CN109522788B - City range extraction method and device based on random forest classification algorithm and electronic equipment - Google Patents
City range extraction method and device based on random forest classification algorithm and electronic equipment Download PDFInfo
- Publication number
- CN109522788B CN109522788B CN201811159030.1A CN201811159030A CN109522788B CN 109522788 B CN109522788 B CN 109522788B CN 201811159030 A CN201811159030 A CN 201811159030A CN 109522788 B CN109522788 B CN 109522788B
- Authority
- CN
- China
- Prior art keywords
- urban
- remote sensing
- night light
- vegetation index
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- 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
Abstract
The invention relates to a method and a device for extracting an urban area based on a random forest classification algorithm and electronic equipment. The city range extraction method based on the random forest classification algorithm comprises the following steps: acquiring a night light city index of the sample area according to the night light remote sensing data and the vegetation index data; selecting a training sample from the night light remote sensing image and the vegetation index image of the sample area, and establishing and training an optimal random forest algorithm model according to the selected training sample, the night light remote sensing data, the vegetation index data and the night light urban index of the sample area; and inputting the night light remote sensing image and the vegetation index image of the area to be identified into an optimal random forest algorithm model, and judging whether the area is in an urban range. The urban area extraction method based on the random forest classification algorithm can identify urban areas and non-urban areas according to the night light remote sensing images and the vegetation indexes.
Description
Technical Field
The invention relates to the technical field of urban development research information, in particular to a method and a device for extracting an urban range based on a random forest classification algorithm and electronic equipment.
Background
At present, the urban range extraction mainly adopts the aerospace remote sensing technology, compared with the traditional ground actual measurement method, the remote sensing technology has the characteristics of small workload, low cost, short period, high efficiency and the like, and can meet the requirement of the current research on urbanization. The traditional method for extracting the city range by using the remote sensing technology generally utilizes multispectral remote sensing images with higher resolution within one year to extract, and the extraction process mainly comprises the steps of geometric correction, atmospheric correction, inlaying, cutting, classifying and the like of the images. The multispectral remote sensing image is greatly influenced by weather, so that the obtained image is difficult to operate due to different imaging time when the obtained image is subjected to geometric correction, mosaic and other work.
Disclosure of Invention
Based on the above, the invention aims to provide an urban area extraction method based on a random forest classification algorithm, which can identify urban areas and non-urban areas according to night light remote sensing images and vegetation indexes.
The invention is realized by the following scheme:
a city range extraction method based on a random forest classification algorithm comprises the following steps:
acquiring a night light remote sensing image and a vegetation index image of the sample area;
acquiring night light remote sensing data and vegetation index data of the sample area, and acquiring a night light city index of the sample area according to the night light remote sensing data and the vegetation index data;
acquiring urban pixels and non-urban pixels in a sample area according to night lamplight remote sensing data and vegetation index data of the sample area;
selecting training samples from urban pixels and non-urban pixels in the sample area, and establishing and training an optimal random forest algorithm model according to the selected training samples and night light remote sensing data, vegetation index data and night light urban indexes of the sample area, wherein the selected training samples and the night light remote sensing data, vegetation index data and night light urban indexes of the sample area are used as input samples of the training optimal random forest algorithm model, and the urban pixels and the non-urban pixels of the sample area are used as output samples of the training optimal random forest algorithm model;
acquiring a night light remote sensing image and a vegetation index image of an area to be identified;
and inputting the night light remote sensing image and the vegetation index image of the area to be identified into an optimal random forest algorithm model, and judging whether the area is in an urban range.
According to the urban range extraction method based on the random forest classification algorithm, an optimal random forest algorithm model is trained according to the night light remote sensing image and the vegetation index image and data of the sample area, whether the area to be identified is an urban range or not can be judged according to the night light remote sensing image and the vegetation index image of the area to be identified through the model, the defect of satellite remote sensing monitoring can be overcome, and urban range data are perfected.
In one embodiment, the method for acquiring the urban image elements and the non-urban image elements in the sample area according to the night light remote sensing data and the vegetation index data of the sample area comprises the following steps:
if the nighttime lamplight remote sensing data in a certain pixel in the sample area is larger than a first set threshold value and the vegetation index data in the pixel is smaller than a second set threshold value, the pixel is an urban pixel, and if the vegetation index data in the certain pixel is larger than the second set threshold value, the pixel is a non-urban pixel.
In one embodiment, selecting training samples from the urban and non-urban image elements in the sample area comprises:
training samples are sampled hierarchically based on administrative division boundaries.
In one embodiment, the training samples are sampled hierarchically based on the administrative division boundary, and the method further comprises the following steps:
generating a random number with a value of 0-1;
obtaining a ratio value between the number of selected samples in each layer and the total number of urban pixels and non-urban pixels;
and if the random number is smaller than the proportion value, taking the selected sample as a training sample.
In one embodiment, before the urban pixel and the non-urban pixel in the sample area are obtained according to the night light remote sensing data and the vegetation index data of the sample area, the method further comprises the following steps:
and removing water pixels in the night lamplight remote sensing image according to the water distribution data.
Further, the invention also provides a city range extraction device based on the random forest classification algorithm, which comprises the following steps:
the first data acquisition module is used for acquiring a night lamplight remote sensing image and a vegetation index image of the sample area;
the second data acquisition module is used for acquiring night light remote sensing data and vegetation index data of the sample area and acquiring a night light city index of the sample area according to the night light remote sensing data and the vegetation index data;
the sample processing module is used for acquiring urban pixels and non-urban pixels in the sample area according to the night lamplight remote sensing data and the vegetation index data of the sample area;
the random forest training module is used for selecting training samples from the urban pixels and the non-urban pixels in the sample area, and establishing and training an optimal random forest algorithm model according to the selected training samples and the night light remote sensing data, vegetation index data and night light urban indexes of the sample area, wherein the selected training samples and the night light remote sensing data, vegetation index data and night light urban indexes of the sample area are used as input samples of the optimal random forest algorithm model, and the urban pixels and the non-urban pixels of the sample area are used as output samples of the optimal random forest algorithm model;
the third data acquisition module is used for acquiring a night light remote sensing image and a vegetation index image of the area to be identified;
and the city range judging module is used for inputting the night light remote sensing image and the vegetation index image of the area to be identified into an optimal random forest algorithm model and judging whether the area is a city range.
According to the urban range extraction device based on the random forest classification algorithm, an optimal random forest algorithm model is trained according to the night light remote sensing image and the vegetation index image and data of the sample area, whether the area to be identified is an urban range can be judged according to the night light remote sensing image and the vegetation index image of the area to be identified through the model, the defect of satellite remote sensing monitoring can be overcome, and urban range data can be perfected.
In one embodiment, the random forest training module comprises:
and the pixel distinguishing unit is used for judging whether the night lamplight remote sensing data in the sample area is larger than a first set threshold value or not and whether the vegetation index data in the pixel is smaller than a second set threshold value or not, if so, the pixel is an urban pixel, and if the vegetation index data in a certain pixel is larger than the second set threshold value, the pixel is a non-urban pixel.
In one embodiment, further comprising:
and the water body removing module is used for removing water body pixels in the night lamplight remote sensing image according to the water body distribution data.
Further, the present invention also provides a computer readable medium, on which a computer program is stored, which when executed by a processor implements any one of the above-mentioned methods for extracting a city range based on a random forest classification algorithm.
Further, the present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable by the processor, and when the processor executes the computer program, the city range extraction method based on the random forest classification algorithm is implemented as any one of the above methods.
For a better understanding and practice, the invention is described in detail below with reference to the accompanying drawings.
Drawings
FIG. 1 is a flow chart of a city range extraction method based on a random forest classification algorithm in an embodiment;
FIG. 2 is a flow diagram of selecting a training sample in one embodiment;
FIG. 3 is a flow chart of sampling and selecting training samples for different administrative regions;
FIG. 4 is a flow chart of a city range extraction method based on a random forest classification algorithm in an embodiment;
FIG. 5 is a schematic diagram of an embodiment of a random forest classification algorithm-based city range extraction apparatus;
fig. 6 is a schematic structural diagram of an electronic device in an embodiment.
Detailed Description
Referring to fig. 1, in an embodiment, a city range extraction method based on a random forest classification algorithm includes the following steps:
step S10: and acquiring a night lamplight remote sensing image and a vegetation index image of the sample area.
Step S20: and acquiring night light remote sensing data and vegetation index data of the sample area, and acquiring a night light city index of the sample area according to the night light remote sensing data and the vegetation index data.
The night light remote sensing image is a visible light source image of a land or a water body acquired by a remote sensing sensor on a satellite under the condition of no cloud at night, the night light remote sensing data (NTL) is data corresponding to the night light remote sensing image and is a numerical value indication of light brightness in each resolution ratio in the night light remote sensing image, and the value of the night light remote sensing data is usually 0-63. The vegetation index (NDVI) is a vegetation coverage index formed by combining satellite visible light and near infrared wave bands according to the spectral characteristics of vegetation, and qualitatively and quantitatively evaluates vegetation coverage and growth vigor thereof. The value of the vegetation index is usually-1 to 1, the vegetation index is usually a constant less than zero in ice and snow coverage, water bodies and desert areas, and the vegetation index image is an image formed by a data table corresponding to the vegetation index data and the night light remote sensing image. The night light city Index (the Vegetation Adjusted NTL Urban Index, VANUI) is the night light remote sensing data Adjusted according to the Vegetation Index data, and in this embodiment, the calculation formula of the night light city Index is as follows: and (3) VANUI (NTL) (1-NDVI), wherein the night light remote sensing image and data and the vegetation index image and data of the sample area can be obtained through satellite remote sensing data.
Step S30: and acquiring the urban pixels and the non-urban pixels in the sample area according to the night lamplight remote sensing data and the vegetation index data in the sample area.
In this embodiment, the height of the night light remote sensing data and the height of the vegetation index data are determined to determine that a certain pixel of the sample area belongs to an urban pixel or a non-urban pixel, wherein the area with the higher night light remote sensing data and the lower vegetation index data is the urban pixel, and the area with the higher vegetation index data is the non-urban pixel.
Step S40: selecting training samples from the urban pixels and the non-urban pixels in the sample area, and establishing and training an optimal random forest algorithm model according to the selected training samples and the night light remote sensing data, vegetation index data and night light urban indexes of the sample area, wherein the selected training samples and the night light remote sensing data, vegetation index data and night light urban indexes of the sample area are used as input samples of the training optimal random forest algorithm model, and the urban pixels and the non-urban pixels of the sample area are used as output samples of the training optimal random forest algorithm model.
Step S50: and acquiring a night light remote sensing image and a vegetation index image of the area to be identified.
The region to be identified is a region which needs to be identified as an urban region or a non-urban region in the sample region.
The night light remote sensing image and data and the vegetation index image and data of the area to be identified can be obtained through satellite remote sensing data.
Step S60: and inputting the night light remote sensing image and the vegetation index image of the area to be identified into an optimal random forest algorithm model, and judging whether the area is in an urban range.
According to the urban range extraction method based on the random forest classification algorithm, an optimal random forest algorithm model is trained according to the night light remote sensing image and the vegetation index image and data of the sample area, whether the area to be identified is an urban range or not can be judged according to the night light remote sensing image and the vegetation index image of the area to be identified through the model, the defect of satellite remote sensing monitoring can be overcome, and urban range data are perfected.
In one embodiment, for better screening of a training sample, before acquiring urban pixels and non-urban pixels in a sample area according to night light remote sensing data and vegetation index data of the sample area, the method further comprises the following steps:
and removing water pixels in the night lamplight remote sensing image according to the water distribution data.
The water body distribution data are distribution data about the water body corresponding to the night light remote sensing influence, can be obtained through satellite remote sensing, and water body distribution coordinates in the water body distribution data are input into the night light remote sensing image, so that the water body coordinates in the night light remote sensing image can be moved out.
Referring to fig. 2, in an embodiment, in step S30, according to the night light remote sensing data and the vegetation index data of the sample area, acquiring an urban pixel and a non-urban pixel in the sample area includes the following steps:
step S31: if the nighttime lamplight remote sensing data in a certain pixel in the sample area is larger than a first set threshold value and the vegetation index data in the pixel is smaller than a second set threshold value, the pixel is an urban pixel, and if the vegetation index data in the certain pixel is larger than the second set threshold value, the pixel is a non-urban pixel.
The first set threshold and the second set threshold are set values. The pixels are minimum resolution units in the light remote sensing images or the vegetation index images, if light remote sensing data in a certain pixel are larger than a first set threshold and the vegetation index data are smaller than a second set threshold, the pixel is likely to be an urban pixel, if the vegetation index data in the certain pixel are larger than the second set threshold, the pixel is likely to be a non-urban pixel, and other areas, such as the pixels with the light remote sensing data smaller than the first set threshold and the vegetation index data smaller than the second set threshold, are difficult to judge whether the pixels are the urban pixel or the non-urban pixel, so that the pixels are not selected as training samples. In this embodiment, the first set threshold is 40, and the second set threshold is 0.4.
Since the developed degrees of different administrative areas are different in the same city, and therefore, the urban area judgment criteria are different, in one embodiment, the training samples are selected based on hierarchical sampling of administrative area boundaries, and the training samples are selected by respectively sampling different administrative areas.
Referring to fig. 3, in one embodiment, sampling and selecting training samples in different administrative areas includes the following steps:
step S32: generating a random number with a value between 0 and 1.
Step S33: and obtaining a ratio value between the number of the selected samples in each layer and the total number of the urban pixels and the non-urban pixels.
Step S34: and if the random number is smaller than the proportion value, taking the selected sample as a training sample.
Wherein, the calculation formula of the proportional value is as follows: and P is Ns/N, wherein Ns is the number of samples selected in each layer, and N is the total number of the urban pixels and the non-urban pixels. When a pixel is selected, if the random number is smaller than the proportion value P, the pixel is selected as a sample pixel, and if the random number is larger than the proportion value P, the pixel is not selected. In this embodiment, the number of samples Ns is 15% of the total number of urban and non-urban pixels.
Referring to fig. 4, in a specific embodiment, the method for extracting a city range based on a random forest classification algorithm of the present invention includes the following steps:
step S401: and acquiring a night lamplight remote sensing image and a vegetation index image of the sample area.
Step S402: and removing water pixels in the night lamplight remote sensing image according to the water distribution data.
Step S403: and acquiring night light remote sensing data and vegetation index data of the sample area, and acquiring a night light city index of the sample area according to the night light remote sensing data and the vegetation index data.
Step S404: and distinguishing the urban pixels and the non-urban pixels in the sample area, layering based on administrative division boundaries, and selecting training samples from the urban pixels and the non-urban pixels.
Step S405: and generating a random number with a value of 0-1, and acquiring a ratio value between the number of the selected samples in each layer and the total number of the urban pixels and the non-urban pixels.
Step S406: and if the random number is smaller than the proportion value, taking the selected sample as a training sample.
Step S407: and establishing an original sample set S according to the selected training sample and the night light remote sensing data, vegetation index data and night light city index of the sample area.
Step S408: k training sample sets are extracted from the original sample set S by the Bootstrap method.
Step S409: and learning the k training sets to generate k decision tree models. In the process of generating the decision tree, 4 input variables are totally arranged, n variables are randomly extracted from the 4 variables, each internal node is split by utilizing the optimal splitting mode on the n characteristic variables, and the value of n is a constant in the process of forming the random forest model.
Step S410: and combining the results of the k decision trees, and repeatedly training to form an optimal random forest algorithm model.
Step S411: and acquiring a night light remote sensing image and a vegetation index image of the area to be identified.
Step S412: and inputting the night light remote sensing image and the vegetation index image of the area to be identified into an optimal random forest algorithm model, and judging whether the area is in an urban range.
According to the urban range extraction method based on the random forest classification algorithm, an optimal random forest algorithm model is trained according to the night light remote sensing image and the vegetation index image and data of the sample area, whether the area to be identified is an urban range or not can be judged according to the night light remote sensing image and the vegetation index image of the area to be identified through the model, the defect of satellite remote sensing monitoring can be overcome, and urban range data are perfected; by removing water body data, dividing sampling samples more accurately and adopting a random number sampling strategy, a random forest regression model can be established more accurately, and a more accurate city range can be automatically extracted.
Referring to fig. 5, in an embodiment, the city range extracting apparatus 500 based on the random forest classification algorithm of the present invention includes:
the first data acquisition module 501 is used for acquiring a night light remote sensing image and a vegetation index image of the sample area;
the second data acquisition module 502 is used for acquiring night light remote sensing data and vegetation index data of the sample area and acquiring a night light city index of the sample area according to the night light remote sensing data and the vegetation index data;
the sample processing module 503 is configured to obtain an urban pixel and a non-urban pixel in the sample area according to the night light remote sensing data and the vegetation index data of the sample area;
a random forest training module 504, configured to select a training sample from the urban pixel and the non-urban pixel in the sample area, and establish and train an optimal random forest algorithm model according to the selected training sample and the nighttime light remote sensing data, vegetation index data, and nighttime light urban index of the sample area, where the selected training sample and the nighttime light remote sensing data, vegetation index data, and nighttime light urban index of the sample area are used as input samples of the training optimal random forest algorithm model, and the urban pixel and the non-urban pixel of the sample area are used as output samples of the training optimal random forest algorithm model;
the third data acquisition module 505 is configured to acquire a night light remote sensing image and a vegetation index image of an area to be identified;
and the city range judging module 506 is used for inputting the night light remote sensing image and the vegetation index image of the area to be identified into an optimal random forest algorithm model and judging whether the area is a city range.
In an embodiment, the system further includes a water removing module 507, configured to remove water pixels in the night light remote sensing image according to water distribution data.
In one embodiment, the sample processing module 503 comprises:
the pixel distinguishing unit 5031 is configured to determine whether light remote sensing data in a certain pixel is greater than a first set threshold, and whether vegetation index data in the pixel is less than a second set threshold, if so, the pixel is an urban pixel, and if vegetation index data in the certain pixel is greater than the second set threshold, the pixel is a non-urban pixel.
In an embodiment, the random forest training module 504 further includes:
a random number generation unit 5041, configured to generate a random number with a value of 0-1;
a proportion value obtaining unit 5042, configured to obtain a proportion value between the number of selected samples in each layer and the total number of urban pixels and non-urban pixels;
a determining unit 5043, configured to use the selected sample as a training sample when the random number is smaller than the ratio value.
According to the urban range extraction device based on the random forest classification algorithm, an optimal random forest algorithm model is trained according to the night light remote sensing image and the vegetation index image and data of the sample area, whether the area to be identified is an urban range or not can be judged according to the night light remote sensing image and the vegetation index image of the area to be identified through the model, the defect of satellite remote sensing monitoring can be overcome, and urban range data are perfected; by removing water body data, dividing sampling samples more accurately and adopting a random number sampling strategy, a random forest regression model can be established more accurately, and a more accurate city range can be automatically extracted.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
Referring to fig. 6, in an embodiment, the electronic device 60 of the present invention includes a memory 61 and a processor 62, and a computer program stored in the memory 61 and executable by the processor 62, wherein the processor 62, when executing the computer program, implements the method for estimating precipitation data based on the random forest classification algorithm according to any one of the above embodiments.
In the present embodiment, the controller 62 may be one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors, or other electronic components. Storage medium 61 may take the form of a computer program product embodied on one or more storage media including, but not limited to, disk storage, CD-ROM, optical storage, etc., having program code embodied therein. Computer readable storage media, which include both non-transitory and non-transitory, removable and non-removable media, may implement any method or technology for storage of information. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of the storage medium of the computer include, but are not limited to: phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technologies, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic tape storage or other magnetic storage devices, or any other non-transmission medium, may be used to store information that may be accessed by a computing device.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention.
Claims (8)
1. A city range extraction method based on a random forest classification algorithm is characterized by comprising the following steps:
acquiring a night light remote sensing image and a vegetation index image of the sample area;
acquiring night light remote sensing data and vegetation index data of the sample area, and acquiring a night light city index of the sample area according to the night light remote sensing data and the vegetation index data;
acquiring an urban pixel and a non-urban pixel in a sample area according to night light remote sensing data and vegetation index data of the sample area, wherein if the night light remote sensing data in a certain pixel in the sample area is larger than a first set threshold and the vegetation index data in the pixel is smaller than a second set threshold, the pixel is the urban pixel, and if the vegetation index data in the certain pixel is larger than the second set threshold, the pixel is the non-urban pixel, wherein the area with higher night light remote sensing data and lower vegetation index data is the urban pixel, and the area with higher vegetation index data is the non-urban pixel;
selecting training samples from urban pixels and non-urban pixels in the sample area, and establishing and training an optimal random forest algorithm model according to the selected training samples and night light remote sensing data, vegetation index data and night light urban indexes of the sample area, wherein the selected training samples and the night light remote sensing data, vegetation index data and night light urban indexes of the sample area are used as input samples of the training optimal random forest algorithm model, and the urban pixels and the non-urban pixels of the sample area are used as output samples of the training optimal random forest algorithm model;
acquiring a night light remote sensing image and a vegetation index image of an area to be identified;
and inputting the night light remote sensing image and the vegetation index image of the area to be identified into an optimal random forest algorithm model, and judging whether the area is in an urban range.
2. The random forest classification algorithm-based city range extraction method as claimed in claim 1, wherein the training samples are selected from the city pixels and the non-city pixels in the sample area, comprising the steps of:
training samples are sampled hierarchically based on administrative division boundaries.
3. The random forest classification algorithm-based city range extraction method as claimed in claim 2, wherein training samples are hierarchically sampled based on administrative division boundaries, and further comprising the steps of:
generating a random number with a value of 0-1;
obtaining a ratio value between the number of selected samples in each layer and the total number of the urban pixels and the non-urban pixels;
and if the random number is smaller than the proportion value, taking the selected sample as a training sample.
4. The urban area extraction method based on the random forest classification algorithm according to claim 1, wherein before the urban pixels and the non-urban pixels in the sample area are obtained according to the night light remote sensing data and the vegetation index data of the sample area, the method further comprises the following steps:
and removing water pixels in the night lamplight remote sensing image according to the water distribution data.
5. The utility model provides a city scope extraction element based on random forest classification algorithm which characterized in that includes:
the first data acquisition module is used for acquiring a night lamplight remote sensing image and a vegetation index image of the sample area;
the second data acquisition module is used for acquiring night light remote sensing data and vegetation index data of the sample area and acquiring a night light city index of the sample area according to the night light remote sensing data and the vegetation index data;
the system comprises a sample processing module and a random forest training module, wherein the sample processing module is used for acquiring urban pixels and non-urban pixels in a sample area according to night light remote sensing data and vegetation index data of the sample area, the area with higher night light remote sensing data and lower vegetation index data is an urban pixel, and the area with higher vegetation index data is a non-urban pixel, and the random forest training module comprises:
the pixel distinguishing unit is used for judging whether the night lamplight remote sensing data in the sample area is larger than a first set threshold value or not and whether the vegetation index data in the pixel is smaller than a second set threshold value or not, if so, the pixel is an urban pixel, and if the vegetation index data in a certain pixel is larger than the second set threshold value, the pixel is a non-urban pixel;
the random forest training module is used for selecting training samples from the urban pixels and the non-urban pixels in the sample area, and establishing and training an optimal random forest algorithm model according to the selected training samples and the night light remote sensing data, vegetation index data and night light urban indexes of the sample area, wherein the selected training samples and the night light remote sensing data, vegetation index data and night light urban indexes of the sample area are used as input samples of the optimal random forest algorithm model, and the urban pixels and the non-urban pixels of the sample area are used as output samples of the optimal random forest algorithm model;
the third data acquisition module is used for acquiring a night light remote sensing image and a vegetation index image of the area to be identified;
and the city range judging module is used for inputting the night light remote sensing image and the vegetation index image of the area to be identified into an optimal random forest algorithm model and judging whether the area is a city range.
6. The random forest classification algorithm-based city range extraction device according to claim 5, further comprising:
and the water body removing module is used for removing water body pixels in the night lamplight remote sensing image according to the water body distribution data.
7. A computer-readable medium having a computer program stored thereon, characterized in that:
the computer program when executed by a processor implements a city range extraction method based on a random forest classification algorithm as claimed in any one of claims 1 to 4.
8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein:
the processor, when executing the computer program, implements a city range extraction method based on a random forest classification algorithm as claimed in any one of claims 1 to 4.
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 CN109522788A (en) | 2019-03-26 |
| CN109522788B true 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) |
Families Citing this family (39)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| CN111476308A (en) * | 2020-04-10 | 2020-07-31 | 高崟 | Remote sensing image classification method and device based on prior geometric constraint and electronic equipment |
| 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 |
| 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 |
| CN111832575A (en) * | 2020-07-16 | 2020-10-27 | 黄河勘测规划设计研究院有限公司 | Water surface area extraction method and device based on remote sensing image |
| 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 |
| 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 |
| 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 |
| 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 |
| 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 |
| 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 |
| CN113989657A (en) * | 2021-10-11 | 2022-01-28 | 中国测绘科学研究院 | Method and device for detecting farmland range change based on invariant information sample screening |
| CN114266138B (en) * | 2021-11-29 | 2022-09-16 | 西南大学 | Method for recognizing and verifying urban edge area by using cloud data |
| 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 |
| 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 |
| 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 |
| 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 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8306942B2 (en) * | 2008-05-06 | 2012-11-06 | Lawrence Livermore National Security, Llc | Discriminant forest classification method and system |
| US8224097B2 (en) * | 2008-06-12 | 2012-07-17 | Sri International | Building segmentation for densely built urban regions using aerial LIDAR data |
| 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 |
| 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 |
| US11315045B2 (en) * | 2016-12-29 | 2022-04-26 | Intel Corporation | Entropy-based weighting in random forest models |
| 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 (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
Non-Patent Citations (4)
| Title |
|---|
| Classification of Urban Environment Using Feature Extraction and Random Forest;Camila Souza dos Anjos等;《IEEE》;20171204;第1205-1208页 * |
| 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;郜燕芳等;《冰川冻土》;20180831;第40卷(第4期);第828-836页 * |
| 基于随机森林的遥感干旱监测模型的构建;沈润平等;《地球信息科学》;20170131;第19卷(第1期);第125-133页 * |
| 面向对象随机森林方法在湿地植被分类的应用;李方方等;《遥感信息》;20180228;第33卷(第1期);第111-116页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109522788A (en) | 2019-03-26 |
| WO2020063461A1 (en) | 2020-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109522788B (en) | City range extraction method and device based on random forest classification algorithm and electronic equipment | |
| CN110263717B (en) | Method for determining land utilization category of street view image | |
| CN109840553B (en) | Extraction method and system of cultivated land crop type, storage medium and electronic equipment | |
| CN109472393B (en) | Spatial downscaling precipitation data detection method and device and electronic equipment | |
| CN112164062A (en) | Wasteland information extraction method and device based on remote sensing time sequence analysis | |
| Zhang et al. | Estimation of forest leaf area index using height and canopy cover information extracted from unmanned aerial vehicle stereo imagery | |
| CN109522516B (en) | Soil humidity detection method and device based on random forest regression algorithm and electronic equipment | |
| CN104881865A (en) | Forest disease and pest monitoring and early warning method and system based on unmanned plane image analysis | |
| CN111340826A (en) | Single tree crown segmentation algorithm for aerial image based on superpixels and topological features | |
| CN111028255A (en) | Farmland area pre-screening method and device based on prior information and deep learning | |
| WO2020063529A1 (en) | Precipitation data detection method and apparatus based on random forests algorithm and electronic device | |
| Kang et al. | Identifying tree crown areas in undulating eucalyptus plantations using JSEG multi-scale segmentation and unmanned aerial vehicle near-infrared imagery | |
| Grigillo et al. | Automated building extraction from IKONOS images in suburban areas | |
| Fang et al. | An improved tree crown delineation method based on live crown ratios from airborne LiDAR data | |
| CN111914767A (en) | Scattered-pollution enterprise detection method and system based on multi-source remote sensing data | |
| CN112053371A (en) | Water body extraction method and device in remote sensing image | |
| Miao et al. | Measurement method of maize morphological parameters based on point cloud image conversion | |
| Zhang et al. | UAV‐derived imagery for vegetation structure estimation in rangelands: validation and application | |
| CN103065296B (en) | High-resolution remote sensing image residential area extraction method based on edge feature | |
| CN115760885B (en) | High-closure-degree wetland forest parameter extraction method based on consumer-level unmanned aerial vehicle image | |
| Liu et al. | Tree Crown width estimation, using discrete airborne LiDAR data | |
| Tan et al. | Automatic extraction of built-up area based on deep convolution neural network | |
| WO2018098736A1 (en) | Remote sensing imaging-based cyanobacteria monitoring method and device | |
| AU2014259264B2 (en) | Inter-tidal mapping | |
| Chen et al. | Improving fractional vegetation cover estimation with shadow effects using high dynamic range images |
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 |
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 |
|
| CP01 | Change in the name or title of a patent holder |