CN110517575A - A kind of surface water body drafting method and device - Google Patents
A kind of surface water body drafting method and device Download PDFInfo
- Publication number
- CN110517575A CN110517575A CN201910771960.0A CN201910771960A CN110517575A CN 110517575 A CN110517575 A CN 110517575A CN 201910771960 A CN201910771960 A CN 201910771960A CN 110517575 A CN110517575 A CN 110517575A
- Authority
- CN
- China
- Prior art keywords
- water body
- target area
- probability
- sample set
- pixel
- 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
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
Abstract
A kind of surface water body drafting method and device, are related to technical field of data processing.Wherein, when surface water body drafting method is included in generation surface water body charting results, the initial data of target area is obtained first, then initial data is pre-processed to obtain the target area pixel collection of candidate water body sample set and target area；Then the target area pixel collection is handled by the prediction neural network and the candidate water body sample set that construct in advance, obtains the first probability that the target area pixel concentrates each pixel to be marked as water body；Then each pixel is concentrated to belong to the second probability of water body according to the first probability calculation target area pixel；Finally, generating the surface water body charting results of target area according to the second probability.This method can be avoided the interference of barrier (such as built-up areas, shade), promote water body cartographic accuracy, while reducing artificial interference and participation, improve the degree of automation of remote sensing water body drawing.
Description
Technical field
This application involves technical field of data processing, in particular to a kind of surface water body drafting method and device.
Background technique
The surface water bodies such as river, lake, reservoir play important in maintaining socio-economic development and ecosystem balance
Role.However, the influence of climate variation and mankind's activity, the spatial distribution and physical and chemical composition of surface water body are occurring huge
Therefore big variation monitors surface water dynamic in time, water-related for water resources management, water damage prevention and treatment, water environment protection etc. to grind
Study carefully and planning is of great significance.Existing water body drafting method is often water body index method, directly by carrying out to initial data
Band math inhibits non-water body signal to enhance water body signal, realizes the identification and extraction of Water-Body Information.However, in practice
It was found that existing water body drafting method is easy when generating water body drawing by (such as yin of cloud, cloud of barrier in initial data
Shadow, the shade of building, shade of landform etc.) interference, and then the water body Drawing Error caused is big, and accuracy is low.
Summary of the invention
The embodiment of the present application is designed to provide a kind of surface water body drafting method and device, can be avoided barrier
Interference, and then promote the precision of water body drawing.
The embodiment of the present application first aspect provides a kind of surface water body drafting method, comprising:
The initial data charted for the surface water body of target area is obtained, and the initial data is pre-processed,
Obtain the target area pixel collection of candidate water body sample set and the target area；
By the prediction neural network that constructs in advance and the candidate water body sample set to the target area pixel collection into
Row processing, obtains the first probability that the target area pixel concentrates each pixel to be marked as water body；
Each pixel is concentrated to belong to the second probability of water body according to target area pixel described in first probability calculation；
The surface water body charting results of the target area are generated according to second probability.
During above-mentioned realization, when generating surface water body charting results, the initial data of target area is first obtained, so
Initial data is pre-processed again afterwards to obtain the target area pixel collection of candidate water body sample set and target area, then root again
Verifying sample set is determined according to candidate water body sample set and the target area pixel collection of target area, and calculates target area pixel
It concentrates each pixel to be marked as the first probability of water body, further, is concentrated according to the first probability calculation target area pixel
Each pixel belongs to the second probability of water body, finally, generating surface water body charting results according to the second probability.
Further, the initial data includes the satellite remote sensing date of the target area and the crowd of the target area
Packet map datum；
The initial data is pre-processed, the target area picture of candidate water body sample set and the target area is obtained
Metaset, comprising:
The target area pixel collection of the target area is obtained according to the satellite remote sensing date, and to the satellite remote sensing
Data are pre-processed, and target water body index is obtained；
The crowdsourcing map datum is pre-processed according to the target water body index, obtains candidate water body sample set.
During above-mentioned realization, satellite remote sensing date and crowdsourcing map datum are combined, are conducive to promote atural object
It identifies the precision extracted, while also satellite remote sensing date and crowdsourcing map datum is pre-processed respectively, be further reduced big
Gas is conducive to promote generation water body charting results to the error message in the interference and crowdsourcing map datum of satellite remote-sensing image
Precision.
Further, the satellite remote sensing date is pre-processed, obtains target water body index, comprising:
Radiation calibration processing is carried out to the satellite remote sensing date, obtains radiance data；
Atmospheric correction processing is carried out to the radiance data, obtains Reflectivity for Growing Season；
Target water body index is calculated according to the Reflectivity for Growing Season.
It is fixed by radiate to satellite remote sensing date before calculating target water body index during above-mentioned realization
Mark and atmospheric correction processing, obtain Reflectivity for Growing Season, finally calculate target water body index further according to Reflectivity for Growing Season, can subtract
The interference of few Atmospheric Absorption and scattering, so that calculated target water body index is relatively reliable, error is small.
Further, the crowdsourcing map datum is pre-processed according to the target water body index, obtains candidate water
Body sample set, comprising:
Element information relevant to water body is extracted from the crowdsourcing map datum, the element information includes area pattern
Information and linear element information；
The buffer area of the linear element information is established in the crowdsourcing map datum；
The buffer area of the area pattern information and the linear element information is merged into processing, obtains merging number
According to；
Rasterizing processing is carried out to the merging data, obtains crowdsourcing map water body mask data；
Processing is filtered to the crowdsourcing map water body mask data according to the target water body index, obtains candidate water
Body sample set.
During above-mentioned realization, crowdsourcing map datum includes the water-related letters such as river abundant, streams, river levee
Breath, while also including other non-Water-Body Informations.By establishing the buffer area of linear element information, can obtain more comprehensive
Water-Body Information.On the other hand, by being filtered processing to crowdsourcing map water body mask data, crowdsourcing map water body can be removed
False water pixel in mask data obtains high-precision candidate water body sample set, is conducive to promote generation water body charting results
Precision.
Further, by the prediction neural network that constructs in advance and the candidate water body sample set to the target area
Pixel collection is handled, and the first probability that the target area pixel concentrates each pixel to be marked as water body is obtained, comprising:
The background sample data of default first quantity are chosen from the satellite remote sensing date, and to the background sample number
Processing is merged according to the candidate water body sample set, obtains merging sample set；
Training sample subset and verifying sample set are determined from the merging sample set；
Obtain the prediction neural network that constructs in advance, and by the training sample subset to the prediction neural network into
Row processing obtains processing data；
Parameter correction is carried out to the prediction neural network according to processing data and the verifying sample set, is trained
Prediction neural network afterwards；
The target area pixel collection is handled by the prediction neural network after the training, obtains the target
Region pixel concentrates each pixel to be marked as the first probability of water body.
During above-mentioned realization, calculates each target area pixel and each pixel is concentrated to be marked as the first general of water body
Rate constantly, first determines verifying sample set and training sample according to candidate water body sample set and the target area pixel collection of target area
Then this subset is trained processing to the prediction neural network constructed in advance by training sample subset again, after being trained
Prediction neural network, target area pixel is then calculated by the prediction neural network after training again and concentrates each pixel quilt
Labeled as the first probability of water body.
Further, the target area pixel according to first probability calculation concentrates each pixel to belong to the of water body
Two probability, comprising:
The water body sample in the verifying sample set is handled by the prediction neural network after the training, is obtained
The third probability of each water body sample into the verifying sample set；
According to the third probability, the mathematical expectation of probability of all water body samples in the verifying sample set is calculated；
According to the mathematical expectation of probability and first probability, calculates the target area pixel and each pixel is concentrated to belong to water
Second probability of body.
During above-mentioned realization, the prediction neural network after first passing through training calculates each water body in verifying sample set
The third probability of sample；Then according to third probability, the mathematical expectation of probability of all water body samples in verifying sample set is calculated, finally
Further according to mathematical expectation of probability and the first probability, the second probability that target area pixel concentrates each pixel to belong to water body is calculated.
Further, the surface water body charting results of the target area are generated according to second probability, comprising:
Segmentation threshold section is obtained, and positive sample-background scene confusion matrix is constructed according to the merging sample set；
Target Segmentation threshold value is determined according to the positive sample-background scene confusion matrix and the segmentation threshold section；
According to the Target Segmentation threshold value and second probability, the surface water body drawing knot of the target area is generated
Fruit.
During above-mentioned realization, before generating surface water body charting results, it is thus necessary to determine that Target Segmentation threshold value, according to
Positive sample-background scene confusion matrix and segmentation threshold section are capable of determining that most suitable segmentation threshold (i.e. Target Segmentation threshold
Value), and then generate optimal surface water body charting results.
Further, positive sample-background scene confusion matrix is constructed according to the merging sample set, comprising:
Segmentation sample set is determined from the merging sample set, and determination is multiple to be selected from the segmentation threshold section
Segmentation threshold；
Obscured according to the corresponding positive sample-background scene of each segmentation threshold to be selected of segmentation sample set building
Matrix.
During above-mentioned realization, after determining segmentation threshold to be selected, segmentation is determined from above-mentioned merging sample set
Then sample set constructs the corresponding positive sample-background scene of each segmentation threshold to be selected further according to segmentation sample set and obscures
Matrix.
Further, target point is determined according to the positive sample-background scene confusion matrix and the segmentation threshold section
Cut threshold value, comprising:
According to the corresponding positive sample-background scene confusion matrix of each segmentation threshold to be selected, the segmentation to be selected is calculated
The corresponding surface water body cartographic accuracy of threshold value；
The corresponding segmentation threshold to be selected of maximum surface water body cartographic accuracy is determined as Target Segmentation threshold value.
During above-mentioned realization, the corresponding segmentation threshold to be selected of maximum surface water body cartographic accuracy is determined as target
Segmentation threshold, as optimal segmenting threshold, corresponding surface water body cartographic accuracy is maximum, and then is obtained according to optimal segmenting threshold
It charts to optimal surface water body.
The embodiment of the present application second aspect provides a kind of surface water body mapping arrangements, comprising:
Data acquisition module, for obtaining the initial data for being used for the surface water body of target area and charting；
Preprocessing module obtains candidate water body sample set and the target for pre-processing to the initial data
The target area pixel collection in region；
First probability evaluation entity, for the prediction neural network and the candidate water body sample set pair by constructing in advance
The target area pixel collection is handled, and obtains the target area pixel and each pixel is concentrated to be marked as the first of water body
Probability；
Second probability evaluation entity concentrates each pixel for the target area pixel according to first probability calculation
Belong to the second probability of water body；
Drawing module, for generating the surface water body charting results of the target area according to second probability.
During above-mentioned realization, when generating surface water body charting results, data acquisition module first obtains target area
Initial data, and initial data is pre-processed to obtain the target area pixel of candidate water body sample set and target area
Collection, then the first probability evaluation entity determines verifying further according to candidate water body sample set and the target area pixel collection of target area
Sample set, and the first probability that target area pixel concentrates each pixel to be marked as water body is calculated, further, second is general
Rate computing module concentrates each pixel to belong to the second probability of water body according to the first probability calculation target area pixel, finally, making
Module generates surface water body charting results according to the second probability.
The embodiment of the present application third aspect provides a kind of computer equipment, including memory and processor, the storage
Device is for storing computer program, and the processor runs the computer program so that the computer equipment executes first party
The some or all of surface water body drafting method disclosed in face.
The embodiment of the present application fourth aspect provides a kind of computer readable storage medium, is stored with described in the third aspect
The computer program used in computer equipment.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application will make below to required in the embodiment of the present application
Attached drawing is briefly described, it should be understood that the following drawings illustrates only some embodiments of the application, therefore should not be seen
Work is the restriction to range, for those of ordinary skill in the art, without creative efforts, can be with
Other relevant attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of flow diagram for surface water body drafting method that the embodiment of the present application one provides；
Fig. 2 is a kind of flow diagram for surface water body drafting method that the embodiment of the present application two provides；
Fig. 3 is a kind of segmentation threshold-cartographic accuracy curve relation figure that the embodiment of the present application two provides；
Fig. 4 is a kind of structural schematic diagram for surface water body mapping arrangements that the embodiment of the present application three provides；
Fig. 5 is the structural schematic diagram for another surface water body mapping arrangements that the embodiment of the present application three provides.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application is described.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.Meanwhile the application's
In description, term " first ", " second " etc. are only used for distinguishing description, are not understood to indicate or imply relative importance.
Embodiment 1
Fig. 1 is please referred to, Fig. 1 is a kind of schematic process flow diagram of surface water body drafting method provided by the embodiments of the present application.
As shown in Figure 1, the surface water body drafting method includes:
S101, the initial data charted for the surface water body of target area is obtained, and initial data is pre-processed,
Obtain the target area pixel collection of candidate water body sample set and target area.
In the embodiment of the present application, when needing to obtain the surface water body charting results of target area, then what is got is used for
The initial data of surface water body drawing is the geodata of the target area.Wherein, which can be satellite remote sensing shadow
As data, map datum etc., this embodiment of the present application is not construed as limiting.
S102, by the prediction neural network that constructs in advance and candidate water body sample set to target area pixel collection at
Reason, obtains the first probability that target area pixel concentrates each pixel to be marked as water body.
In the embodiment of the present application, prediction neural network can be BP (back propagation, backpropagation) nerve net
Network, convolutional neural networks (Convolutional Neural Networks, CNN), RBF (Radial Basis Function,
Radial base) neural network etc., this embodiment of the present application is not construed as limiting.
S103, each pixel is concentrated to belong to the second probability of water body according to the first probability calculation target area pixel.
S104, the surface water body charting results that target area is generated according to the second probability.
In the embodiment of the present application, calculate target area pixel concentrate each pixel belong to water body the second probability it
Afterwards, Threshold segmentation processing is carried out to the second probability further according to segmentation threshold, the second probability is greater than to the pixel of Target Segmentation threshold value
It is determined as water body, the pixel that the second probability is less than Target Segmentation threshold value is determined as non-water body, and then obtain corresponding earth's surface
Water body charting results.
As it can be seen that implementing surface water body drafting method described in Fig. 1, the interference of barrier can be avoided, and then promote water
The precision of system figure.
Embodiment 2
Fig. 2 is please referred to, Fig. 2 is a kind of schematic process flow diagram of surface water body drafting method provided by the embodiments of the present application.
As shown in Fig. 2, the surface water body drafting method includes:
S201, the initial data charted for the surface water body of target area is obtained.
In the embodiment of the present application, initial data includes satellite remote sensing date and crowdsourcing map datum.Specifically, satellite remote sensing
Data can be sentry No. two image (Sentinel-2MSI) data, WorldView satellite data, QuickBird satellite numbers
According to, high score No.1 satellite data, No. three satellite datas of resource, Landsat satellite data etc., crowdsourcing map datum can be to open
Street map (OpenStreetMap, OSM) data etc., are not construed as limiting this embodiment of the present application.
In the embodiment of the present application, No. two images (Sentinel-2MSI) of sentry are a kind of novel multi-spectrum remote sensing images,
With 13 spectral bands, (wherein 4 wave bands are 10 meters of spatial resolutions, and 6 wave bands are 20 meters of spatial resolutions, 3 wave bands
For 60 meters of spatial resolutions), revisiting period is about 5 days.Compared to common intermediate resolution remotely-sensed data (Landsat
ETM/ETM+/OLI satellite image), which has higher spatial resolution and revisiting period, more suitable for surface water system
Figure and dynamic monitoring.
S202, according to satellite remote sensing date obtain target area target area pixel collection, and to satellite remote sensing date into
Row pretreatment, obtains target water body index.
As an alternative embodiment, pre-processing to satellite remote sensing date, target water body index is obtained, it can be with
The following steps are included:
Radiation calibration processing is carried out to satellite remote sensing date, obtains radiance data；
Atmospheric correction processing is carried out to radiance data, obtains Reflectivity for Growing Season；
Target water body index is calculated according to earth's surface albedometer.
In the above-described embodiment, target water body index can be automatic water body extracting index (Automate Water
Extraction Index, AWEI), normalization aqua index (Normalized Difference Water Index, NDWI),
Multiband aqua index (Multi-Band Water Index, MBWI), multispectral aqua index (Multi-spectral Water
Index, MuWI)), the one of which in all band aqua index (All Bans Water Index, ABWI) etc., to this this Shen
Please embodiment be not construed as limiting.
In the above-described embodiment, Reflectivity for Growing Season refers to the ratio between ground return amount of radiation and incident radiation amount, characterization ground
In face of the absorption and albedo of solar radiation.Reflectivity is bigger, and ground absorption solar radiation is fewer；Reflectivity is smaller, ground
It is more to absorb solar radiation.
In the above-described embodiment, radiation calibration is the spectral reflectivity or spectral radiance that user needs to calculate atural object
When, or need to different time, different sensors obtain image be compared when, the luminance grayscale values of image are converted to
The process of absolute radiance.
In the above-described embodiment, atmospheric correction refers to the global radiation brightness for the ground target that sensor finally measures not
It is the reflection of earth's surface real reflectance, wherein containing the amount of radiation error caused by Atmospheric Absorption, especially scattering process.Greatly
Gas correction is exactly to eliminate these radiation errors as caused by atmospheric effect, the process of the true surface reflectivity of inverting atural object.
It is further comprising the steps of after step S202:
S203, crowdsourcing map datum is pre-processed according to target water body index, obtains candidate water body sample set.
In the embodiment of the present application, OpenStreetMap (OSM) data are a kind of street maps that user generates, and are contained
Geography information abundant, such as water body, building.OSM data are provided by numerous volunteers, and lack necessary quality control.
Therefore, OSM data are pre-processed by step S203, falseness Water-Body Information that may be present in OSM data can be rejected,
Promote the accuracy of water body sample.
As an alternative embodiment, being pre-processed according to target water body index to crowdsourcing map datum, obtain
Candidate water body sample set, may comprise steps of:
Element information relevant to water body is extracted from crowdsourcing map datum, element information includes area pattern information and line
Property element information；
The buffer area of linear element information is established in crowdsourcing map datum；
The buffer area of area pattern information and linear element information is merged into processing, obtains merging data；
Rasterizing processing is carried out to merging data, obtains crowdsourcing map water body mask data；
Processing is filtered to crowdsourcing map water body mask data according to target water body index, obtains candidate water body sample
Collection.
In the above-described embodiment, when crowdsourcing map datum is OSM data, for OSM data, first from OSM data
Extracting relevant to water body element information, element information relevant to water body includes area pattern information and linear element information,
Wherein, area pattern information includes reservoir (reservoir) element information, water body (water) element information, river levee
(riverbank) one or more of element information etc.；Linear element information includes river (canal) element information, river
One or more of (river) element information, streams (stream) element information etc. are flowed, this embodiment of the present application is not made
It limits.
In the above-described embodiment, reservoir (reservoir) element can be extracted from the landuse figure layer of OSM data
Information；From the natural figure layer of OSM data, water body (water) element information, river levee (riverbank) element letter are extracted
Breath；From the waterway figure layer of OSM data, river (canal) element information, river (river) element information and small stream are chosen
Flow (stream) element information.
In the above-described embodiment, waterway figure in crowdsourcing map datum can be constructed according to pre-set buffer area size
The buffer area of layer element, wherein pre-set buffer area size includes pre-set buffer area radius.In actual use, pre-set buffer area
Radius can be 70 meters etc., be not construed as limiting to this embodiment of the present application.
In the above-described embodiment, when can carry out rasterizing processing to merging data according to default grid point value, wherein grid
Lattice value can be 10 meters etc., without limitation to this embodiment of the present application.
As further alternative embodiment, crowdsourcing map water body mask data was carried out according to target water body index
Filter processing, obtains candidate water body sample set, may comprise steps of:
Filtering Model is constructed according to target water body index；
Processing is filtered to crowdsourcing map water body mask data according to Filtering Model, obtains candidate water body sample set.
It in the above-described embodiment, is that can reject crowdsourcing map according to building Filtering Model with target water body index > 0
False water body pixel in water body mask data, and then candidate water body sample set is obtained, effectively avoid false water body pel data
Influence to surface water body drawing is conducive to the precision for promoting surface water body drawing.
In the embodiment of the present application, implements above-mentioned steps S202~step S203, initial data can be pre-processed, be obtained
To the target area pixel collection of candidate water body sample set and target area.
It is further comprising the steps of after step S203:
S204, the background sample data that default first quantity is chosen from satellite remote sensing date, and to background sample data
Processing is merged with candidate water body sample set, obtains merging sample set, and determine training sample subset in sample set from merging
With verifying sample set.
In the embodiment of the present application, training sample subset and verifying can be determined from merging sample set according to default division rule
Sample set.Wherein, default division rule includes according to preset ratio random division, and preset ratio includes in training sample subset
Sample size accounts for the total sample number amount in sample set that mergesVerifying sample set accounts for the total sample number amount in sample set that merges
In the above-described embodiment, from merging in sample set, determining training sample subset and the mode of verifying sample set can
Think and randomly select, do not need the selection for being manually trained sample, reduces drain on manpower and material resources, be also beneficial to lift scheme
Trained efficiency.
The prediction neural network that S205, acquisition construct in advance, and prediction neural network is carried out by training sample subset
Processing obtains processing data, and carries out parameter correction to prediction neural network according to processing data and verifying sample set, obtains
Prediction neural network after to training.
In the above-described embodiment, processing is merged to background sample data and candidate water body sample set, is merged
Sample set, specifically, first randomly selecting the default preliminary pixel for choosing quantity (such as accounting 5%) from candidate water body sample set
Quantity (such as accounting is chosen as preliminary water body sample, and according to default preliminary) back is randomly selected from satellite remote sensing date
Scape sample data.Further, background sample data and preliminary water body sample are merged to obtain and merges sample set, according still further to
Default sample chooses quantity (for example, the quantity of pixel and target area pixel concentrate the quantity of pixel in training sample subset
Ratio is 3:1) training sample subset and verifying sample set are determined from merging sample set, wherein and training sample subset is used for
The model training of prediction neural network, model parameter of the verifying sample set for prediction neural network are preferred.
S206, target area pixel collection is handled by the prediction neural network after training, obtains target area picture
The each pixel of member concentration is marked as the first probability of water body.
In the embodiment of the present application, implement above-mentioned steps S204~step S206, it can be neural by the prediction constructed in advance
Network and candidate water body sample set handle target area pixel collection, obtain target area pixel and each pixel is concentrated to be marked
It is denoted as the first probability of water body.
S207, the water body sample in verifying sample set is handled by the prediction neural network after training, is obtained
The third probability of each water body sample in sample set is verified, and according to third probability, calculates all water in verifying sample set
The mathematical expectation of probability of body sample.
S208, according to mathematical expectation of probability and the first probability, calculate target area pixel and each pixel concentrated to belong to the of water body
Two probability.
In the embodiment of the present application, for each pixel that target area pixel is concentrated, calculates it and belong to the second general of water body
The calculation formula of rate is as follows:
Wherein, PWPixel is concentrated to belong to the second probability of water body for target area pixel, c is to own in verifying sample set
The mathematical expectation of probability of water body sample, PobsThe first probability of water body is marked as target area pixel concentration pixel.
In the embodiment of the present application, implement above-mentioned steps S205~step S206, it can be according to the first probability calculation target area
Domain pixel concentrates each pixel to belong to the second probability of water body.
S209, the surface water body charting results that target area is generated according to the second probability.
As an alternative embodiment, the surface water body charting results of target area are generated according to the second probability, it can
With the following steps are included:
Segmentation threshold section is obtained, and constructs positive sample-background scene confusion matrix according to sample set is merged；
Target Segmentation threshold value is determined according to positive sample-background scene confusion matrix and segmentation threshold section；
According to Target Segmentation threshold value and the second probability, the surface water body charting results of target area are generated.
In the above-described embodiment, segmentation threshold section can be to preset, and be specifically as follows [- 1,1] etc., to this
Application embodiment is not construed as limiting.
As further alternative embodiment, positive sample-background scene confusion matrix is constructed according to sample set is merged, it can
With the following steps are included:
Segmentation sample set is determined in sample set from merging, and multiple segmentation thresholds to be selected are determined from segmentation threshold section
Value；
Corresponding positive sample-background scene the confusion matrix of each segmentation threshold to be selected is constructed according to segmentation sample set.
In the above-described embodiment, for segmentation threshold to be selected for one, corresponding positive sample-background scene is obscured
Matrix is expressed as follows:
Because segmentation sample set be obtained by merging sample set, and merge sample set be according to background sample data and
What candidate water body sample set obtained, therefore, s=1 indicates the pixel for belonging to candidate water body sample set in segmentation pixel subset, s=0
Indicate that the pixel for belonging to background sample data in segmentation sample set, the prediction neural network after y '=1 indicates trained carry out
The result of prediction is water body, and the result predicted of prediction neural network after y '=0 indicates trained is non-water body, TP ',
FP ', FN ', TN ' indicate positive sample-background scene confusion matrix element.
In the above-described embodiment, segmentation sample of default second quantity can be randomly selected from merging sample set
Collection, wherein default second quantity can be pixel sum in merging sample setDeng not limiting this embodiment of the present application
It is fixed.
It is true according to positive sample-background scene confusion matrix and segmentation threshold section as further alternative embodiment
Set the goal segmentation threshold, may comprise steps of:
According to the corresponding positive sample-background scene confusion matrix of each segmentation threshold to be selected, the segmentation threshold to be selected is calculated
Corresponding surface water body cartographic accuracy；
The corresponding segmentation threshold to be selected of maximum surface water body cartographic accuracy is determined as Target Segmentation threshold value.
In the above-described embodiment, for each segmentation threshold to be selected, its corresponding surface water body cartographic accuracy is calculated
Formula is as follows:
Wherein, FmIt FP ', FN ', is TP ', this to be selected point for the corresponding surface water body cartographic accuracy of segmentation threshold to be selected
Cut the corresponding positive sample of threshold value-background scene confusion matrix element.
It is a kind of segmentation threshold-cartographic accuracy curve relation figure provided in this embodiment also referring to Fig. 3, Fig. 3.Such as figure
Shown in 3, in the curve relation figure, abscissa is each segmentation threshold to be selected in segmentation threshold section, and ordinate is surface water
Body cartographic accuracy Fm(i.e. F-score).As seen from Figure 3, with the increase of segmentation threshold to be selected, FmFirst stable rear increase is presented most
The variation tendency reduced afterwards, when segmentation threshold is 0.32, water body cartographic accuracy (Fm) reach maximum value, i.e. Max (F-score)
=1.7825, hence, it can be determined that Target Segmentation threshold value is 0.32 out.
In the above-described embodiment, after determining Target Segmentation threshold value, the second probability is greater than Target Segmentation threshold value
Pixel be determined as water body pixel, the pixel that the second probability is less than Target Segmentation threshold value is determined as non-aqueous body image member, so
To corresponding surface water body charting results.
As it can be seen that implementing surface water body drafting method described in Fig. 2, the interference of barrier can be avoided, and then promote water
The precision of system figure.
Embodiment 3
Fig. 4 is please referred to, Fig. 4 is a kind of structural schematic block diagram of surface water body mapping arrangements provided by the embodiments of the present application.
As shown in figure 4, the surface water body mapping arrangements include:
Data acquisition module 310, for obtaining the initial data for being used for the surface water body of target area and charting.
Preprocessing module 320 obtains candidate water body sample set and target area for pre-processing to initial data
Target area pixel collection.
First probability evaluation entity 330, for the prediction neural network and candidate water body sample set pair by constructing in advance
Target area pixel collection is handled, and the first probability that target area pixel concentrates each pixel to be marked as water body is obtained.
Second probability evaluation entity 340, for concentrating each pixel to belong to according to the first probability calculation target area pixel
Second probability of water body.
Drawing module 350, for generating the surface water body charting results of target area according to the second probability.
In the embodiment of the present application, initial data includes satellite remote sensing date and crowdsourcing map datum etc., to this application reality
Example is applied to be not construed as limiting.
Fig. 5 is please referred to, Fig. 5 is the structural representation frame of another surface water body mapping arrangements provided by the embodiments of the present application
Figure.Wherein, surface water body mapping arrangements shown in fig. 5 are that surface water body mapping arrangements as shown in Figure 4 optimize,
As shown in figure 5, preprocessing module 320, comprising:
Pixel acquisition submodule 321, for obtaining the target area pixel collection of target area according to satellite remote sensing date.
First pretreatment submodule 322 obtains target water body index for pre-processing to satellite remote sensing date.
Second pretreatment submodule 323 is obtained for being pre-processed according to target water body index to crowdsourcing map datum
Candidate water body sample set.
Submodule 321 is pre-processed as into a kind of optional embodiment, first, comprising:
Radiation calibration unit obtains radiance data for carrying out radiation calibration processing to satellite remote sensing date.
Atmospheric correction unit obtains Reflectivity for Growing Season for carrying out atmospheric correction processing to radiance data.
Computing unit, for calculating target water body index according to earth's surface albedometer.
As further alternative embodiment, the second pretreatment submodule 322, comprising:
Information extraction unit, for extracting element information relevant to water body, element information packet from crowdsourcing map datum
Include area pattern information and linear element information.
Unit is established in buffer area, for establishing the buffer area of linear element information in crowdsourcing map datum.
Combining unit is closed for the buffer area of area pattern information and linear element information to be merged processing
And data.
Rasterizing unit obtains crowdsourcing map water body mask data for carrying out rasterizing processing to merging data.
Filter element is obtained for being filtered processing to crowdsourcing map water body mask data according to target water body index
Candidate water body sample set.
As an alternative embodiment, the first probability evaluation entity 330, comprising:
Sample acquisition submodule 331, for choosing the background sample data of default first quantity from satellite remote sensing date,
And processing is merged to background sample data and candidate water body sample set, it obtains merging sample set；And from merge sample set
Middle determining training sample subset and verifying sample set；
Training submodule 332, for obtaining the prediction neural network constructed in advance, and by training sample subset to prediction
Neural network is handled, and processing data are obtained；And according to processing data and verifying sample set to prediction neural network into
Row parameter correction, the prediction neural network after being trained；
Submodule 333 is handled, for being handled by the prediction neural network after training target area pixel collection, is obtained
Each pixel is concentrated to be marked as the first probability of water body to target area pixel.
As an alternative embodiment, the second probability evaluation entity 340, comprising:
First computational submodule 341, for passing through the prediction neural network after training to the water body in verifying sample set
Sample is handled, and the third probability of each water body sample in sample set is verified；And according to third probability, calculating is tested
Demonstrate,prove the mathematical expectation of probability of all water body samples in sample set.
Second computational submodule 342, for calculating target area pixel and concentrating each according to mathematical expectation of probability and the first probability
Pixel belongs to the second probability of water body.
As an alternative embodiment, drawing module 350, comprising:
Section acquisition submodule 351, for obtaining segmentation threshold section.
Matrix constructs submodule 352, for constructing positive sample-background scene confusion matrix according to merging sample set.
Threshold value determines submodule 353, for determining mesh according to positive sample-background scene confusion matrix and segmentation threshold section
Mark segmentation threshold.
Drawing generates submodule 354, for generating the surface water of target area according to Target Segmentation threshold value and the second probability
Body charting results.
As further alternative embodiment, matrix constructs submodule 352, comprising:
Subset determing unit, for determining segmentation sample set from merging sample set.
Threshold value determination unit to be selected, for determining multiple segmentation thresholds to be selected from segmentation threshold section.
Matrix construction unit, for constructing the corresponding positive sample-back of each segmentation threshold to be selected according to segmentation sample set
Scape scene confusion matrix.
As further alternative embodiment, threshold value determines submodule 353, comprising:
Accuracy computation unit, for according to the corresponding positive sample-background scene confusion matrix of each segmentation threshold to be selected, meter
Calculate the corresponding surface water body cartographic accuracy of the segmentation threshold to be selected.
Target determination unit, for the corresponding segmentation threshold to be selected of maximum surface water body cartographic accuracy to be determined as target
Segmentation threshold.
As it can be seen that implementing surface water body mapping arrangements described in the present embodiment, the interference of barrier, Jin Erti can be avoided
Rise the precision of water body drawing.
In addition, the present invention also provides a kind of computer equipments.The computer equipment includes memory and processor, storage
Device can be used for storing computer program, and processor is by operation computer program, so that the computer equipment be made to execute above-mentioned side
The function of method or the modules in above-mentioned surface water body mapping arrangements.
Memory may include storing program area and storage data area, wherein storing program area can storage program area, at least
Application program needed for one function (such as sound-playing function, image player function etc.) etc.；Storage data area can store root
Created data (such as audio data, phone directory etc.) etc. are used according to mobile terminal.In addition, memory may include high speed
Random access memory, can also include nonvolatile memory, a for example, at least disk memory, flush memory device or
Other volatile solid-state parts.
The present embodiment additionally provides a kind of computer storage medium, for storing calculating used in above-mentioned computer equipment
Machine program.
In several embodiments provided herein, it should be understood that disclosed device and method can also pass through
Other modes are realized.The apparatus embodiments described above are merely exemplary, for example, flow chart and block diagram in attached drawing
Show the device of multiple embodiments according to the application, the architectural framework in the cards of method and computer program product,
Function and operation.In this regard, each box in flowchart or block diagram can represent the one of a module, section or code
Part, a part of the module, section or code, which includes that one or more is for implementing the specified logical function, to be held
Row instruction.It should also be noted that function marked in the box can also be to be different from some implementations as replacement
The sequence marked in attached drawing occurs.For example, two continuous boxes can actually be basically executed in parallel, they are sometimes
It can execute in the opposite order, this depends on the function involved.It is also noted that every in block diagram and or flow chart
The combination of box in a box and block diagram and or flow chart can use the dedicated base for executing defined function or movement
It realizes, or can realize using a combination of dedicated hardware and computer instructions in the system of hardware.
In addition, each functional module in each embodiment of the application can integrate one independent portion of formation together
Point, it is also possible to modules individualism, an independent part can also be integrated to form with two or more modules.
It, can be with if the function is realized and when sold or used as an independent product in the form of software function module
It is stored in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) execute each embodiment the method for the application all or part of the steps.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
The above description is only an example of the present application, the protection scope being not intended to limit this application, for ability
For the technical staff in domain, various changes and changes are possible in this application.Within the spirit and principles of this application, made
Any modification, equivalent substitution, improvement and etc. should be included within the scope of protection of this application.It should also be noted that similar label and
Letter indicates similar terms in following attached drawing, therefore, once it is defined in a certain Xiang Yi attached drawing, then in subsequent attached drawing
In do not need that it is further defined and explained.
The above, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, it is any
Those familiar with the art within the technical scope of the present application, can easily think of the change or the replacement, and should all contain
Lid is within the scope of protection of this application.Therefore, the protection scope of the application shall be subject to the protection scope of the claim.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
Claims (10)
1. a kind of surface water body drafting method characterized by comprising
The initial data charted for the surface water body of target area is obtained, and the initial data is pre-processed, is obtained
The target area pixel collection of candidate water body sample set and the target area；
By the prediction neural network that constructs in advance and the candidate water body sample set to the target area pixel collection at
Reason, obtains the first probability that the target area pixel concentrates each pixel to be marked as water body；
Each pixel is concentrated to belong to the second probability of water body according to target area pixel described in first probability calculation；
The surface water body charting results of the target area are generated according to second probability.
2. surface water body drafting method according to claim 1, which is characterized in that the initial data includes the target
The crowdsourcing map datum of the satellite remote sensing date in region and the target area；
The initial data is pre-processed, the target area pixel of candidate water body sample set and the target area is obtained
Collection, comprising:
The target area pixel collection of the target area is obtained according to the satellite remote sensing date, and to the satellite remote sensing date
It is pre-processed, obtains target water body index；
The crowdsourcing map datum is pre-processed according to the target water body index, obtains candidate water body sample set.
3. surface water body drafting method according to claim 2, which is characterized in that carried out to the satellite remote sensing date pre-
Processing, obtains target water body index, comprising:
Radiation calibration processing is carried out to the satellite remote sensing date, obtains radiance data；
Atmospheric correction processing is carried out to the radiance data, obtains Reflectivity for Growing Season；
Target water body index is calculated according to the Reflectivity for Growing Season.
4. surface water body drafting method according to claim 3, which is characterized in that according to the target water body index to institute
It states crowdsourcing map datum to be pre-processed, obtains candidate water body sample set, comprising:
Element information relevant to water body is extracted from the crowdsourcing map datum, the element information includes area pattern information
With linear element information；
The buffer area of the linear element information is established in the crowdsourcing map datum；
The buffer area of the area pattern information and the linear element information is merged into processing, obtains merging data；
Rasterizing processing is carried out to the merging data, obtains crowdsourcing map water body mask data；
Processing is filtered to the crowdsourcing map water body mask data according to the target water body index, obtains candidate water body sample
This collection.
5. surface water body drafting method according to claim 2, which is characterized in that pass through the prediction nerve net constructed in advance
Network and the candidate water body sample set handle the target area pixel collection, obtain the target area pixel and concentrate often
A pixel is marked as the first probability of water body, comprising:
Choose the background sample data of default first quantity from the satellite remote sensing date, and to the background sample data and
Candidate's water body sample set merges processing, obtains merging sample set；
Training sample subset and verifying sample set are determined from the merging sample set；
Obtain the prediction neural network that constructs in advance, and by the training sample subset to the prediction neural network at
Reason obtains processing data；
Parameter correction is carried out to the prediction neural network according to processing data and the verifying sample set, after being trained
Prediction neural network；
The target area pixel collection is handled by the prediction neural network after the training, obtains the target area
Pixel concentrates each pixel to be marked as the first probability of water body.
6. surface water body drafting method according to claim 5, which is characterized in that according to first probability calculation
Target area pixel concentrates each pixel to belong to the second probability of water body, comprising:
The water body sample in the verifying sample set is handled by the prediction neural network after the training, obtains institute
State the third probability of each water body sample in verifying sample set；
According to the third probability, the mathematical expectation of probability of all water body samples in the verifying sample set is calculated；
According to the mathematical expectation of probability and first probability, calculates the target area pixel and each pixel is concentrated to belong to water body
Second probability.
7. surface water body drafting method according to claim 5, which is characterized in that according to second probability generation
The surface water body charting results of target area, comprising:
Segmentation threshold section is obtained, and positive sample-background scene confusion matrix is constructed according to the merging sample set；
Target Segmentation threshold value is determined according to the positive sample-background scene confusion matrix and the segmentation threshold section；
According to the Target Segmentation threshold value and second probability, the surface water body charting results of the target area are generated.
8. surface water body drafting method according to claim 7, which is characterized in that just according to merging sample set building
Sample-background scene confusion matrix, comprising:
Segmentation sample set is determined from the merging sample set, and multiple segmentations to be selected are determined from the segmentation threshold section
Threshold value；
Square is obscured according to the corresponding positive sample-background scene of each segmentation threshold to be selected of segmentation sample set building
Battle array.
9. surface water body drafting method according to claim 8, which is characterized in that according to the positive sample-background scene
Confusion matrix and the segmentation threshold section determine Target Segmentation threshold value, comprising:
According to the corresponding positive sample-background scene confusion matrix of each segmentation threshold to be selected, the segmentation threshold to be selected is calculated
Corresponding surface water body cartographic accuracy；
The corresponding segmentation threshold to be selected of maximum surface water body cartographic accuracy is determined as Target Segmentation threshold value.
10. a kind of surface water body mapping arrangements characterized by comprising
Data acquisition module, for obtaining the initial data for being used for the surface water body of target area and charting；
Preprocessing module obtains candidate water body sample set and the target area for pre-processing to the initial data
Target area pixel collection；
First probability evaluation entity, for passing through the prediction neural network constructed in advance and the candidate water body sample set to described
Target area pixel collection is handled, and obtains the target area pixel and each pixel is concentrated to be marked as the first general of water body
Rate；
Second probability evaluation entity concentrates each pixel to belong to for the target area pixel according to first probability calculation
Second probability of water body；
Drawing module, for generating the surface water body charting results of the target area according to second probability.
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 true CN110517575A (en) | 2019-11-29 |
| CN110517575B 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) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652882A (en) * | 2020-07-07 | 2020-09-11 | 中国水利水电科学研究院 | Large-scale surface water product drawing precision evaluation method |
| CN111913965A (en) * | 2020-08-03 | 2020-11-10 | 北京吉威空间信息股份有限公司 | Method for analyzing spatial big data buffer area |
| CN112950780A (en) * | 2021-03-12 | 2021-06-11 | 北京理工大学 | Intelligent network map generation method and system based on remote sensing image |
| CN113327214A (en) * | 2021-05-19 | 2021-08-31 | 中国科学院地理科学与资源研究所 | Continuous time series water body remote sensing mapping method |
| 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 |
Citations (13)
| 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 |
| CN103456122A (en) * | 2013-08-26 | 2013-12-18 | 中国科学技术大学 | Forest fire smoke recognizing method and device |
| CN106023133A (en) * | 2016-04-26 | 2016-10-12 | 武汉大学 | High resolution remote sensing image water body extraction method based on multi-feature combined treatment |
| CN107271367A (en) * | 2017-05-05 | 2017-10-20 | 北京师范大学 | A kind of identifying water boy method and device |
| 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 |
| CN109063754A (en) * | 2018-07-18 | 2018-12-21 | 武汉大学 | A kind of remote sensing image multiple features combining classification method based on OpenStreetMap |
| CN109211793A (en) * | 2018-09-12 | 2019-01-15 | 中国科学技术大学 | In conjunction with the excessive fire slash recognition methods of the fire of 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 |
| 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 |
| CN109977801A (en) * | 2019-03-08 | 2019-07-05 | 中国水利水电科学研究院 | A kind of quick Dynamic Extraction method and system of region water body of optical joint and radar |
| 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 |
-
2019
- 2019-08-21 CN CN201910771960.0A patent/CN110517575B/en active Active
Patent Citations (13)
| 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 |
| CN103456122A (en) * | 2013-08-26 | 2013-12-18 | 中国科学技术大学 | Forest fire smoke recognizing method and device |
| CN106023133A (en) * | 2016-04-26 | 2016-10-12 | 武汉大学 | High resolution remote sensing image water body extraction method based on multi-feature combined treatment |
| CN107271367A (en) * | 2017-05-05 | 2017-10-20 | 北京师范大学 | A kind of identifying water boy method and device |
| 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 |
| CN109063754A (en) * | 2018-07-18 | 2018-12-21 | 武汉大学 | A kind of remote sensing image multiple features combining classification method based on OpenStreetMap |
| CN109211793A (en) * | 2018-09-12 | 2019-01-15 | 中国科学技术大学 | In conjunction with the excessive fire slash recognition methods of the fire of 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 |
| 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 |
| CN109977801A (en) * | 2019-03-08 | 2019-07-05 | 中国水利水电科学研究院 | A kind of quick Dynamic Extraction method and system of region water body of optical joint and radar |
| 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 |
Non-Patent Citations (1)
| Title |
|---|
| 刘文涛,李世华等: "基于全卷积神经网络的建筑物屋顶自动提取", 《地球信息科学学报》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652882A (en) * | 2020-07-07 | 2020-09-11 | 中国水利水电科学研究院 | Large-scale surface water product drawing precision evaluation method |
| 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 |
| CN111913965A (en) * | 2020-08-03 | 2020-11-10 | 北京吉威空间信息股份有限公司 | Method for analyzing spatial big data buffer area |
| 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 |
| CN112950780A (en) * | 2021-03-12 | 2021-06-11 | 北京理工大学 | Intelligent network map generation method and system based on remote sensing image |
| 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 |
| CN113327214A (en) * | 2021-05-19 | 2021-08-31 | 中国科学院地理科学与资源研究所 | Continuous time series water body remote sensing mapping method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110517575B (en) | 2021-03-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110517575A (en) | A kind of surface water body drafting method and device | |
| Tian et al. | Aboveground mangrove biomass estimation in Beibu Gulf using machine learning and UAV remote sensing | |
| Burns et al. | Assessing the impact of acute disturbances on the structure and composition of a coral community using innovative 3D reconstruction techniques | |
| CN107063197B (en) | Reservoir characteristic curve extraction method based on spatial information technology | |
| Esbah et al. | Analyzing landscape changes in the Bafa Lake Nature Park of Turkey using remote sensing and landscape structure metrics | |
| CN109840553B (en) | Extraction method and system of cultivated land crop type, storage medium and electronic equipment | |
| CN106407927A (en) | Salient visual method based on polarization imaging and applicable to underwater target detection | |
| Steinfeld et al. | Semi‐automated GIS techniques for detecting floodplain earthworks | |
| Xu et al. | Monitoring coastal reclamation changes across Jiangsu Province during 1984–2019 using landsat data | |
| Ai et al. | The impact of rapid urban expansion on coastal mangroves: a case study in Guangdong Province, China | |
| CN104121850A (en) | Canopy density measurement method and device | |
| Li et al. | Extracting check dam areas from high‐resolution imagery based on the integration of object‐based image analysis and deep learning | |
| CN112819066A (en) | Res-UNet single tree species classification technology | |
| 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 | |
| CN113220810B (en) | Multi-source species distribution data processing method and device | |
| Hu et al. | Multi-information PointNet++ fusion method for DEM construction from airborne LiDAR data | |
| CN117197174A (en) | Water body information extraction method, device, equipment and medium based on remote sensing data | |
| Martinez-Sanchez et al. | Skyline variations allow estimating distance to trees on landscape photos using semantic segmentation | |
| CN114386828A (en) | Sea island landscape three-dimensional index construction method | |
| CN112966657A (en) | Remote sensing automatic classification method for large-scale water body coverage | |
| CN113392377A (en) | Swamp wetland remote sensing extraction method and system based on soil humidity | |
| Wang et al. | Combining SAR images with land cover products for rapid urban flood mapping | |
| Sun et al. | Check dam extraction from remote sensing images using deep learning and geospatial analysis: A case study in the Yanhe River Basin of the Loess Plateau, China | |
| Shih et al. | Waterline digital elevation model development to quantify inundation duration and coastal protection of tidal wetlands | |
| Wang et al. | More appropriate DenseNetBL classifier for small sample tree species classification using UAV-based RGB imagery |
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 |