CN109522788A - City scope extracting method, device and electronic equipment based on random forest sorting algorithm - Google Patents
City scope extracting method, device and electronic equipment based on random forest sorting algorithm Download PDFInfo
- Publication number
- CN109522788A CN109522788A CN201811159030.1A CN201811159030A CN109522788A CN 109522788 A CN109522788 A CN 109522788A CN 201811159030 A CN201811159030 A CN 201811159030A CN 109522788 A CN109522788 A CN 109522788A
- Authority
- CN
- China
- Prior art keywords
- city
- pixel
- sample
- night lights
- vegetation index
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to a kind of city scope extracting method, device and electronic equipments based on random forest sorting algorithm.City scope extracting method based on random forest sorting algorithm of the invention includes the following steps: and according to the night lights city index of night lights remotely-sensed data and vegetation index data acquisition sample areas；Training sample is chosen from the night lights remote sensing image and vegetation index image of sample areas, and according to the night lights remotely-sensed data of the training sample and sample areas elected, vegetation index data and night lights city index, establishes and train optimal stochastic forest algorithm model；The night lights remote sensing image in region to be identified and vegetation index image are inputted into optimal stochastic forest algorithm model, judge whether the region is city scope.City scope extracting method based on random forest sorting algorithm of the invention can identify urban area and non-urban area according to night lights remote sensing image and vegetation index.
Description
Technical field
The present invention relates to Study of Urban information technology fields, are based on random forest sorting algorithm more particularly to one kind
City scope extracting method, device and electronic equipment.
Background technique
City scope, which is extracted, at present mainly uses airborne and spaceborne RS technology, compared to the side of traditional ground actual measurement
Method, remote sensing technology has the features such as workload is small, at low cost, the period is short, high-efficient, and can satisfy current research urbanization
Demand.The mode that traditional remote sensing technology extracts city scope is usually utilize high-resolution in 1 year multispectral distant
Sense image extracts, and extraction process mainly includes the geometric correction of image, atmospheric correction, inlays, cuts, classify etc. and walk
Suddenly.Be affected since multispectral remote sensing image is obtained by weather, obtained image carry out geometric correction,
Since the difference of imaging time causes its operation relatively difficult when the work such as inlaying.
Summary of the invention
Based on this, the object of the present invention is to provide a kind of city scope extraction sides based on random forest sorting algorithm
Method can identify urban area and non-urban area according to night lights remote sensing image and vegetation index.
The present invention is achieved by the following scheme:
A kind of city scope extracting method based on random forest sorting algorithm, includes the following steps:
Obtain the night lights remote sensing image and vegetation index image of sample areas；
The night lights remotely-sensed data and vegetation index data of sample areas are obtained, and according to the night lights remote sensing number
According to the night lights city index with vegetation index data acquisition sample areas；
According to the night lights remotely-sensed data and vegetation index data of sample areas, the city in the sample areas is obtained
Pixel and non-city pixel；
Choose training sample from the city pixel and non-city pixel in the sample areas, and according to being elected
The night lights remotely-sensed data of training sample and sample areas, vegetation index data and night lights city index are established simultaneously
Training optimal stochastic forest algorithm model, wherein the night lights remote sensing number of the training sample elected and sample areas
According to, the input sample of vegetation index data and night lights city index as the trained optimal stochastic forest algorithm model,
The output sample of the city pixel of the sample areas and non-city pixel as the trained optimal stochastic forest algorithm model；
Obtain the night lights remote sensing image and vegetation index image in region to be identified；
The night lights remote sensing image in the region to be identified and vegetation index image are inputted into optimal stochastic forest algorithm
Model judges whether the region is city scope.
City scope extracting method of the present invention based on random forest sorting algorithm, according to the night of sample areas
Light remote sensing image and vegetation index image and data train optimal random forests algorithm model, can be with by the model
Judge whether the region to be identified is city model according to the night lights remote sensing image in region to be identified and vegetation index image
It encloses, the defect of Satellite Remote Sensing can be made up, improve city scope data.
In one embodiment, according to the night lights remotely-sensed data of sample areas and vegetation index data, described in acquisition
City pixel and non-city pixel in sample areas, include the following steps:
If the night lights remotely-sensed data in the sample areas in a certain pixel is greater than the first given threshold, and the picture
Vegetation index data in member are less than the second given threshold, then the pixel is city pixel, if the vegetation in a certain pixel refers to
Number data are greater than the second given threshold, then the pixel is non-city pixel.
In one embodiment, training sample is chosen from the city pixel and non-city pixel in the sample areas,
Include the following steps:
Based on administrative division boundary stratified sampling training sample.
In one embodiment, it is based on administrative division boundary stratified sampling training sample, further includes following steps:
Generating a value is the random number between 0~1；
Obtain the ratio value chosen between sample size and city pixel and non-city pixel total quantity in each layer；
If the random number is less than the ratio value, choose sample as training sample this.
In one embodiment, according to the night lights remotely-sensed data of sample areas and vegetation index data, described in acquisition
Further include following steps before city pixel and non-city pixel in sample areas:
According to water body distributed data, the water body pixel in the night lights remote sensing image is removed.
Further, the present invention also provides a kind of city scope extraction elements based on random forest sorting algorithm, comprising:
First data acquisition module, for obtaining the night lights remote sensing image and vegetation index image of sample areas；
Second data acquisition module, for obtaining the night lights remotely-sensed data and vegetation index data of sample areas, and
According to the night lights city index of the night lights remotely-sensed data and vegetation index data acquisition sample areas；
Sample process module obtains institute for the night lights remotely-sensed data and vegetation index data according to sample areas
State the city pixel and non-city pixel in sample areas；
Random forest training module, for choosing training from the city pixel and non-city pixel in the sample areas
Sample, and according to the night lights remotely-sensed data, vegetation index data and night of the training sample and sample areas elected
Between light city index, establish simultaneously training optimal stochastic forest algorithm model, wherein the training sample and sample elected
Night lights remotely-sensed data, vegetation index data and the night lights city index in region are as the trained optimal stochastic forest
The input sample of algorithm model, the city pixel of the sample areas and non-city pixel are as the trained optimal stochastic forest
The output sample of algorithm model；
Third data acquisition module, for obtaining the night lights remote sensing image and vegetation index image in region to be identified；
City scope judgment module, for by the night lights remote sensing image and vegetation index image in the region to be identified
Optimal stochastic forest algorithm model is inputted, judges whether the region is city scope.
City scope extraction element of the present invention based on random forest sorting algorithm, according to the night of sample areas
Light remote sensing image and vegetation index image and data train optimal random forests algorithm model, can be with by the model
Judge whether the region to be identified is city model according to the night lights remote sensing image in region to be identified and vegetation index image
It encloses, the defect of Satellite Remote Sensing can be made up, improve city scope data.
In one embodiment, the random forest training module, comprising:
Whether pixel discrimination unit, the night lights remotely-sensed data for judging in the sample areas are greater than the first setting
Threshold value, and whether the vegetation index data in the pixel less than the second given threshold, if it is, the pixel is city pixel,
If the vegetation index data in a certain pixel are greater than the second given threshold, which is non-city pixel.
In one embodiment, further includes:
Water body remove module, for removing the water body picture in the night lights remote sensing image according to water body distributed data
Member.
Further, the present invention also provides a kind of computer-readable medium, it is stored thereon with computer program, the computer
Such as city scope extracting method of the above-mentioned any one based on random forest sorting algorithm is realized when program is executed by processor.
Further, the present invention also provides a kind of electronic equipment, including memory, processor and it is stored in the storage
Device and the computer program that can be executed by the processor when processor executes the computer program, are realized as above-mentioned
City scope extracting method of any one based on random forest sorting algorithm.
In order to better understand and implement, the invention will now be described in detail with reference to the accompanying drawings.
Detailed description of the invention
Fig. 1 is the city scope extracting method flow chart based on random forest sorting algorithm in a kind of embodiment；
Fig. 2 is that training sample flow chart is chosen in a kind of embodiment；
Fig. 3 is different administrative region, and training sample flow chart is chosen in sampling；
Fig. 4 is the city scope extracting method flow chart based on random forest sorting algorithm in a kind of embodiment；
Fig. 5 is the city scope extraction element structural schematic diagram based on random forest sorting algorithm in a kind of embodiment；
Fig. 6 is electronic devices structure schematic diagram in a kind of embodiment.
Specific embodiment
Referring to Fig. 1, in one embodiment, the city scope extracting method based on random forest sorting algorithm includes such as
Lower step:
Step S10: the night lights remote sensing image and vegetation index image of sample areas are obtained.
Step S20: the night lights remotely-sensed data and vegetation index data of sample areas are obtained, and according to the night lamp
The night lights city index of light remotely-sensed data and vegetation index data acquisition sample areas.
The night lights remote sensing image, for the land that in the case where night is cloudless, the remote sensor on satellite is obtained
The visible light source image on ground or water body, the night lights remotely-sensed data (NTL) are right for the night lights remote sensing image
The data answered, for in night lights remote sensing image, the numerical value of the lamplight brightness in each resolution ratio, the night lights
Remotely-sensed data value is usually 0-63.The vegetation index (NDVI) is the spectral characteristic according to vegetation, by satellite visible and closely
Infrared band is combined, the vegetative coverage index of formation, qualitative and quantitative assessment vegetative coverage and its growth vigor.Vegetation refers to
Several values is usually -1 to 1, and in snow and ice cover, water body and desert areas, vegetation index is typically less than zero constant, the plant
By index image, figure is formed by for tables of data corresponding to the vegetation index data and the night lights remote sensing image
Picture.According to night lights city index (the Vegetation Adjusted NTL Urban Index, VANUI)
Night lights remotely-sensed data after the vegetation index data point reuse, in the present embodiment, night lights city index
Calculation formula are as follows: VANUI=NTL* (1-NDVI), wherein the night lights remote sensing image and data of sample areas and vegetation refer to
Number image and data can be obtained by satellite remote sensing date.
Step S30: according to the night lights remotely-sensed data and vegetation index data of sample areas, the sample areas is obtained
In city pixel and non-city pixel.
In the present embodiment, by judging the height of night lights remotely-sensed data, and, the height of vegetation index data,
The a certain pixel for coming judgement sample region belongs to city pixel or non-city pixel, wherein night lights remotely-sensed data is higher
Region and the lower region of vegetation index data are city pixel, and the higher region of vegetation index data is then non-city picture
Member.
Step S40: training sample is chosen from the city pixel and non-city pixel in the sample areas, and according to institute
The night lights remotely-sensed data of the training sample elected and sample areas, vegetation index data and night lights city refer to
Number is established and trains optimal stochastic forest algorithm model, wherein the night lamp of the training sample and sample areas that are elected
Light remotely-sensed data, vegetation index data and night lights city index are defeated as the trained optimal stochastic forest algorithm model
Enter sample, the city pixel of the sample areas and non-city pixel as the defeated of the trained optimal stochastic forest algorithm model
Sample out.
Step S50: the night lights remote sensing image and vegetation index image in region to be identified are obtained.
The region to be identified is to need to be identified as the region of city scope or non-city scope in sample areas.
Wherein, the night lights remote sensing image in region to be identified and data and vegetation index image and data can pass through satellite
Remotely-sensed data obtains.
Step S60: the night lights remote sensing image in the region to be identified and vegetation index image are inputted into optimal stochastic
Forest algorithm model judges whether the region is city scope.
City scope extracting method of the present invention based on random forest sorting algorithm, according to the night of sample areas
Light remote sensing image and vegetation index image and data train optimal random forests algorithm model, can be with by the model
Judge whether the region to be identified is city model according to the night lights remote sensing image in region to be identified and vegetation index image
It encloses, the defect of Satellite Remote Sensing can be made up, improve city scope data.
In one embodiment, in order to preferably screen training sample, according to the night lights remotely-sensed data of sample areas
Before obtaining the city pixel and non-city pixel in the sample areas further include following steps with vegetation index data:
According to water body distributed data, the water body pixel in the night lights remote sensing image is removed.
The water body distributed data is to influence the corresponding distributed data about water body with the night lights remote sensing, can
It is obtained, the water body distribution coordinate in water body distributed data is inputted in night lights remote sensing image, Ke Yiyi by satellite remote sensing
Water body coordinate in night lights remote sensing image out.
Referring to Fig. 2, in one embodiment, according to the night lights remotely-sensed data of sample areas and plant in step S30
By exponent data, the city pixel and non-city pixel in the sample areas are obtained, is specifically comprised the following steps:
Step S31: if the night lights remotely-sensed data in the sample areas in a certain pixel is greater than the first setting threshold
Value, and the vegetation index data in the pixel less than the second given threshold, then the pixel is city pixel, if in a certain pixel
Vegetation index data be greater than the second given threshold, then the pixel be non-city pixel.
Wherein, the first given threshold and the second given threshold are setting value.The pixel is light remote sensing image or vegetation
Minimum resolution unit in index image if the light remotely-sensed data in a certain pixel is greater than the first given threshold, and is planted
By exponent data less than the second given threshold, then illustrate that the pixel is possible for city pixel, if the vegetation in a certain pixel
Exponent data is greater than the second given threshold, then illustrates that the pixel is possible for non-city pixel, and other regions, such as light are distant
Feel data less than the first given threshold, and vegetation index data are less than the pixel of the second given threshold, then compare be difficult to judge be
It is no not therefore to be selected as training sample for city pixel or non-city pixel.In the present embodiment, the first given threshold is 40, the
Two given thresholds are 0.4.
Since in same city, the development degree of different administrative regions is different, and therefore, city scope judgment criteria is
Therefore difference in one embodiment, training sample, different administrative areas is chosen based on administrative region boundary stratified sampling
Domain samples respectively, to choose training sample.
Referring to Fig. 3, in one embodiment, in different administrative regions, it includes following step that training sample is chosen in sampling
It is rapid:
Step S32: generating a value is the random number between 0~1.
Step S33: the ratio chosen between sample size and city pixel and non-city pixel total quantity in each layer is obtained
Example value.
Step S34: if the random number is less than the ratio value, choose sample as training sample this.
Wherein, the calculation formula of the ratio value are as follows: P=Ns/N, wherein Ns is the sample size chosen in each layer, N
For city pixel and non-city pixel total quantity.When choosing pixel, if the random number is less than the ratio value P, the picture
Member is selected as sample pixel, if the random number is greater than the ratio value P, which is not selected.In the present embodiment
In, sample size Ns is the 15% of city pixel and non-city pixel total quantity.
Referring to Fig. 4, in a specific embodiment, the present invention is based on the city scopes of random forest sorting algorithm to mention
Method is taken to include the following steps:
Step S401: the night lights remote sensing image and vegetation index image of sample areas are obtained.
Step S402: according to water body distributed data, the water body pixel in the night lights remote sensing image is removed.
Step S403: the night lights remotely-sensed data and vegetation index data of sample areas are obtained, and according to the night
The night lights city index of light remotely-sensed data and vegetation index data acquisition sample areas.
Step S404: being distinguished the city pixel and non-city pixel of sample areas, and be layered based on administrative division boundary, from
Training sample is chosen in city pixel and non-city pixel.
Step S405: generating a value is the random number between 0~1, and obtains in each layer and choose sample size and city
Ratio value between pixel and non-city pixel total quantity.
Step S406: if the random number is less than the ratio value, choose sample as training sample this.
Step S407: referred to according to the night lights remotely-sensed data of the training sample and sample areas that are elected, vegetation
Number data and night lights city index, establish original sample collection S.
Step S408: k training sample set is extracted in original sample collection S by Bootstrap method.
Step S409: learning k training set, generates k decision-tree model with this.In Decision Tree Construction
In, 4 input variables are shared, n variable is randomly selected from 4 variables, each internal node is become using this n feature
Optimal divisional mode divides in amount, and n value is constant constant in the forming process of Random Forest model.
Step S410: the result of k decision tree is combined, and through repetition training, forms optimal stochastic forest algorithm mould
Type.
Step S411: the night lights remote sensing image and vegetation index image in region to be identified are obtained.
Step S412: the night lights remote sensing image in the region to be identified and vegetation index image are inputted into optimal stochastic
Forest algorithm model judges whether the region is city scope.
City scope extracting method of the present invention based on random forest sorting algorithm, according to the night of sample areas
Light remote sensing image and vegetation index image and data train optimal random forests algorithm model, can be with by the model
Judge whether the region to be identified is city model according to the night lights remote sensing image in region to be identified and vegetation index image
It encloses, the defect of Satellite Remote Sensing can be made up, improve city scope data；By removal water body data, and to sampling sample
This more accurate division, and Random sampling strategy is used, random forest regression model can be more accurately established, is automatically extracted
To more accurate city scope.
Referring to Fig. 5, in one embodiment, the present invention is based on the city scope extraction elements of random forest sorting algorithm
500, comprising:
First data acquisition module 501, for obtaining the night lights remote sensing image and vegetation index image of sample areas；
Second data acquisition module 502, for obtaining the night lights remotely-sensed data and vegetation index data of sample areas,
And according to the night lights city index of the night lights remotely-sensed data and vegetation index data acquisition sample areas；
Sample process module 503 is obtained for the night lights remotely-sensed data and vegetation index data according to sample areas
City pixel and non-city pixel in the sample areas；
Random forest training module 504, for being chosen from the city pixel and non-city pixel in the sample areas
Training sample, and according to the night lights remotely-sensed data of the training sample and sample areas elected, vegetation index data
With night lights city index, establish simultaneously training optimal stochastic forest algorithm model, wherein the training sample that is elected and
Night lights remotely-sensed data, vegetation index data and the night lights city index of sample areas are as the trained optimal stochastic
The input sample of forest algorithm model, the city pixel of the sample areas and non-city pixel are as the trained optimal stochastic
The output sample of forest algorithm model；
Third data acquisition module 505, for obtaining the night lights remote sensing image and vegetation index shadow in region to be identified
Picture；
City scope judgment module 506, for by the night lights remote sensing image and vegetation index in the region to be identified
Image inputs optimal stochastic forest algorithm model, judges whether the region is city scope.
It in one embodiment, further include water body remove module 507, for removing the night according to water body distributed data
Between water body pixel in light remote sensing image.
In one embodiment, the sample process module 503, comprising:
Whether pixel discrimination unit 5031, the light remotely-sensed data for judging in a certain pixel are greater than the first setting threshold
Value, and whether the vegetation index data in the pixel less than the second given threshold, if it is, the pixel is city pixel, such as
Vegetation index data in a certain pixel of fruit are greater than the second given threshold, then the pixel is non-city pixel.
In one embodiment, the random forest training module 504, further includes:
Random number generation unit 5041 is the random number between 0~1 for generating a value；
Ratio value acquiring unit 5042 chooses sample size and city pixel and non-city pixel for obtaining in each layer
Ratio value between total quantity；
Judging unit 5043, for choosing sample as training sample this when the random number is less than the ratio value
This.
City scope extraction element of the present invention based on random forest sorting algorithm, according to the night of sample areas
Light remote sensing image and vegetation index image and data train optimal random forests algorithm model, can be with by the model
Judge whether the region to be identified is city model according to the night lights remote sensing image in region to be identified and vegetation index image
It encloses, the defect of Satellite Remote Sensing can be made up, improve city scope data；By removal water body data, and to sampling sample
This more accurate division, and Random sampling strategy is used, random forest regression model can be more accurately established, is automatically extracted
To more accurate city scope.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
Referring to Fig. 6, in one embodiment, electronic equipment 60 of the invention includes memory 61 and processor 62, with
And the computer program that is stored in the memory 61 and can be executed by the processor 62, the processor 62 execute the meter
When calculation machine program, realize such as the precipitation data estimation side based on random forest sorting algorithm in above-mentioned any one embodiment
Method.
In the present embodiment, controller 62 can be one or more application specific integrated circuit (ASIC), digital signal
Processor (DSP), digital signal processing appts (DSPD), programmable logic device (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components.Storage medium 61 can be used it is one or more its
In include program code storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) on implement
Computer program product form.Computer-readable storage media includes permanent and non-permanent, removable and non-removable
Dynamic media can be accomplished by any method or technique information storage.Information can be computer readable instructions, data structure,
The module of program or other data.The example of the storage medium of computer includes but is not limited to: phase change memory (PRAM), it is static with
Machine access memory (SRAM), dynamic random access memory (DRAM), other kinds of random access memory (RAM), only
It reads memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory techniques, read-only
Compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic magnetic
Disk storage or other magnetic storage devices or any other non-transmission medium, can be used for storage can be accessed by a computing device letter
Breath.
In several embodiments provided herein, it should be understood that disclosed system, device and method can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown or
The mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, the indirect coupling of device or unit
It closes or communicates to connect, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to protection of the invention
Range.
Claims (10)
1. a kind of city scope extracting method based on random forest sorting algorithm, which comprises the steps of:
Obtain the night lights remote sensing image and vegetation index image of sample areas；
Obtain the night lights remotely-sensed data and vegetation index data of sample areas, and according to the night lights remotely-sensed data and
The night lights city index of vegetation index data acquisition sample areas；
According to the night lights remotely-sensed data and vegetation index data of sample areas, the city pixel in the sample areas is obtained
With non-city pixel；
Training sample is chosen from the city pixel and non-city pixel in the sample areas, and according to the training elected
The night lights remotely-sensed data of sample and sample areas, vegetation index data and night lights city index are established and are trained
Optimal stochastic forest algorithm model, wherein the night lights remotely-sensed data of the training sample elected and sample areas is planted
Input sample by exponent data and night lights city index as the trained optimal stochastic forest algorithm model, the sample
The output sample of the city pixel of one's respective area and non-city pixel as the trained optimal stochastic forest algorithm model；
Obtain the night lights remote sensing image and vegetation index image in region to be identified；
The night lights remote sensing image in the region to be identified and vegetation index image are inputted into optimal stochastic forest algorithm model,
Judge whether the region is city scope.
2. the city scope extracting method according to claim 1 based on random forest sorting algorithm, which is characterized in that root
According to the night lights remotely-sensed data and vegetation index data of sample areas, the city pixel in the sample areas and non-city are obtained
City's pixel, includes the following steps:
If the night lights remotely-sensed data in the sample areas in a certain pixel is greater than the first given threshold, and in the pixel
Vegetation index data less than the second given threshold, then the pixel be city pixel, if the vegetation index number in a certain pixel
According to the second given threshold is greater than, then the pixel is non-city pixel.
3. the city scope extracting method according to claim 1 based on random forest sorting algorithm, which is characterized in that from
Training sample is chosen in city pixel and non-city pixel in the sample areas, is included the following steps:
Based on administrative division boundary stratified sampling training sample.
4. the city scope extracting method according to claim 3 based on random forest sorting algorithm, which is characterized in that base
Further include following steps in administrative division boundary stratified sampling training sample:
Generating a value is the random number between 0~1；
Obtain the ratio value chosen between sample size and the city pixel and non-city pixel total quantity in each layer；
If the random number is less than the ratio value, choose sample as training sample this.
5. the city scope extracting method according to claim 1 based on random forest sorting algorithm, which is characterized in that root
According to the night lights remotely-sensed data and vegetation index data of sample areas, the city pixel in the sample areas and non-city are obtained
Further include following steps before city's pixel:
According to water body distributed data, the water body pixel in the night lights remote sensing image is removed.
6. a kind of city scope extraction element based on random forest sorting algorithm characterized by comprising
First data acquisition module, for obtaining the night lights remote sensing image and vegetation index image of sample areas；
Second data acquisition module, for obtaining the night lights remotely-sensed data and vegetation index data of sample areas, and according to
The night lights city index of the night lights remotely-sensed data and vegetation index data acquisition sample areas；
Sample process module obtains the sample for the night lights remotely-sensed data and vegetation index data according to sample areas
City pixel and non-city pixel in one's respective area；
Random forest training module, for choosing training sample from the city pixel and non-city pixel in the sample areas
This, and according to the night lights remotely-sensed data, vegetation index data and night of the training sample and sample areas elected
Light city index is established and trains optimal stochastic forest algorithm model, wherein the training sample elected and sample area
Night lights remotely-sensed data, vegetation index data and the night lights city index in domain are calculated as the trained optimal stochastic forest
The input sample of method model, the city pixel of the sample areas and non-city pixel are calculated as the trained optimal stochastic forest
The output sample of method model；
Third data acquisition module, for obtaining the night lights remote sensing image and vegetation index image in region to be identified；
City scope judgment module, for inputting the night lights remote sensing image in the region to be identified and vegetation index image
Optimal stochastic forest algorithm model judges whether the region is city scope.
7. the city scope extraction element according to claim 6 based on random forest sorting algorithm, which is characterized in that institute
State random forest training module, comprising:
Whether pixel discrimination unit, the night lights remotely-sensed data for judging in the sample areas are greater than the first setting threshold
Value, and whether the vegetation index data in the pixel less than the second given threshold, if it is, the pixel is city pixel, such as
Vegetation index data in a certain pixel of fruit are greater than the second given threshold, then the pixel is non-city pixel.
8. the city scope extraction element according to claim 6 based on random forest sorting algorithm, which is characterized in that also
Include:
Water body remove module, for removing the water body pixel in the night lights remote sensing image according to water body distributed data.
9. a kind of computer-readable medium, is stored thereon with computer program, it is characterised in that:
Realize that Claims 1-4 any one such as is based on random forest sorting algorithm when the computer program is executed by processor
City scope extracting method.
10. a kind of electronic equipment, including memory, processor and it is stored in the memory and can be executed by the processor
Computer program, it is characterised in that:
When the processor executes the computer program, it is random gloomy to realize that any one as described in Claims 1-4 is based on
The city scope extracting method of woods sorting algorithm.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811159030.1A CN109522788B (en) | 2018-09-30 | 2018-09-30 | City range extraction method and device based on random forest classification algorithm and electronic equipment |
| PCT/CN2019/106857 WO2020063461A1 (en) | 2018-09-30 | 2019-09-20 | Urban extent extraction method and apparatus based on random forest classification algorithm, and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811159030.1A CN109522788B (en) | 2018-09-30 | 2018-09-30 | City range extraction method and device based on random forest classification algorithm and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109522788A true CN109522788A (en) | 2019-03-26 |
| CN109522788B CN109522788B (en) | 2020-11-06 |
Family
ID=65771619
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811159030.1A Active CN109522788B (en) | 2018-09-30 | 2018-09-30 | City range extraction method and device based on random forest classification algorithm and electronic equipment |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109522788B (en) |
| WO (1) | WO2020063461A1 (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020063461A1 (en) * | 2018-09-30 | 2020-04-02 | 广州地理研究所 | Urban extent extraction method and apparatus based on random forest classification algorithm, and electronic device |
| CN111144340A (en) * | 2019-12-30 | 2020-05-12 | 中山大学 | Method and system for automatically monitoring human activities in natural reserve area based on night light and high-resolution remote sensing image |
| CN111178179A (en) * | 2019-12-16 | 2020-05-19 | 广州地理研究所 | Method and device for identifying urban functional area based on pixel scale |
| CN111178180A (en) * | 2019-12-16 | 2020-05-19 | 广州地理研究所 | Hyperspectral image feature selection method and device based on improved ant colony algorithm |
| CN111695606A (en) * | 2020-05-25 | 2020-09-22 | 中国科学院东北地理与农业生态研究所 | Multi-type city green land classification method |
| CN113158899A (en) * | 2021-04-22 | 2021-07-23 | 中国科学院地理科学与资源研究所 | Village and town development state measurement method based on remote sensing luminous dark target enhancement technology |
| CN113378785A (en) * | 2021-07-05 | 2021-09-10 | 中煤航测遥感集团有限公司 | Forest type identification method and device |
| CN113392859A (en) * | 2020-03-11 | 2021-09-14 | Tcl科技集团股份有限公司 | Method and device for determining type of city functional area |
| CN115424134A (en) * | 2022-08-26 | 2022-12-02 | 南方海洋科学与工程广东省实验室(广州) | Pollution sea-entering flux prediction method and device based on remote sensing image |
| CN116151610A (en) * | 2022-12-16 | 2023-05-23 | 中国水利水电科学研究院 | Method for simulating risk exposure space of disaster-bearing body on underlying surface of heterogeneous city |
Families Citing this family (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111476308B (en) * | 2020-04-10 | 2024-06-11 | 高崟 | Remote sensing image classification method and device based on priori geometric constraint and electronic equipment |
| CN111709379B (en) * | 2020-06-18 | 2023-04-18 | 广西壮族自治区农业科学院 | Remote sensing image-based hilly area citrus planting land plot monitoring method and system |
| CN111767813B (en) * | 2020-06-19 | 2021-09-28 | 宁波大学 | Multi-feature and long-time-sequence probability synergistic mangrove forest extraction method |
| CN111723525B (en) * | 2020-06-23 | 2023-10-31 | 南通大学 | PM2.5 inversion method based on multi-source data and neural network model |
| CN111832575B (en) * | 2020-07-16 | 2024-04-30 | 黄河勘测规划设计研究院有限公司 | Remote sensing image-based water surface area extraction method and device |
| CN112131946B (en) * | 2020-08-21 | 2023-06-23 | 长沙银汉空间科技有限公司 | Automatic extraction method for vegetation and water information of optical remote sensing image |
| CN112287756A (en) * | 2020-09-25 | 2021-01-29 | 北京佳格天地科技有限公司 | Ground object identification method, device, storage medium and terminal |
| CN112257531B (en) * | 2020-10-13 | 2023-07-28 | 西安电子科技大学 | Remote sensing monitoring method for forest land change based on diversity feature combination |
| CN112257626A (en) * | 2020-10-29 | 2021-01-22 | 辽宁工程技术大学 | Method and system for sampling remote sensing data |
| CN112381013B (en) * | 2020-11-18 | 2024-02-27 | 南通市测绘院有限公司 | Urban vegetation inversion method and system based on high-resolution remote sensing image |
| CN112683811B (en) * | 2020-11-27 | 2023-01-10 | 南京观微空间科技有限公司 | Forest canopy spectrum quaternary phase change monitoring method based on medium-resolution image |
| CN112395383B (en) * | 2020-11-30 | 2024-03-22 | 湘潭大学 | Population data spatialization method, device and equipment based on multi-source information fusion |
| CN112508386A (en) * | 2020-12-04 | 2021-03-16 | 湘潭大学 | Noctilucent and POI data fused urban built-up area extraction method |
| CN112800826A (en) * | 2020-12-15 | 2021-05-14 | 四川航天神坤科技有限公司 | City change monitoring method and system based on remote sensing image classification |
| CN112560990A (en) * | 2020-12-25 | 2021-03-26 | 济南浪潮高新科技投资发展有限公司 | Remote sensing image city expansion monitoring method based on data mining |
| CN112860767B (en) * | 2021-02-09 | 2023-04-21 | 中国科学院地理科学与资源研究所 | Identification of urban and rural transition zone based on time sequence data |
| CN113158934A (en) * | 2021-04-28 | 2021-07-23 | 中国科学院空天信息创新研究院 | High-resolution remote sensing image-based urban land use classification method, device and equipment |
| CN113392585B (en) * | 2021-06-10 | 2023-11-03 | 京师天启(北京)科技有限公司 | Method for spatialization of sensitive crowd around polluted land |
| CN113761790B (en) * | 2021-07-27 | 2024-04-23 | 河海大学 | Fruit tree leaf nitrogen content estimation method based on Stacking integrated learning |
| CN113723228B (en) * | 2021-08-16 | 2024-04-26 | 北京大学 | Method and device for determining earth surface type duty ratio, electronic equipment and storage medium |
| CN113780174B (en) * | 2021-09-10 | 2023-09-15 | 福州大学 | Storm-type landslide identification method for high vegetation platform combined with random forest algorithm |
| CN113780232B (en) * | 2021-09-23 | 2024-02-02 | 中国科学院东北地理与农业生态研究所 | Urban wetland dynamic monitoring method |
| CN113989657B (en) * | 2021-10-11 | 2024-05-10 | 中国测绘科学研究院 | Cultivated land range change detection method and device based on invariable information sample screening |
| CN113936054A (en) * | 2021-10-18 | 2022-01-14 | 中国科学院东北地理与农业生态研究所 | Method for detecting area and distribution range of plastic greenhouse |
| CN114266138B (en) * | 2021-11-29 | 2022-09-16 | 西南大学 | Method for recognizing and verifying urban edge area by using cloud data |
| CN114529721B (en) * | 2022-02-08 | 2024-05-10 | 山东浪潮科学研究院有限公司 | Urban remote sensing image vegetation coverage recognition method based on deep learning |
| CN114782835B (en) * | 2022-05-05 | 2024-04-26 | 中国农业大学 | Crop lodging area proportion detection method and device |
| CN114581784B (en) * | 2022-05-07 | 2022-08-12 | 自然资源部第二海洋研究所 | Construction method of long-time-sequence yearly mangrove remote sensing monitoring product |
| CN114782255B (en) * | 2022-06-16 | 2022-09-02 | 武汉大学 | Semantic-based noctilucent remote sensing image high-resolution reconstruction method |
| CN114782837B (en) * | 2022-06-17 | 2022-10-18 | 中化现代农业有限公司 | Plant estimation method, plant estimation device, electronic equipment and storage medium |
| CN115223062B (en) * | 2022-06-30 | 2023-10-20 | 桂林理工大学 | Eucalyptus artificial forest area stand accumulation amount time difference correction method based on UAV data |
| CN115272878A (en) * | 2022-07-22 | 2022-11-01 | 中国自然资源航空物探遥感中心 | Plateau ice lake automatic extraction method based on machine learning |
| CN115937454B (en) * | 2023-01-09 | 2023-05-30 | 航天宏图信息技术股份有限公司 | Method and device for automatically placing tree models in large-scale city scene |
| CN117423003B (en) * | 2023-12-19 | 2024-03-19 | 山东科技大学 | Winter wheat seedling condition grading remote sensing monitoring method in overwintering period |
| CN117495425B (en) * | 2023-12-29 | 2024-04-12 | 武汉大学 | Asset financial estimation method and system based on multidimensional noctilucent features |
| CN117787813B (en) * | 2024-02-23 | 2024-05-31 | 中国农业科学院农业资源与农业区划研究所 | Fine-granularity feature constraint-based urban village identification method and device and electronic equipment |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090281981A1 (en) * | 2008-05-06 | 2009-11-12 | Chen Barry Y | Discriminant Forest Classification Method and System |
| US20090310867A1 (en) * | 2008-06-12 | 2009-12-17 | Bogdan Calin Mihai Matei | Building segmentation for densely built urban regions using aerial lidar data |
| CN104850853A (en) * | 2015-06-02 | 2015-08-19 | 中国地质大学（武汉） | Rapid city extraction method and rapid city extraction device |
| CN105260729A (en) * | 2015-11-20 | 2016-01-20 | 武汉大学 | Satellite remote sensing image cloud amount calculation method on the basis of random forest |
| CN105844300A (en) * | 2016-03-24 | 2016-08-10 | 河南师范大学 | Optimized classification method and optimized classification device based on random forest algorithm |
| CN106056134A (en) * | 2016-05-20 | 2016-10-26 | 重庆大学 | Semi-supervised random forests classification method based on Spark |
| CN106127121A (en) * | 2016-06-15 | 2016-11-16 | 四川省遥感信息测绘院 | A kind of built-up areas intellectuality extracting method based on nighttime light data |
| CN106682830A (en) * | 2016-12-26 | 2017-05-17 | 中国科学院深圳先进技术研究院 | Urbanization ghost town index evaluating method and device |
| US20180189667A1 (en) * | 2016-12-29 | 2018-07-05 | Intel Corporation | Entropy-based weighting in random forest models |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103620381B (en) * | 2011-06-29 | 2015-09-30 | 富士通株式会社 | Floristics recognition device and method |
| CN105608474B (en) * | 2015-12-30 | 2019-07-09 | 二十一世纪空间技术应用股份有限公司 | Region adaptivity plant extraction method based on high resolution image |
| CN107895136B (en) * | 2017-08-15 | 2020-09-15 | 东北大学 | Coal mine area identification method and system |
| CN108280440A (en) * | 2018-02-09 | 2018-07-13 | 三亚中科遥感研究所 | A kind of fruit-bearing forest recognition methods and system |
| CN109522788B (en) * | 2018-09-30 | 2020-11-06 | 广州地理研究所 | City range extraction method and device based on random forest classification algorithm and electronic equipment |
-
2018
- 2018-09-30 CN CN201811159030.1A patent/CN109522788B/en active Active
-
2019
- 2019-09-20 WO PCT/CN2019/106857 patent/WO2020063461A1/en active Application Filing
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090281981A1 (en) * | 2008-05-06 | 2009-11-12 | Chen Barry Y | Discriminant Forest Classification Method and System |
| US20090310867A1 (en) * | 2008-06-12 | 2009-12-17 | Bogdan Calin Mihai Matei | Building segmentation for densely built urban regions using aerial lidar data |
| CN104850853A (en) * | 2015-06-02 | 2015-08-19 | 中国地质大学（武汉） | Rapid city extraction method and rapid city extraction device |
| CN105260729A (en) * | 2015-11-20 | 2016-01-20 | 武汉大学 | Satellite remote sensing image cloud amount calculation method on the basis of random forest |
| CN105844300A (en) * | 2016-03-24 | 2016-08-10 | 河南师范大学 | Optimized classification method and optimized classification device based on random forest algorithm |
| CN106056134A (en) * | 2016-05-20 | 2016-10-26 | 重庆大学 | Semi-supervised random forests classification method based on Spark |
| CN106127121A (en) * | 2016-06-15 | 2016-11-16 | 四川省遥感信息测绘院 | A kind of built-up areas intellectuality extracting method based on nighttime light data |
| CN106682830A (en) * | 2016-12-26 | 2017-05-17 | 中国科学院深圳先进技术研究院 | Urbanization ghost town index evaluating method and device |
| US20180189667A1 (en) * | 2016-12-29 | 2018-07-05 | Intel Corporation | Entropy-based weighting in random forest models |
Non-Patent Citations (5)
| Title |
|---|
| CAMILA SOUZA DOS ANJOS等: "Classification of Urban Environment Using Feature Extraction and Random Forest", 《IEEE》 * |
| 李方方等: "面向对象随机森林方法在湿地植被分类的应用", 《遥感信息》 * |
| 沈润平等: "基于随机森林的遥感干旱监测模型的构建", 《地球信息科学》 * |
| 田绍鸿等: "采用随机森林法的天绘数据干旱区城市土地覆盖分类方法", 《国土资源遥感》 * |
| 郜燕芳等: "基于随机森林模型的城市不透水面提取研究—以呼和浩特市为例", 《冰川冻土》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020063461A1 (en) * | 2018-09-30 | 2020-04-02 | 广州地理研究所 | Urban extent extraction method and apparatus based on random forest classification algorithm, and electronic device |
| CN111178179A (en) * | 2019-12-16 | 2020-05-19 | 广州地理研究所 | Method and device for identifying urban functional area based on pixel scale |
| CN111178180A (en) * | 2019-12-16 | 2020-05-19 | 广州地理研究所 | Hyperspectral image feature selection method and device based on improved ant colony algorithm |
| CN111178179B (en) * | 2019-12-16 | 2023-08-11 | 广州地理研究所 | Urban functional area identification method and device based on pixel scale |
| CN111178180B (en) * | 2019-12-16 | 2023-08-11 | 广州地理研究所 | Hyperspectral image feature selection method and device based on improved ant colony algorithm |
| CN111144340A (en) * | 2019-12-30 | 2020-05-12 | 中山大学 | Method and system for automatically monitoring human activities in natural reserve area based on night light and high-resolution remote sensing image |
| CN113392859A (en) * | 2020-03-11 | 2021-09-14 | Tcl科技集团股份有限公司 | Method and device for determining type of city functional area |
| CN111695606A (en) * | 2020-05-25 | 2020-09-22 | 中国科学院东北地理与农业生态研究所 | Multi-type city green land classification method |
| CN111695606B (en) * | 2020-05-25 | 2022-10-04 | 中国科学院东北地理与农业生态研究所 | Multi-type city green land classification method |
| CN113158899B (en) * | 2021-04-22 | 2022-07-29 | 中国科学院地理科学与资源研究所 | Village and town development state measurement method based on remote sensing luminous dark target enhancement technology |
| CN113158899A (en) * | 2021-04-22 | 2021-07-23 | 中国科学院地理科学与资源研究所 | Village and town development state measurement method based on remote sensing luminous dark target enhancement technology |
| CN113378785A (en) * | 2021-07-05 | 2021-09-10 | 中煤航测遥感集团有限公司 | Forest type identification method and device |
| CN115424134A (en) * | 2022-08-26 | 2022-12-02 | 南方海洋科学与工程广东省实验室(广州) | Pollution sea-entering flux prediction method and device based on remote sensing image |
| CN115424134B (en) * | 2022-08-26 | 2024-04-09 | 南方海洋科学与工程广东省实验室(广州) | Pollution sea flux prediction method and device based on remote sensing image |
| CN116151610A (en) * | 2022-12-16 | 2023-05-23 | 中国水利水电科学研究院 | Method for simulating risk exposure space of disaster-bearing body on underlying surface of heterogeneous city |
| CN116151610B (en) * | 2022-12-16 | 2023-10-03 | 中国水利水电科学研究院 | Method for simulating risk exposure space of disaster-bearing body on underlying surface of heterogeneous city |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020063461A1 (en) | 2020-04-02 |
| CN109522788B (en) | 2020-11-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109522788A (en) | City scope extracting method, device and electronic equipment based on random forest sorting algorithm | |
| CN111738314B (en) | Deep learning method of multi-modal image visibility detection model based on shallow fusion | |
| CN106951836B (en) | crop coverage extraction method based on prior threshold optimization convolutional neural network | |
| Schiewe | Segmentation of high-resolution remotely sensed data-concepts, applications and problems | |
| Clode et al. | The automatic extraction of roads from LIDAR data | |
| Zhang et al. | Object-oriented method for urban vegetation mapping using IKONOS imagery | |
| CN103927741B (en) | SAR image synthesis method for enhancing target characteristics | |
| CN110059608A (en) | A kind of object detecting method, device, electronic equipment and storage medium | |
| CN104318546B (en) | Multi-scale analysis-based greenhouse field plant leaf margin extraction method and system | |
| KR101183105B1 (en) | Method of establishing information of cloud data and establishing system of information of cloud data | |
| Varhola et al. | Estimation of watershed-level distributed forest structure metrics relevant to hydrologic modeling using LiDAR and Landsat | |
| CN104050481B (en) | Multi-template infrared image real-time pedestrian detection method combining contour feature and gray level | |
| CN108197583A (en) | The building change detecting method of optimization and image structure feature is cut based on figure | |
| CN105139375B (en) | Combining global DEM and stereoscopic vision a kind of satellite image cloud detection method of optic | |
| CN104598908A (en) | Method for recognizing diseases of crop leaves | |
| CN109191432A (en) | The remote sensing images cloud detection method of optic of filtering multi-resolution decomposition is converted based on domain | |
| JP2010086276A (en) | Method and system for tree classification, method and system for generating current status information of forest, and method and system for selecting target area for tree thinning | |
| CN109472393A (en) | Space NO emissions reduction precipitation data detection method, device and electronic equipment | |
| CN109522516A (en) | Soil moisture detection method, device and electronic equipment based on random forest regression algorithm | |
| Wang et al. | Landscape-level vegetation classification and fractional woody and herbaceous vegetation cover estimation over the dryland ecosystems by unmanned aerial vehicle platform | |
| CN109447325A (en) | Precipitation data detection method, device and electronic equipment based on random forests algorithm | |
| Hu et al. | A fast and simple method of building detection from LiDAR data based on scan line analysis | |
| Mustafa et al. | Identification and mapping of tree species in urban areas using worldview-2 imagery | |
| CN104952070A (en) | Near-rectangle guide based remote-sensing cornfield image segmentation method | |
| Wang et al. | Color-and texture-based image segmentation for improved forest delineation |
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 | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 510075 one of the compound No. 100, Xianlie Middle Road, Yuexiu District, Guangzhou City, Guangdong Province Patentee after: Guangzhou Institute of geography, Guangdong Academy of Sciences Address before: 510075 one of the compound No. 100, Xianlie Middle Road, Yuexiu District, Guangzhou City, Guangdong Province Patentee before: GUANGZHOU INSTITUTE OF GEOGRAPHY |