CN114677589A - City functional area identification method, device, equipment and medium based on remote sensing interpretation - Google Patents
City functional area identification method, device, equipment and medium based on remote sensing interpretation Download PDFInfo
- Publication number
- CN114677589A CN114677589A CN202210275644.6A CN202210275644A CN114677589A CN 114677589 A CN114677589 A CN 114677589A CN 202210275644 A CN202210275644 A CN 202210275644A CN 114677589 A CN114677589 A CN 114677589A
- Authority
- CN
- China
- Prior art keywords
- area
- remote sensing
- functional area
- city
- city functional
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method, a device, equipment and a medium for identifying an urban functional area based on remote sensing interpretation, wherein the method comprises the following steps: constructing a remote sensing sample library based on the homeland survey data and the remote sensing image data of the target area; the homeland survey data comprises vector data for marking city functional areas of the target area; training the deep learning model by utilizing a remote sensing sample library based on the deep learning model to obtain a recognition model; based on a map service protocol constructed with a user terminal in advance, identifying the area selection operation of a user in a target area through the user terminal, and obtaining an area to be identified according to the area selection operation; and identifying the city functional area of the area to be identified through the identification model to obtain an identification result, and feeding the identification result back to the user terminal. The method has the advantages of short time consumption and higher marking accuracy when the remote sensing sample library is constructed, so that the accuracy of identifying the urban functional area is improved.
Description
Technical Field
The invention relates to the technical field of geographic information identification, in particular to a method, a device, terminal equipment and a computer readable storage medium for identifying an urban functional area based on remote sensing interpretation.
Background
With the development of remote sensing technology, the remote sensing image interpretation is applied more and more widely, for example, the remote sensing image interpretation technology is used for identifying urban functional areas.
In order to further improve the efficiency and accuracy of remote sensing image interpretation and thus realize faster and more accurate identification of urban functional areas, the prior art generally adopts a method of combining remote sensing image interpretation technology and deep learning, and firstly constructs a remote sensing image sample library, and then trains a deep learning network by using the constructed remote sensing image sample library so as to obtain a deep learning model capable of identifying urban functional areas. However, the method in the prior art needs a lot of manual labeling work when constructing the remote sensing image sample library, the time consumption is long, the accuracy of manual labeling is low, and it is difficult to ensure that the deep learning model obtained by training has high recognition accuracy, so that the accuracy of recognition of the urban functional area is low.
Disclosure of Invention
The invention provides a method, a device, equipment and a medium for identifying a city functional area based on remote sensing interpretation, which are used for solving the problem of lower identification accuracy of the city functional area in the prior art, and utilize homeland survey data to assist the construction of a remote sensing sample library.
In order to solve the above technical problem, a first aspect of an embodiment of the present invention provides a method for identifying an urban functional area based on remote sensing interpretation, including:
constructing a remote sensing sample library of a target area based on homeland survey data and remote sensing image data of the target area; the homeland survey data comprises vector data for marking city functional areas of the target area;
training the deep learning model by utilizing the remote sensing sample library based on a preset deep learning model to obtain an identification model for identifying the urban functional area of the target area;
identifying the area selection operation of the user in the target area through the user terminal based on a map service protocol constructed with the user terminal in advance, and obtaining an area to be identified according to the area selection operation;
and identifying the city functional area of the area to be identified through the identification model to obtain an identification result, and feeding the identification result back to the user terminal.
As a preferred scheme, before constructing a remote sensing sample library of a target area based on homeland survey data and remote sensing image data of the target area, the method further comprises:
And preprocessing the homeland survey data.
As a preferred scheme, the preprocessing the homeland survey data includes:
classifying the city functional areas of the target area according to the remote sensing image data to obtain at least one first city functional area category;
and mapping at least one second city functional area category preset in the homeland survey data according to the at least one first city functional area category so as to map the second city functional area category to the first city functional area category.
As a preferred scheme, the preprocessing the homeland survey data further includes:
when the city functional area category of a first pattern spot in the homeland survey data is different from the city functional area category of a second pattern spot in the remote sensing image data, replacing the city functional area category of the first pattern spot with the city functional area category of the second pattern spot; wherein the target area comprises at least one pattern spot, and the position of the first pattern spot in the target area is the same as the position of the second pattern spot in the target area;
when a plot containing a plurality of patches exists in the homeland survey data and the set of the city functional area categories of the plurality of patches is the first city functional area category, taking the first city functional area category corresponding to the set of the city functional area categories of the plurality of patches as the city functional area category of the plot; wherein the target area comprises at least one parcel;
And when the boundary of the first pattern spot in the homeland survey data is inconsistent with the boundary of the second pattern spot in the remote sensing image data, adjusting the boundary of the first pattern spot according to the boundary of the second pattern spot.
As a preferred scheme, before training the deep learning model by using the remote sensing sample library based on a preset deep learning model to obtain an identification model for identifying the urban functional area of the target area, the method further comprises:
performing data enhancement processing on the image data in the remote sensing sample library through data enhancement operation;
then, based on the preset deep learning model, training the deep learning model by using the remote sensing sample library to obtain an identification model for identifying the urban functional area of the target area, specifically:
training the deep learning model by using a remote sensing sample library subjected to data enhancement processing based on a preset deep learning model to obtain the recognition model;
wherein the data enhancement operations include at least a resampling operation, a rotation operation, a cropping operation, a scaling operation, a flipping operation, a shifting operation, and a color space conversion operation.
Preferably, the method obtains the deep learning model in advance through the following steps:
based on the DeepLabV3+ semantic segmentation model, a residual error network is used as an encoder of the DeepLabV3+ semantic segmentation model, and model parameters of the DeepLabV3+ semantic segmentation model are pre-trained by utilizing an ImageNet data set to obtain the deep learning model.
As a preferred scheme, the identifying the city functional area of the area to be identified through the identification model to obtain an identification result, specifically:
and identifying the city functional area of the area to be identified by adopting a window cutting algorithm through the identification model to obtain an identification result.
The second aspect of the embodiments of the present invention provides a city functional area identification based on remote sensing interpretation, including:
the sample library construction module is used for constructing a remote sensing sample library of the target area based on the homeland survey data and the remote sensing image data of the target area; the homeland survey data comprises vector data for marking city functional areas of the target area;
the identification model acquisition module is used for training the deep learning model by utilizing the remote sensing sample library based on a preset deep learning model to obtain an identification model for identifying the urban functional area of the target area;
The area to be identified acquisition module is used for identifying the area selection operation of the user in the target area through the user terminal based on a map service protocol which is constructed with the user terminal in advance, and acquiring the area to be identified according to the area selection operation;
and the identification module is used for identifying the city functional area of the area to be identified through the identification model, obtaining an identification result and feeding the identification result back to the user terminal.
A third aspect of embodiments of the present invention provides a terminal device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor, when executing the computer program, implements the method for identifying a functional area of a city based on remote sensing interpretation according to any one of the first aspect.
A fourth aspect of the embodiments of the present invention provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program, where when the computer program runs, a device on which the computer-readable storage medium is located is controlled to execute the method for identifying a functional area of a city based on remote sensing interpretation according to any one of the first aspects.
Compared with the prior art, the method and the device have the advantages that the homeland survey data is used for assisting the construction of the remote sensing sample library, and the homeland survey data comprises vector data for marking the urban functional area of the target area, so that a large amount of marking work is not needed manually, time consumption is short, marking accuracy is high, the deep learning model obtained through training is guaranteed to have high recognition accuracy, and accuracy of recognizing the urban functional area is improved.
Drawings
Fig. 1 is a schematic flow diagram of a method for identifying an urban functional area based on remote sensing interpretation according to an embodiment of the present invention;
fig. 2 is a schematic diagram of semantic segmentation of a remote sensing image according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a mapping relationship of city functional area categories according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a DeepLabV3+ semantic segmentation model provided by an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an urban functional area identification device based on remote sensing interpretation according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, in a first aspect, the embodiment of the present invention provides a method for identifying an urban functional area based on remote sensing interpretation, including steps S1 to S4, which are specifically as follows:
step S1, constructing a remote sensing sample library of the target area based on the homeland survey data and the remote sensing image data of the target area; the homeland survey data comprises vector data for marking city functional areas of the target area;
step S2, training the deep learning model by using the remote sensing sample library based on a preset deep learning model to obtain an identification model for identifying the urban functional area of the target area;
step S3, based on a map service protocol constructed with a user terminal in advance, identifying the area selection operation of the user in the target area through the user terminal, and obtaining the area to be identified according to the area selection operation;
and step S4, recognizing the city functional area of the area to be recognized through the recognition model to obtain a recognition result, and feeding the recognition result back to the user terminal.
It should be noted that, the homeland survey is used as a basic survey work for the unified survey and monitoring of natural resources, and the land utilization basic data in the national range is formed by comprehensively surveying the land resource use condition of the country, so as to grasp the detailed and accurate state of homeland utilization and the natural resource change condition. Preferably, the third national homeland survey data is adopted as the homeland survey data in the embodiment of the present invention.
Further, the embodiment of the invention utilizes the remote sensing sample library to train the deep learning model based on a preset deep learning model, and obtains an identification model for identifying the urban functional area of the target area. In the process of constructing the remote sensing sample library, the embodiment of the invention utilizes the homeland survey data to assist the city functional area class labeling work of the remote sensing image data, so that the remote sensing image data is endowed with semantic information, when the deep learning model is trained by utilizing the remote sensing sample library, the deep learning model can carry out semantic segmentation on the remote sensing image according to the semantic information in the remote sensing image data, as shown in figure 2, the semantic information of each land block in a target area is determined, and thus, an identification model which can be used for identifying the city functional area of the target area is obtained.
Further, based on a map service protocol constructed with the user terminal in advance, the user terminal identifies the area selection operation of the user in the target area, and the area to be identified is obtained according to the area selection operation. It should be noted that the map service is a method for making a map accessible through a network by using a software interface such as SuperMap, ArcGIS, and the like. After the map is made, the map is released to the server site as a service, and the user can use the map service in an application program of the user terminal. According to the embodiment of the invention, the region selection operation of the user in the target region is identified through the user terminal, for example, a polygonal region is drawn on a map, an administrative range is selected or a vector file in a specified range is uploaded, so that the region to be identified can be obtained according to the region selection operation.
Further, the city functional area of the area to be identified is identified through the identification model, an identification result is obtained, and the identification result is fed back to the user terminal.
Preferably, based on a map service protocol, the embodiment of the invention can also share the identification result of the urban functional area of the area to be identified through an intelligent gateway technology, thereby meeting the requirement that different users simultaneously interpret the area to be identified.
As one of the preferred embodiments, in order to further improve the identification accuracy of the urban functional area, the embodiment of the present invention adopts a high-resolution ortho image of 0.5m as the remote sensing image data.
According to the city functional area identification method based on remote sensing interpretation, the construction of the remote sensing sample base is assisted by the homeland survey data, and since the homeland survey data comprise vector data for marking the city functional area of the target area, a large amount of marking work is not needed manually, the time consumption is short, the marking accuracy is high, the deep learning model obtained through training is ensured to have high identification accuracy, and therefore the accuracy for identifying the city functional area is improved.
As a preferred scheme, before constructing a remote sensing sample library of a target area based on homeland survey data and remote sensing image data of the target area, the method further comprises:
And preprocessing the homeland survey data.
It should be noted that, because there may be a difference between a preset city functional area category in the homeland survey data and a city functional area category obtained by identification according to the remote sensing image data, and because the acquisition time of the homeland leveling data may be different from the acquisition time of the remote sensing image data, there may be a difference between a vector boundary in the homeland survey data and a surface feature boundary in the remote sensing image data, in order to improve the construction accuracy of the remote sensing sample library, in the embodiment of the present invention, the homeland survey data is preprocessed before the remote sensing sample library of the target area is constructed based on the homeland survey data and the remote sensing image data of the target area.
As a preferred scheme, the preprocessing the homeland survey data includes:
classifying the city functional areas of the target area according to the remote sensing image data to obtain at least one first city functional area category;
and mapping at least one second city functional area category preset in the homeland survey data according to the at least one first city functional area category so as to map the second city functional area category to the first city functional area category.
Specifically, according to the remote sensing image data, the city functional areas of the target area are classified to obtain at least one first city functional area category, and the first city functional area category is one of 14 types of city functional areas, namely low-density residential land, high-density residential land, commercial service land, industrial land, public service land, education and scientific research land, artificial digging land, structures, forest land, grassland, cultivated land, roads, water surface and bare land. Further, since the degree of refinement of the preset city functional area category in the homeland survey data is greater, according to the at least one first city functional area category, mapping processing is performed on the at least one second city functional area category preset in the homeland survey data, so as to map the second city functional area category to the first city functional area category, a specific mapping relationship is shown in fig. 3, "three-tone" land category in the figure represents the city functional area category set in the third national homeland survey, including a primary category and a secondary category, and "custom category" in the figure represents the first city functional area category obtained by performing category division on the city functional area of the target area according to the remote sensing image data.
As a preferred scheme, the preprocessing the homeland survey data further includes:
when the city functional area category of a first pattern spot in the homeland survey data is different from the city functional area category of a second pattern spot in the remote sensing image data, replacing the city functional area category of the first pattern spot with the city functional area category of the second pattern spot; wherein the target area comprises at least one pattern spot, and the position of the first pattern spot in the target area is the same as the position of the second pattern spot in the target area;
when a plot containing a plurality of patches exists in the homeland survey data and the set of the city functional area categories of the plurality of patches is the first city functional area category, taking the first city functional area category corresponding to the set of the city functional area categories of the plurality of patches as the city functional area category of the plot; wherein the target area comprises at least one parcel;
and when the boundary of the first pattern spot in the homeland survey data is inconsistent with the boundary of the second pattern spot in the remote sensing image data, adjusting the boundary of the first pattern spot according to the boundary of the second pattern spot.
It can be understood that the type of land in the same position may change with the migration of time, so that the situation that the city functional area category of the first pattern spot in the homeland survey data is inconsistent with the city functional area category of the second pattern spot in the remote sensing image data occurs, and the acquisition time of the remote sensing image data is often behind the acquisition time of the homeland survey data, so that when the city functional area category of the first pattern spot in the homeland survey data is inconsistent with the city functional area category of the second pattern spot in the remote sensing image data, the city functional area category of the first pattern spot is replaced by the city functional area category of the second pattern spot in the homeland survey data, so as to update the labeling information of the homeland survey data and ensure the construction accuracy of the remote sensing sample library.
Further, the map spots in the homeland survey data are relatively fine, and one land may contain map spots of a plurality of different city functional area categories, so that it is difficult to determine which city functional area category the land belongs to, therefore, in the embodiment of the present invention, when a land containing a plurality of map spots exists in the homeland survey data, and the set of the city functional area categories of the plurality of map spots is the first city functional area category, the first city functional area category corresponding to the set of the city functional area categories of the plurality of map spots is taken as the city functional area category of the land, for example, a land contains a woodland, a grassland, a playground and a residential land, and the set of the city functional area categories is a higher-level city functional area category, that is, an educational and scientific research land, and then the educational and scientific research land is taken as the city functional area category of the land.
Further, because there is a problem that an object blocks the ground in the remote-sensing image data, which may cause that the boundary of a patch in the remote-sensing image data is inconsistent with the boundary of a patch in the remote-sensing image data, in the embodiment of the present invention, when the boundary of a first patch in the remote-sensing image data is inconsistent with the boundary of a second patch in the remote-sensing image data, the boundary of the first patch is adjusted according to the boundary of the second patch, for example, there is a road in the remote-sensing image data, a part of which is shielded by a tall building, and the road does not embody a tall building part in the remote-sensing image data, so it is necessary to extract a range including the tall building from a vector range of the road in the remote-sensing image data by, for example, a manual correction manner, and modify the range including the tall building into a city functional area category corresponding to the tall building.
In addition, for the part with insufficient fineness marked in the homeland survey data, the vector data of the part is perfected in a manual re-marking mode.
As a preferred scheme, before training the deep learning model by using the remote sensing sample library based on a preset deep learning model to obtain an identification model for identifying the urban functional area of the target area, the method further comprises:
Performing data enhancement processing on the image data in the remote sensing sample library through data enhancement operation;
then, based on the preset deep learning model, training the deep learning model by using the remote sensing sample library to obtain an identification model for identifying the urban functional area of the target area, specifically:
training the deep learning model by using a remote sensing sample library subjected to data enhancement processing based on a preset deep learning model to obtain the recognition model;
wherein the data enhancement operations include at least a resampling operation, a rotation operation, a cropping operation, a scaling operation, a flipping operation, a shifting operation, and a color space conversion operation.
It should be noted that the data enhancement processing is a method of constructing more samples based on a small number of samples to expand the data volume. One picture is enhanced into at least four pictures by rotating, turning, cutting and other operations. If similar operations are carried out on all remote sensing images in the remote sensing sample library, the data in the remote sensing sample library can be increased by four times.
Therefore, in order to fully utilize the constructed remote sensing sample library, the embodiment of the invention performs data enhancement processing on the remote sensing sample library through data enhancement operation, wherein the data enhancement operation at least comprises resampling operation, rotation operation, cutting operation, scaling operation, turning operation, shifting operation and color space conversion operation, and then trains the deep learning model by using the remote sensing sample library after the data enhancement processing based on a preset deep learning model to obtain the identification model.
As one of the preferable embodiments, the embodiment of the invention adopts an online enhancement mode, namely, a data enhancement operation is preposed in the deep learning model, so that the data of the remote sensing sample library is subjected to data enhancement processing in the deep learning model. Because the deep learning model is generally deployed in a server, the problem that the local hard disk space is excessively occupied due to sudden increase of data volume after the data enhancement processing is carried out on the sample library in an online enhancing mode can be solved by adopting an online enhancing mode.
Preferably, the method obtains the deep learning model in advance through the following steps:
based on the DeepLabV3+ semantic segmentation model, a residual error network is used as an encoder of the DeepLabV3+ semantic segmentation model, and model parameters of the DeepLabV3+ semantic segmentation model are pre-trained by utilizing an ImageNet data set to obtain the deep learning model.
It should be noted that, in order to solve the problem of information loss caused by continuous downsampling, the embodiment of the present invention adopts a semantic segmentation model based on deplab V3+, and the structure of the semantic segmentation model is shown in fig. 4, where the deplab V3+ network mainly includes a backbone network, an Aperture Spatial Pyramid Pooling (ASPP) module, and a decoder module, the ASPP is used to further utilize multi-scale information of an object, the decoder module refers to the design of UNet and aims to obtain a fine object boundary, the backbone network of the deplab V3+ is an improved Xception, and the size of a final output feature map is one eighth of the size of an original image.
Specifically, in the embodiment of the present invention, based on the deep labv3+ semantic segmentation model, a residual error network is used as an encoder of the deep labv3+ semantic segmentation model, the residual error network may be, but is not limited to, a ResNet network, and the model parameters of the deep labv3+ semantic segmentation model are pre-trained by using an ImageNet data set to obtain the deep learning model. In addition, as one of the preferred embodiments, the embodiment of the invention utilizes a polynomial error attenuation learning strategy to carry out fine adjustment on the preset initial learning rate of the DeepLabV3+ semantic segmentation model.
As a preferred scheme, the identifying the city functional area of the area to be identified through the identification model to obtain an identification result, specifically:
and identifying the city functional area of the area to be identified by adopting a window cutting algorithm through the identification model to obtain an identification result.
Referring to fig. 5, a second aspect of the embodiments of the present invention provides an apparatus for identifying an urban functional area based on remote sensing interpretation, including:
the sample library construction module 501 is used for constructing a remote sensing sample library of a target area based on homeland survey data and remote sensing image data of the target area; the homeland survey data comprise vector data for marking city functional areas of the target area;
The identification model acquisition module 502 is used for training the deep learning model by utilizing the remote sensing sample library based on a preset deep learning model to obtain an identification model for identifying the urban functional area of the target area;
a to-be-identified region acquisition module 503, configured to identify, by a user terminal, a region selection operation of a user in the target region based on a map service protocol pre-established with the user terminal, and acquire a to-be-identified region according to the region selection operation;
the identifying module 504 is configured to identify the city functional area of the area to be identified through the identification model, obtain an identification result, and feed the identification result back to the user terminal.
Preferably, the apparatus further comprises a preprocessing module configured to:
the method comprises the steps of preprocessing homeland survey data based on the homeland survey data and remote sensing image data of a target area before constructing a remote sensing sample library of the target area.
As a preferred scheme, the preprocessing module is configured to preprocess the homeland survey data, and includes:
classifying the city functional areas of the target area according to the remote sensing image data to obtain at least one first city functional area category;
And mapping at least one second city functional area category preset in the homeland survey data according to the at least one first city functional area category so as to map the second city functional area category to the first city functional area category.
As a preferred scheme, the preprocessing module is configured to preprocess the homeland survey data, and further includes:
when the city functional area category of a first pattern spot in the homeland survey data is different from the city functional area category of a second pattern spot in the remote sensing image data, replacing the city functional area category of the first pattern spot with the city functional area category of the second pattern spot; wherein the target area comprises at least one pattern spot, and the position of the first pattern spot in the target area is the same as the position of the second pattern spot in the target area;
when a plot containing a plurality of patches exists in the homeland survey data and the set of the city functional area categories of the plurality of patches is the first city functional area category, taking the first city functional area category corresponding to the set of the city functional area categories of the plurality of patches as the city functional area category of the plot; wherein the target area comprises at least one parcel;
And when the boundary of the first pattern spot in the homeland survey data is inconsistent with the boundary of the second pattern spot in the remote sensing image data, adjusting the boundary of the first pattern spot according to the boundary of the second pattern spot.
Preferably, the apparatus further comprises a data enhancement module, configured to:
performing data enhancement processing on the image data in the remote sensing sample library through data enhancement operation;
then, the recognition model obtaining module 502 is configured to train the deep learning model by using the remote sensing sample library based on a preset deep learning model, and obtain a recognition model for recognizing the urban functional area of the target area, specifically:
training the deep learning model by using a remote sensing sample library subjected to data enhancement processing based on a preset deep learning model to obtain the recognition model;
wherein the data enhancement operation includes at least a resampling operation, a rotation operation, a clipping operation, a scaling operation, a flipping operation, a shifting operation, and a color space conversion operation.
Preferably, the apparatus further includes a deep learning model obtaining module, configured to:
Based on the DeepLabV3+ semantic segmentation model, a residual error network is used as an encoder of the DeepLabV3+ semantic segmentation model, and model parameters of the DeepLabV3+ semantic segmentation model are pre-trained by utilizing an ImageNet data set to obtain the deep learning model.
As a preferred scheme, the identifying module 504 is configured to identify the city functional area of the area to be identified through the identification model, and obtain an identification result, specifically:
and identifying the city functional area of the area to be identified by adopting a window cutting algorithm through the identification model to obtain an identification result.
It should be noted that, the device for identifying an urban functional area based on remote sensing interpretation provided in the embodiment of the present invention can implement all the processes of the method for identifying an urban functional area based on remote sensing interpretation described in any of the above embodiments, and the functions and implemented technical effects of each module in the device are respectively the same as those of the method for identifying an urban functional area based on remote sensing interpretation described in the above embodiment, and are not described herein again.
A third aspect of the embodiments of the present invention provides a terminal device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor, when executing the computer program, implements the method for identifying a city functional area based on remote sensing interpretation as described in any embodiment of the first aspect.
The terminal device can be a desktop computer, a notebook, a palm computer, a cloud server and other computing devices. The terminal device may include, but is not limited to, a processor, a memory. The terminal device may also include input and output devices, network access devices, buses, etc.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like, said processor being the control center of said terminal device, and various interfaces and lines are used to connect the various parts of the whole terminal device.
The memory may be used to store the computer programs and/or modules, and the processor may implement various functions of the terminal device by executing or executing the computer programs and/or modules stored in the memory and calling data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, etc. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
A fourth aspect of the embodiments of the present invention provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program, where when the computer program runs, a device on which the computer-readable storage medium is located is controlled to execute the method for identifying a functional city area based on remote sensing interpretation according to any one of the embodiments of the first aspect.
Through the above description of the embodiments, those skilled in the art will clearly understand that the present invention may be implemented by software plus a necessary hardware platform, and may also be implemented by hardware entirely. With this understanding in mind, all or part of the technical solutions of the present invention that contribute to the background art may be embodied in the form of a software product, which can be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods according to the embodiments or some parts of the embodiments of the present invention.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (10)
1. A city functional area identification method based on remote sensing interpretation is characterized by comprising the following steps:
constructing a remote sensing sample library of a target area based on homeland survey data and remote sensing image data of the target area; the homeland survey data comprises vector data for marking city functional areas of the target area;
training the deep learning model by utilizing the remote sensing sample library based on a preset deep learning model to obtain an identification model for identifying the urban functional area of the target area;
identifying the area selection operation of the user in the target area through the user terminal based on a map service protocol constructed with the user terminal in advance, and obtaining an area to be identified according to the area selection operation;
and identifying the city functional area of the area to be identified through the identification model to obtain an identification result, and feeding the identification result back to the user terminal.
2. The method for identifying an urban functional area based on remote sensing interpretation according to claim 1, wherein before constructing a remote sensing sample library of a target area based on homeland survey data and remote sensing image data of the target area, the method further comprises:
And preprocessing the homeland survey data.
3. The city functional area identification method based on remote sensing interpretation as recited in claim 2, wherein the preprocessing of the homeland survey data comprises:
classifying the city functional areas of the target area according to the remote sensing image data to obtain at least one first city functional area category;
and mapping at least one second city functional area category preset in the homeland survey data according to the at least one first city functional area category so as to map the second city functional area category to the first city functional area category.
4. The remote sensing interpretation-based urban functional area identification method according to claim 3, wherein the preprocessing of the homeland survey data further comprises:
when the city functional area category of a first pattern spot in the homeland survey data is different from the city functional area category of a second pattern spot in the remote sensing image data, replacing the city functional area category of the first pattern spot with the city functional area category of the second pattern spot; wherein the target area comprises at least one pattern spot, and the position of the first pattern spot in the target area is the same as the position of the second pattern spot in the target area;
When a land parcel comprising a plurality of patches exists in the homeland survey data and the set of the city functional area categories of the plurality of patches is the first city functional area category, taking the first city functional area category corresponding to the set of the city functional area categories of the plurality of patches as the city functional area category of the land parcel; wherein the target area comprises at least one parcel;
and when the boundary of the first pattern spot in the homeland survey data is inconsistent with the boundary of the second pattern spot in the remote sensing image data, adjusting the boundary of the first pattern spot according to the boundary of the second pattern spot.
5. The method for identifying urban functional areas based on remote sensing interpretation according to claim 4, wherein before the method trains the deep learning model by using the remote sensing sample library based on a preset deep learning model to obtain an identification model for identifying the urban functional areas of the target area, the method further comprises the following steps:
performing data enhancement processing on the image data in the remote sensing sample library through data enhancement operation;
then, based on the preset deep learning model, training the deep learning model by using the remote sensing sample library to obtain an identification model for identifying the urban functional area of the target area, specifically:
Training the deep learning model by using a remote sensing sample library subjected to data enhancement processing based on a preset deep learning model to obtain the recognition model;
wherein the data enhancement operations include at least a resampling operation, a rotation operation, a cropping operation, a scaling operation, a flipping operation, a shifting operation, and a color space conversion operation.
6. The city functional area identification method based on remote sensing interpretation as recited in claim 5, wherein the method obtains the deep learning model in advance by the following steps:
based on the DeepLabV3+ semantic segmentation model, a residual error network is used as an encoder of the DeepLabV3+ semantic segmentation model, and model parameters of the DeepLabV3+ semantic segmentation model are pre-trained by using an ImageNet data set to obtain the deep learning model.
7. The city functional area identification method based on remote sensing interpretation as claimed in claim 6, wherein the city functional area of the area to be identified is identified through the identification model to obtain an identification result, specifically:
and identifying the city functional area of the area to be identified by adopting a window cutting algorithm through the identification model to obtain an identification result.
8. A city functional area recognition device based on remote sensing interpretation is characterized by comprising:
the sample library construction module is used for constructing a remote sensing sample library of the target area based on the homeland survey data and the remote sensing image data of the target area; the homeland survey data comprise vector data for marking city functional areas of the target area;
the identification model acquisition module is used for training the deep learning model by utilizing the remote sensing sample library based on a preset deep learning model to obtain an identification model for identifying the urban functional area of the target area;
the area to be identified acquisition module is used for identifying the area selection operation of the user in the target area through the user terminal based on a map service protocol which is constructed with the user terminal in advance, and acquiring the area to be identified according to the area selection operation;
and the identification module is used for identifying the city functional area of the area to be identified through the identification model, obtaining an identification result and feeding the identification result back to the user terminal.
9. A terminal device, characterized by comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor when executing the computer program implements the remote sensing interpretation-based city functional area identification method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored computer program, wherein when the computer program runs, the computer-readable storage medium is controlled to implement the method for identifying a functional city area based on remote sensing interpretation according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210275644.6A CN114677589B (en) | 2022-03-18 | 2022-03-18 | City functional area identification method, device, equipment and medium based on remote sensing interpretation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210275644.6A CN114677589B (en) | 2022-03-18 | 2022-03-18 | City functional area identification method, device, equipment and medium based on remote sensing interpretation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114677589A true CN114677589A (en) | 2022-06-28 |
| CN114677589B CN114677589B (en) | 2022-12-06 |
Family
ID=82074786
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210275644.6A Active CN114677589B (en) | 2022-03-18 | 2022-03-18 | City functional area identification method, device, equipment and medium based on remote sensing interpretation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114677589B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115937439A (en) * | 2023-03-02 | 2023-04-07 | 航天宏图信息技术股份有限公司 | Method and device for constructing three-dimensional model of urban building and electronic equipment |
| CN117575979A (en) * | 2023-08-01 | 2024-02-20 | 广东省国土资源测绘院 | Remote sensing image change detection method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101510374A (en) * | 2009-03-12 | 2009-08-19 | 中国科学院遥感应用研究所 | Method for automatically updating land-use map based on historical data and remote sensing data |
| CN105279223A (en) * | 2015-10-20 | 2016-01-27 | 西南林业大学 | Computer automatic interpretation method for remote sensing image |
| CN109325085A (en) * | 2018-08-08 | 2019-02-12 | 中南大学 | A kind of urban land identification of function and change detecting method |
| CN109446992A (en) * | 2018-10-30 | 2019-03-08 | 苏州中科天启遥感科技有限公司 | Remote sensing image building extracting method and system, storage medium, electronic equipment based on deep learning |
| CN110348506A (en) * | 2019-07-03 | 2019-10-18 | 广州大学 | Land use classes method, storage medium and calculating equipment based on remote sensing images |
| CN112733781A (en) * | 2021-01-20 | 2021-04-30 | 中国科学院地理科学与资源研究所 | City functional area identification method combining POI data, storage medium and electronic equipment |
-
2022
- 2022-03-18 CN CN202210275644.6A patent/CN114677589B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101510374A (en) * | 2009-03-12 | 2009-08-19 | 中国科学院遥感应用研究所 | Method for automatically updating land-use map based on historical data and remote sensing data |
| CN105279223A (en) * | 2015-10-20 | 2016-01-27 | 西南林业大学 | Computer automatic interpretation method for remote sensing image |
| CN109325085A (en) * | 2018-08-08 | 2019-02-12 | 中南大学 | A kind of urban land identification of function and change detecting method |
| CN109446992A (en) * | 2018-10-30 | 2019-03-08 | 苏州中科天启遥感科技有限公司 | Remote sensing image building extracting method and system, storage medium, electronic equipment based on deep learning |
| CN110348506A (en) * | 2019-07-03 | 2019-10-18 | 广州大学 | Land use classes method, storage medium and calculating equipment based on remote sensing images |
| CN112733781A (en) * | 2021-01-20 | 2021-04-30 | 中国科学院地理科学与资源研究所 | City functional area identification method combining POI data, storage medium and electronic equipment |
Non-Patent Citations (1)
| Title |
|---|
| 吴施瑶 等: "基于遥感影像的城市功能区识别", 《信息与电脑(理论版)》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115937439A (en) * | 2023-03-02 | 2023-04-07 | 航天宏图信息技术股份有限公司 | Method and device for constructing three-dimensional model of urban building and electronic equipment |
| CN117575979A (en) * | 2023-08-01 | 2024-02-20 | 广东省国土资源测绘院 | Remote sensing image change detection method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114677589B (en) | 2022-12-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114677589B (en) | City functional area identification method, device, equipment and medium based on remote sensing interpretation | |
| WO2017020528A1 (en) | Lane line recognition modeling method, apparatus, storage medium, and device, recognition method and apparatus, storage medium, and device | |
| Kim et al. | CityCraft: 3D virtual city creation from a single image | |
| CN111091167B (en) | Mark recognition training data synthesis method and device, electronic equipment and storage medium | |
| DE202012013426U1 (en) | Integrate maps and street views | |
| CN110059151B (en) | Map rendering method, map rendering device, map server, and storage medium | |
| CN112712138B (en) | Image processing method, device, equipment and storage medium | |
| CN113223042B (en) | Intelligent acquisition method and equipment for remote sensing image deep learning sample | |
| CN114463623A (en) | Method and device for detecting farmland change based on multi-scale remote sensing image | |
| CN113192192A (en) | Live-action three-dimensional digital twin channel scene construction method | |
| CN115457408A (en) | Land monitoring method and device, electronic equipment and medium | |
| CN115641415A (en) | Method, device, equipment and medium for generating three-dimensional scene based on satellite image | |
| CN107578003A (en) | A kind of remote sensing images transfer learning method based on GEOGRAPHICAL INDICATION image | |
| CN111028244A (en) | Remote sensing image semantic segmentation method based on super-pixel under condition of known sample imbalance | |
| CN117011413B (en) | Road image reconstruction method, device, computer equipment and storage medium | |
| CN111191530B (en) | Town and bare land extraction method, device and equipment based on sentinel remote sensing data | |
| CN111325126B (en) | Interpretation method and device of remote sensing image | |
| CN116597317A (en) | Remote sensing image change detection data generation method, device, equipment and medium | |
| CN106908797A (en) | X-band areal rainfall radar data extracting method based on GIS | |
| Ruiz-Lendínez et al. | Method for an automatic alignment of imagery and vector data applied to cadastral information in Poland | |
| Kumar | Solar potential analysis of rooftops using satellite imagery | |
| CN116310628A (en) | Token mask mechanism-based large-scale village-in-city extraction method | |
| CN116385716A (en) | Three-dimensional map ground object data automatic production method based on remote sensing map | |
| CN113538679A (en) | Mixed real-scene three-dimensional channel scene construction method | |
| CN111062863A (en) | Method, device, equipment and storage medium for binding 3D model and longitude and latitude coordinates |
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 |