocessing the original data to obtain a candidate water body sample set and a target area image element set of the target area.
In the embodiment of the application, when the surface water body mapping result of the target area needs to be obtained, the obtained original data for surface water body mapping is the geographic data of the target area. The geographic data may be satellite remote sensing image data, map data, or the like, and the embodiment of the present application is not limited thereto.
S102, processing the target area pixel set through a pre-constructed prediction neural network and a candidate water body sample set to obtain a first probability that each pixel in the target area pixel set is marked as a water body.
In the embodiment of the present application, the predictive Neural network may be a Back Propagation (BP) Neural network, a Convolutional Neural Network (CNN), a Radial Basis Function (RBF) Neural network, and the like, and the embodiment of the present application is not limited thereto.
S103, calculating a second probability that each pixel in the target area pixel set belongs to the water body according to the first probability.
And S104, generating a surface water body mapping result of the target area according to the second probability.
In the embodiment of the application, after the second probability that each pixel in the target area pixel set belongs to the water body is calculated, threshold segmentation processing is performed on the second probability according to the segmentation threshold, the pixel with the second probability being greater than the target segmentation threshold is determined as the water body, the pixel with the second probability being less than the target segmentation threshold is determined as the non-water body, and then the corresponding surface water body mapping result is obtained.
Therefore, the method for drawing the surface water body described in fig. 1 can avoid the interference of the obstacles, and further improve the precision of the water body drawing.
Example 2
Referring to fig. 2, fig. 2 is a schematic flow chart of a method for mapping a surface water according to an embodiment of the present application. As shown in fig. 2, the surface water body mapping method includes:
s201, obtaining raw data for surface water body mapping of the target area.
In the embodiment of the application, the original data comprises satellite remote sensing data and crowd-sourced map data. Specifically, the satellite remote sensing data may be Sentinel second image (Sentinel-2 MSI) data, WorldView satellite data, QuickBird satellite data, high-resolution first satellite data, resource third satellite data, Landsat satellite data, and the like, and the crowd-sourced map data may be Open Street Map (OSM) data, and the embodiment of the present application is not limited thereto.
In the embodiment of the application, the Sentinel image two number (Sentinel-2 MSI) is a novel multispectral remote sensing image, and has 13 spectral bands (4 bands are 10 m spatial resolution, 6 bands are 20 m spatial resolution, and 3 bands are 60 m spatial resolution), and the revisit period is about 5 days. Compared with common medium-resolution remote sensing data (Landsat ETM/ETM +/OLI satellite images), the data has higher spatial resolution and revisiting period, and is more suitable for surface water body mapping and dynamic monitoring.
S202, acquiring a target area image element set of a target area according to the satellite remote sensing data, and preprocessing the satellite remote sensing data to obtain a target water body index.
As an optional implementation manner, the preprocessing the satellite remote sensing data to obtain the target water body index may include the following steps:
carrying out radiometric calibration processing on the satellite remote sensing data to obtain radiance data;
atmospheric correction processing is carried out on the radiation brightness data to obtain the earth surface reflectivity;
and calculating the target water body index according to the surface reflectivity.
In the above embodiment, the target Water body Index may be one of an automatic Water body Extraction Index (AWEI), a Normalized Difference Water Index (NDWI), a Multi-Band Water Index (MBWI), a Multi-spectral Water Index (MuWI)), an All-Band Water Index (ABWI), and the like, and this embodiment of the present application is not limited thereto.
In the above embodiments, the surface reflectivity is the ratio of the amount of reflected radiation to the amount of incident radiation on the ground, and is indicative of the absorption and reflection capabilities of the ground for solar radiation. The greater the reflectivity, the less solar radiation is absorbed by the ground; the smaller the reflectivity, the more solar radiation is absorbed by the ground.
In the above embodiment, the radiometric calibration is a process of converting the brightness gray value of the image into absolute radiance when the user needs to calculate the spectral reflectance or spectral radiance of the feature, or needs to compare images acquired by different sensors at different times.
In the above embodiments, atmospheric correction means that the total radiance of the ground target finally measured by the sensor is not a reflection of the true reflectivity of the ground, including the radiation amount error caused by atmospheric absorption, especially scattering. Atmospheric correction is the process of inverting the real surface reflectivity of the ground object by eliminating the radiation errors caused by atmospheric influence.
After step S202, the following steps are also included:
s203, preprocessing crowdsourcing map data according to the target water body index to obtain a candidate water body sample set.
In the embodiment of the present application, openstreetmap (osm) data is a street map generated by a user, and includes rich geographic information, such as water bodies, buildings, and the like. OSM data is provided by a large number of volunteers and lacks the necessary quality control. Therefore, through the preprocessing of the OSM data in the step S203, false water body information possibly existing in the OSM data can be eliminated, and the accuracy of the water body sample is improved.
As an optional implementation, preprocessing the crowdsourcing map data according to the target water body index to obtain a candidate water body sample set, may include the following steps:
extracting element information related to the water body from crowd-sourced map data, wherein the element information comprises planar element information and linear element information;
establishing a buffer area of linear element information in crowdsourcing map data;
merging the planar element information and the linear element information to obtain merged data;
rasterizing the merged data to obtain crowd-sourced map water mask data;
and filtering the crowdsourcing map water mask data according to the target water index to obtain a candidate water sample set.
In the above embodiment, when the crowd-sourced map data is OSM data, for the OSM data, first, element information related to a water body is extracted from the OSM data, and the element information related to the water body includes planar element information and linear element information, where the planar element information includes one or more of reservoir (reservoir) element information, water body (water) element information, river bank (riverbank) element information, and the like; the linear element information includes one or more of river (canal) element information, river (river) element information, stream (stream) element information, and the like, and the present embodiment is not limited thereto.
In the above embodiment, the reservoir (reservoir) element information may be extracted from the landump layer of the OSM data; extracting water body (water) element information and river bank (riverbank) element information from a natural map layer of OSM data; river (canal) element information, river (river) element information, and stream (stream) element information are extracted from the water layer of the OSM data.
In the foregoing embodiment, the buffer of the water layer element in the crowdsourcing map data may be constructed according to a preset buffer size, where the preset buffer size includes a preset buffer radius. In practical use, the preset buffer area radius may be 70 meters, and the like, and the embodiment of the present application is not limited thereto.
In the foregoing embodiment, when rasterizing the merged data according to a preset grid value, where the grid value may be 10 meters, the present application is not limited to this embodiment.
As a further optional implementation, the filtering processing on the crowd-sourced map water mask data according to the target water index to obtain a candidate water sample set may include the following steps:
constructing a filtering model according to the target water body index;
and filtering the crowdsourcing map water mask data according to the filtering model to obtain a candidate water sample set.
In the embodiment, the filtering model is constructed according to the target water body index >0, false water body pixels in crowdsourcing map water body mask data can be eliminated, a candidate water body sample set is obtained, the influence of the false water body pixel data on surface water body drawing is effectively avoided, and the improvement of the surface water body drawing precision is facilitated.
In the embodiment of the present application, by implementing the steps S202 to S203, the raw data can be preprocessed to obtain a candidate water sample set and a target area image element set of a target area.
After step S203, the following steps are also included:
s204, selecting a preset first amount of background sample data from the satellite remote sensing data, merging the background sample data and the candidate water body sample set to obtain a merged sample set, and determining a training sample subset and a verification sample subset from the merged sample set.
In the embodiment of the application, the training sample subset and the verification sample subset can be determined from the merged sample set according to a preset partition rule. The preset division rule comprises random division according to a preset proportion, wherein the preset proportion comprises 35 of the number of samples in the training sample subset accounting for the total number of samples in the combined sample set, and 35 of the number of verification samples in the subset accounting for the total number of samples in the combined sample set
In the above embodiment, the mode of determining the training sample subset and the verification sample subset from the combined sample set may be random selection, and selection of training samples does not need to be performed manually, so that consumption of manpower and material resources is reduced, and the efficiency of model training is improved.
S205, obtaining a pre-constructed prediction neural network, processing the prediction neural network through the training sample subset to obtain processing data, and performing parameter correction on the prediction neural network according to the processing data and the verification sample subset to obtain the trained prediction neural network.
In the above embodiment, the background sample data and the candidate water body sample set are merged to obtain a merged sample set, specifically, pixels with a preset initial selection number (for example, 5%) are randomly selected from the candidate water body sample set as a preliminary water body sample, and the background sample data is randomly selected from the satellite remote sensing data according to the preset initial selection number (for example, 13). Further, combining the background sample data and the preliminary water body sample to obtain a combined sample set, and then determining a training sample subset and a verification sample subset from the combined sample set according to a preset sample selection number (for example, the ratio of the number of pixels in the training sample subset to the number of pixels in the target area pixel set is 3:1), wherein the training sample subset is used for model training of the prediction neural network, and the verification sample subset is used for model parameter optimization of the prediction neural network.
S206, processing the target area pixel set through the trained prediction neural network to obtain a first probability that each pixel in the target area pixel set is marked as a water body.
In the embodiment of the application, by implementing the steps S204 to S206, the target area pixel set can be processed through the pre-constructed prediction neural network and the candidate water body sample set, so as to obtain a first probability that each pixel in the target area pixel set is marked as a water body.
And S207, processing the water body samples in the verification sample subset through the trained predictive neural network to obtain a third probability of each water body sample in the verification sample subset, and calculating a probability mean value of all the water body samples in the verification sample subset according to the third probability.
S208, according to the probability mean value and the first probability, calculating a second probability that each pixel in the pixel set of the target area belongs to the water body.
In the embodiment of the application, for each pixel in the target area pixel set, a calculation formula for calculating the second probability that the pixel belongs to the water body is as follows:
wherein, PWA second probability that the pixels in the pixel set of the target area belong to the water body, c is a probability mean value of all water body samples in the verification sample subset, PobsA first probability that pixels in a target area pixel set are marked as water.
In the embodiment of the present application, by implementing the steps S205 to S206, the second probability that each pixel in the target area pixel set belongs to the water body can be calculated according to the first probability.
And S209, generating a surface water body mapping result of the target area according to the second probability.
As an alternative implementation, generating the surface water body mapping result of the target region according to the second probability may include the following steps:
acquiring a segmentation threshold interval, and constructing a positive sample-background scene confusion matrix according to the merged sample set;
determining a target segmentation threshold according to the positive sample-background scene confusion matrix and the segmentation threshold interval;
and generating a surface water body mapping result of the target area according to the target segmentation threshold and the second probability.
In the above embodiment, the division threshold interval may be preset, and specifically may be [ -1,1], and the like, and this embodiment of the present application is not limited thereto.
As a further alternative embodiment, constructing a positive sample-background scene confusion matrix from the merged sample set may include the following steps:
determining a segmentation sample subset from the combined sample set, and determining a plurality of segmentation thresholds to be selected from a segmentation threshold interval;
and constructing a positive sample-background scene confusion matrix corresponding to each to-be-selected segmentation threshold according to the segmentation sample subset.
In the above embodiment, for one candidate segmentation threshold, the corresponding positive sample-background scene confusion matrix is represented as follows:
because the segmented sample subset is obtained from the merged sample set, and the merged sample set is obtained according to the background sample data and the candidate water body sample set, s ═ 1 represents a pixel belonging to the candidate water body sample set in the segmented pixel subset, s ═ 0 represents a pixel belonging to the background sample data in the segmented sample subset, y ═ 1 represents that the result of prediction by the trained prediction neural network is a water body, y ═ 0 represents that the result of prediction by the trained prediction neural network is a non-water body, and TP ', FP', FN ', TN' all represent elements of the positive sample-background scene confusion matrix.
In the foregoing embodiment, a preset second number of the split sample subsets may be randomly selected from the merged sample set, where the preset second number may be 15 of the total number of pixels in the merged sample set, and the application is not limited to this embodiment.
As a further alternative, determining the target segmentation threshold according to the positive sample-background scene confusion matrix and the segmentation threshold interval may include the following steps:
calculating the surface water body drawing precision corresponding to each to-be-selected segmentation threshold according to the positive sample-background scene confusion matrix corresponding to the to-be-selected segmentation threshold;
and determining the segmentation threshold to be selected corresponding to the maximum surface water body drawing precision as a target segmentation threshold.
In the above embodiment, for each candidate segmentation threshold, the formula for calculating the surface water body mapping accuracy corresponding to the candidate segmentation threshold is as follows:
wherein, FmAnd for the surface water body drawing precision corresponding to the segmentation threshold to be selected, TP ', FP ' and FN ' are all elements of a positive sample-background scene confusion matrix corresponding to the segmentation threshold to be selected.
Referring to fig. 3, fig. 3 is a graph of a segmentation threshold versus a graph accuracy according to the present embodiment. As shown in fig. 3, in the graph, the abscissa represents each of the candidate division threshold values in the division threshold value interval, and the ordinate represents the surface water body drawing accuracy Fm(i.e., F-score). As can be seen from FIG. 3, as the candidate segmentation threshold increases, FmExhibit a first stabilization, a second increase and a final decreaseWhen the division threshold is 0.32, the water body drawing accuracy (F)m) The maximum value, Max (F-score) ═ 1.7825, was reached, and therefore, the target segmentation threshold value could be determined to be 0.32.
In the above embodiment, after the target segmentation threshold is determined, the pixels with the second probability greater than the target segmentation threshold are determined as water pixels, and the pixels with the second probability less than the target segmentation threshold are determined as non-water pixels, so as to obtain the corresponding surface water body mapping result.
Therefore, the method for drawing the surface water body described in fig. 2 can avoid the interference of the obstacles, and further improve the precision of the water body drawing.
Example 3
Referring to fig. 4, fig. 4 is a schematic block diagram illustrating a structure of a surface water body mapping apparatus according to an embodiment of the present application. As shown in fig. 4, the surface water body mapping apparatus includes:
a data acquisition module 310 for acquiring raw data for surface water mapping of a target area.
And the preprocessing module 320 is configured to preprocess the original data to obtain a candidate water sample set and a target area image element set of a target area.
The first probability calculation module 330 is configured to process the target area pixel set through a pre-constructed prediction neural network and a candidate water body sample set, so as to obtain a first probability that each pixel in the target area pixel set is marked as a water body.
And the second probability calculating module 340 is configured to calculate a second probability that each pixel in the target area pixel set belongs to the water body according to the first probability.
And the drawing module 350 is configured to generate a surface water drawing result of the target area according to the second probability.
In the embodiment of the present application, the original data includes satellite remote sensing data, crowd-sourced map data, and the like, and the embodiment of the present application is not limited thereto.
Referring to fig. 5, fig. 5 is a schematic block diagram of a structure of another surface water body mapping apparatus according to an embodiment of the present application. The surface water body mapping apparatus shown in fig. 5 is obtained by optimizing the surface water body mapping apparatus shown in fig. 4, and as shown in fig. 5, the preprocessing module 320 includes:
and the pixel obtaining submodule 321 is configured to obtain a target area pixel set of the target area according to the satellite remote sensing data.
And the first preprocessing submodule 322 is used for preprocessing the satellite remote sensing data to obtain a target water body index.
And the second preprocessing submodule 323 is used for preprocessing crowdsourcing map data according to the target water body index to obtain a candidate water body sample set.
As a further alternative, the first preprocessing submodule 321 includes:
and the radiometric calibration unit is used for carrying out radiometric calibration processing on the satellite remote sensing data to obtain radiance data.
And the atmosphere correction unit is used for carrying out atmosphere correction processing on the radiation brightness data to obtain the earth surface reflectivity.
And the calculating unit is used for calculating the target water body index according to the surface reflectivity.
As a further optional embodiment, the second preprocessing submodule 322 includes:
and the information extraction unit is used for extracting element information related to the water body from the crowd-sourced map data, and the element information comprises planar element information and linear element information.
A buffer area establishing unit for establishing a buffer area of the linear element information in the crowd-sourced map data.
And the merging unit is used for merging the buffer areas of the planar element information and the linear element information to obtain merged data.
And the rasterizing unit is used for rasterizing the merged data to obtain crowd-sourced map water body mask data.
And the filtering unit is used for filtering the crowdsourcing map water mask data according to the target water index to obtain a candidate water sample set.
As an alternative implementation, the first probability calculation module 330 includes:
the sample acquisition submodule 331 is configured to select a preset first amount of background sample data from the satellite remote sensing data, and merge the background sample data and the candidate water body sample set to obtain a merged sample set; and determining a training sample subset and a validation sample subset from the combined sample set;
the training submodule 332 is configured to obtain a pre-constructed predicted neural network, and process the predicted neural network through the training sample subset to obtain processed data; performing parameter correction on the prediction neural network according to the processing data and the verification sample subset to obtain a trained prediction neural network;
the processing submodule 333 is configured to process the target area pixel set through the trained predictive neural network, and obtain a first probability that each pixel in the target area pixel set is marked as a water body.
As an optional implementation, the second probability calculation module 340 includes:
the first calculation submodule 341 is configured to process the water body samples in the verification sample subset through the trained predictive neural network, so as to obtain a third probability of each water body sample in the verification sample subset; and calculating the probability mean value of all water body samples in the verification sample subset according to the third probability.
And the second calculating submodule 342 is configured to calculate a second probability that each pixel in the pixel set of the target area belongs to the water body according to the probability mean and the first probability.
As an alternative implementation, the charting module 350 includes:
the section obtaining submodule 351 is configured to obtain a segmentation threshold section.
And a matrix construction sub-module 352 configured to construct a positive sample-background scene confusion matrix according to the merged sample set.
And the threshold determination submodule 353 is used for determining a target segmentation threshold according to the positive sample-background scene confusion matrix and the segmentation threshold interval.
And the mapping generation submodule 354 is used for generating a surface water mapping result of the target area according to the target segmentation threshold and the second probability.
As a further alternative embodiment, the matrix construction sub-module 352 includes:
a subset determination unit for determining a subset of the split samples from the combined sample set.
And the candidate threshold determining unit is used for determining a plurality of candidate division thresholds from the division threshold interval.
And the matrix construction unit is used for constructing a positive sample-background scene confusion matrix corresponding to each to-be-selected segmentation threshold according to the segmentation sample subsets.
As a further optional embodiment, the threshold determination sub-module 353 includes:
and the precision calculation unit is used for calculating the surface water body drawing precision corresponding to each to-be-selected segmentation threshold according to the positive sample-background scene confusion matrix corresponding to the to-be-selected segmentation threshold.
And the target determining unit is used for determining the segmentation threshold to be selected corresponding to the maximum drawing precision of the surface water body as the target segmentation threshold.
Therefore, the earth surface water body drawing device described in the embodiment can avoid the interference of the obstacles, and further improves the accuracy of water body drawing.
In addition, the invention also provides computer equipment. The computer device comprises a memory and a processor, wherein the memory can be used for storing a computer program, and the processor can execute the computer program to make the computer device execute the functions of the method or the modules in the surface water body mapping device.
The memory may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the mobile terminal, and the like. Further, the memory may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.
The embodiment also provides a computer storage medium for storing a computer program used in the computer device.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Claims (7)
1. A method of surface water mapping, comprising:
acquiring original data for surface water body mapping of a target area, wherein the original data comprises satellite remote sensing data of the target area and crowd-sourced map data of the target area;
acquiring a target area image element set of the target area according to the satellite remote sensing data, and preprocessing the satellite remote sensing data to obtain a target water body index;
preprocessing the crowdsourcing map data according to the target water body index to obtain a candidate water body sample set;
processing the target area pixel set through a pre-constructed prediction neural network and the candidate water body sample set to obtain a first probability that each pixel in the target area pixel set is marked as a water body;
calculating a second probability that each pixel in the target area pixel set belongs to the water body according to the first probability;
generating a surface water body mapping result of the target area according to the second probability;
preprocessing the satellite remote sensing data to obtain a target water body index, wherein the preprocessing comprises the following steps:
carrying out radiometric calibration processing on the satellite remote sensing data to obtain radiance data;
performing atmospheric correction processing on the radiance data to obtain the earth surface reflectivity;
calculating a target water body index according to the surface reflectivity;
preprocessing the crowdsourcing map data according to the target water body index to obtain a candidate water body sample set, wherein the preprocessing comprises the following steps:
extracting element information related to a water body from the crowdsourcing map data, wherein the element information comprises planar element information and linear element information;
establishing a buffer of the linear element information in the crowd-sourced map data;
merging the planar element information and the linear element information to obtain merged data;
rasterizing the merged data to obtain crowd-sourced map water mask data;
and filtering the crowdsourcing map water mask data according to the target water body index to obtain a candidate water body sample set.
2. The surface water body mapping method according to claim 1, wherein the step of processing the target area pixel set through a pre-constructed prediction neural network and the candidate water body sample set to obtain a first probability that each pixel in the target area pixel set is marked as a water body comprises the steps of:
selecting a preset first amount of background sample data from the satellite remote sensing data, and merging the background sample data and the candidate water body sample set to obtain a merged sample set;
determining a training sample subset and a validation sample subset from the merged sample set;
acquiring a pre-constructed prediction neural network, and processing the prediction neural network through the training sample subset to obtain processing data;
performing parameter correction on the prediction neural network according to the processing data and the verification sample subset to obtain a trained prediction neural network;
and processing the target area pixel set through the trained predictive neural network to obtain a first probability that each pixel in the target area pixel set is marked as a water body.
3. The method for surface water body mapping according to claim 2, wherein calculating a second probability that each pixel in the target area pixel set belongs to a water body according to the first probability comprises:
processing the water body samples in the verification sample subset through the trained predictive neural network to obtain a third probability of each water body sample in the verification sample subset;
calculating the probability mean value of all water body samples in the verification sample subset according to the third probability;
and calculating a second probability that each pixel in the target area pixel set belongs to the water body according to the probability mean value and the first probability.
4. The method for surface water body mapping according to claim 2, wherein generating the surface water body mapping result of the target region according to the second probability comprises:
obtaining a segmentation threshold interval, and constructing a positive sample-background scene confusion matrix according to the merged sample set;
determining a target segmentation threshold according to the positive sample-background scene confusion matrix and the segmentation threshold interval;
and generating a surface water body mapping result of the target area according to the target segmentation threshold and the second probability.
5. The method for surface water mapping according to claim 4, wherein constructing a positive sample-background scene confusion matrix from the merged sample set comprises:
determining a segmentation sample subset from the merged sample set, and determining a plurality of segmentation threshold values to be selected from the segmentation threshold value interval;
and constructing a positive sample-background scene confusion matrix corresponding to each to-be-selected segmentation threshold according to the segmentation sample subset.
6. The method for mapping a surface water body according to claim 5, wherein determining a target segmentation threshold from the positive sample-background scene confusion matrix and the segmentation threshold interval comprises:
calculating the surface water body drawing precision corresponding to each to-be-selected segmentation threshold according to the positive sample-background scene confusion matrix corresponding to the to-be-selected segmentation threshold;
and determining the segmentation threshold to be selected corresponding to the maximum surface water body drawing precision as a target segmentation threshold.
7. A surface water system map apparatus, comprising:
the data acquisition module is used for acquiring original data for surface water body mapping of the target area; the original data comprises satellite remote sensing data of the target area and crowd-sourced map data of the target area;
the preprocessing module is used for preprocessing the original data to obtain a candidate water body sample set and a target area image element set of the target area;
the first probability calculation module is used for processing the target area pixel set through a pre-constructed prediction neural network and the candidate water body sample set to obtain a first probability that each pixel in the target area pixel set is marked as a water body;
the second probability calculation module is used for calculating a second probability that each pixel in the target area pixel set belongs to the water body according to the first probability;
the drawing module is used for generating a surface water body drawing result of the target area according to the second probability;
wherein the preprocessing module comprises:
the pixel acquisition sub-module is used for acquiring a target area pixel set of the target area according to the satellite remote sensing data;
the first preprocessing submodule is used for preprocessing the satellite remote sensing data to obtain a target water body index;
the second preprocessing submodule is used for preprocessing the crowdsourcing map data according to the target water body index to obtain a candidate water body sample set;
wherein the first pre-processing sub-module comprises:
the radiometric calibration unit is used for carrying out radiometric calibration processing on the satellite remote sensing data to obtain radiance data;
the atmosphere correction unit is used for carrying out atmosphere correction processing on the radiation brightness data to obtain the earth surface reflectivity;
the calculation unit is used for calculating a target water body index according to the earth surface reflectivity;
wherein the second preprocessing sub-module comprises:
an information extraction unit configured to extract element information related to a water body from the crowd-sourced map data, the element information including planar element information and linear element information;
a buffer area establishing unit configured to establish a buffer area of the linear element information in the crowd-sourced map data;
a merging unit, configured to merge the planar element information and the buffer area of the linear element information to obtain merged data;
the rasterizing unit is used for rasterizing the merged data to obtain crowd-sourced map water body mask data;
and the filtering unit is used for filtering the crowdsourcing map water mask data according to the target water index to obtain a candidate water sample set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910771960.0A CN110517575B (en) | 2019-08-21 | 2019-08-21 | Method and device for mapping surface water body |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910771960.0A CN110517575B (en) | 2019-08-21 | 2019-08-21 | Method and device for mapping surface water body |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110517575A CN110517575A (en) | 2019-11-29 |
| CN110517575B true CN110517575B (en) | 2021-03-02 |
Family
ID=68627173
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910771960.0A Active CN110517575B (en) | 2019-08-21 | 2019-08-21 | Method and device for mapping surface water body |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110517575B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652882B (en) * | 2020-07-07 | 2020-12-08 | 中国水利水电科学研究院 | Large-scale surface water product drawing precision evaluation method |
| CN111913965B (en) * | 2020-08-03 | 2024-02-27 | 北京吉威空间信息股份有限公司 | Space big data buffer area analysis-oriented method |
| CN112950780B (en) * | 2021-03-12 | 2022-09-06 | 北京理工大学 | Intelligent network map generation method and system based on remote sensing image |
| CN113610882A (en) * | 2021-04-23 | 2021-11-05 | 华北水利水电大学 | Surface water body drawing method and device, electronic equipment and storage medium |
| CN113327214B (en) * | 2021-05-19 | 2022-01-07 | 中国科学院地理科学与资源研究所 | Continuous time series water body remote sensing mapping method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846352A (en) * | 2018-06-08 | 2018-11-20 | 广东电网有限责任公司 | A kind of vegetation classification and recognition methods |
| CN109741285A (en) * | 2019-01-28 | 2019-05-10 | 上海海洋大学 | A kind of construction method and system of underwater picture data set |
| CN110008899A (en) * | 2019-04-02 | 2019-07-12 | 北京市遥感信息研究所 | A kind of visible remote sensing image candidate target extracts and classification method |
| CN110046607A (en) * | 2019-04-26 | 2019-07-23 | 西安因诺航空科技有限公司 | A kind of unmanned aerial vehicle remote sensing image board house or building materials test method based on deep learning |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6731804B1 (en) * | 2000-09-28 | 2004-05-04 | The United States Of America As Represented By The Secretary Of The Army | Thermal luminescence liquid monitoring system and method |
| CN103456122B (en) * | 2013-08-26 | 2015-11-25 | 中国科学技术大学 | A kind of forest fire smoke recognition methods and device |
| CN106023133B (en) * | 2016-04-26 | 2019-03-19 | 武汉大学 | A kind of high-resolution remote sensing image Clean water withdraw method based on multiple features combining processing |
| CN107271367B (en) * | 2017-05-05 | 2019-06-28 | 北京师范大学 | A kind of identifying water boy method and device |
| CN109063754B (en) * | 2018-07-18 | 2020-08-07 | 武汉大学 | Remote sensing image multi-feature joint classification method based on OpenStreetMap |
| CN109211793B (en) * | 2018-09-12 | 2020-10-27 | 中国科学技术大学 | Fire spot identification method combining spectral index and neural network |
| CN109344769A (en) * | 2018-09-29 | 2019-02-15 | 中国资源卫星应用中心 | A kind of photovoltaic plant detection method and system based on remote sensing image |
| CN109472804A (en) * | 2018-11-05 | 2019-03-15 | 南方科技大学 | Land table Clean water withdraw method and apparatus based on remote sensing image |
| CN109977801B (en) * | 2019-03-08 | 2020-12-01 | 中国水利水电科学研究院 | Optical and radar combined regional water body rapid dynamic extraction method and system |
-
2019
- 2019-08-21 CN CN201910771960.0A patent/CN110517575B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846352A (en) * | 2018-06-08 | 2018-11-20 | 广东电网有限责任公司 | A kind of vegetation classification and recognition methods |
| CN109741285A (en) * | 2019-01-28 | 2019-05-10 | 上海海洋大学 | A kind of construction method and system of underwater picture data set |
| CN110008899A (en) * | 2019-04-02 | 2019-07-12 | 北京市遥感信息研究所 | A kind of visible remote sensing image candidate target extracts and classification method |
| CN110046607A (en) * | 2019-04-26 | 2019-07-23 | 西安因诺航空科技有限公司 | A kind of unmanned aerial vehicle remote sensing image board house or building materials test method based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110517575A (en) | 2019-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110517575B (en) | Method and device for mapping surface water body | |
| Li et al. | An automatic method for mapping inland surface waterbodies with Radarsat-2 imagery | |
| CN111274918B (en) | River dryout and cutoff monitoring method and device based on multi-source remote sensing image | |
| Ishola et al. | Dynamics of surface urban biophysical compositions and its impact on land surface thermal field | |
| Stanchev et al. | Integrating GIS and high resolution orthophoto images for the development of a geomorphic shoreline classification and risk assessment—a case study of cliff/bluff erosion along the Bulgarian coast | |
| CN112200362A (en) | Landslide prediction method, device, equipment and storage medium | |
| Lê et al. | Multiscale framework for rapid change analysis from SAR image time series: Case study of flood monitoring in the central coast regions of Vietnam | |
| Kumar et al. | Agricultural land use change analysis using remote sensing and GIS: a case study of Allahabad, India | |
| CN113221813A (en) | Coastline remote sensing extraction method | |
| Kundu et al. | Detection and prediction of sundarban reserve forest using the CA-Markov chain model and remote sensing data | |
| CN114494881A (en) | Method, device and terminal for detecting remote sensing image change based on subdivision grid | |
| Ali et al. | Glacier area changes in Novaya Zemlya from 1986–89 to 2019–21 using object-based image analysis in Google Earth Engine | |
| CN116340863B (en) | Air pollutant prediction method and device, electronic equipment and readable storage medium | |
| CN116106904B (en) | Facility deformation monitoring method and facility deformation monitoring equipment for object MT-InSAR | |
| CN115984689A (en) | Multi-scale earth surface complexity feature extraction and land utilization segmentation method | |
| Jafarzadeh et al. | Increasing the spatial accuracy of the land use map using fusion of optical and radar images of Sentinel and Google Earth Engine | |
| Mammadov et al. | APPLYING AN OBJECT-BASED CLASSIFICATION APPROACH THROUGH A CELLULAR AUTOMATA-MARKOV METHOD IN LANDCOVER/LANDUSE CHANGE DETECTION PROCEDURE" CASE OF THE URMIA LAKE" | |
| Wang et al. | Combining SAR images with land cover products for rapid urban flood mapping | |
| Dung et al. | Comparison of resampling methods on different remote sensing images for Vietnam’s Urban Classification | |
| CN113392377A (en) | Swamp wetland remote sensing extraction method and system based on soil humidity | |
| Bai et al. | The spatial and temporal availability differences of cloud-free landsat images over three gorges reservoir area | |
| Wang et al. | Deep Learning-Based Method to Extend the Time Series of Global Annual Viirs-Like Nighttime Light Data | |
| CN118135424B (en) | Ecological environment supervision method and device based on remote sensing image and GIS | |
| Rajendra et al. | Understanding the dynamics of Gautala Autramghat forest: A digital image classification approach | |
| CN115424131B (en) | Cloud detection optimal threshold selection method, cloud detection method and cloud detection system |
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 |