Urban impervious bed extraction method and system, storage medium and electronic equipment

Abstract

The invention provides a city impervious layer extraction method and system, a storage medium and electronic equipment. The method comprises the following steps: acquiring a multi-source and multi-mode remote sensing image of a city; the multi-source and multi-mode remote sensing image is preprocessed; stray light removal is carried out on the preprocessed multi-source and multi-mode remote sensing image; invalid value removal is carried out on the multi-source multi-mode remote sensing image after stray light removal; carrying out data normalization processing on the multi-source and multi-mode remote sensing image after the invalid value is removed; and constructing an urban impervious layer extraction index model based on the multi-source and multi-mode remote sensing image after data normalization so as to realize urban impervious layer extraction based on the urban impervious layer extraction index model. According to the urban impervious layer extraction method and system, the storage medium and the electronic equipment, the urban impervious layer extraction index model is constructed based on the multi-source and multi-mode remote sensing image, so that rapid extraction of global urban impervious layers can be rapidly completed.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for extracting urban impermeable layers, a storage medium and electronic equipment. Background Art

[0002] Cities are not only places where people live, but also centers of economic, cultural and political activities. They are the product of a certain stage of social development. Different land use types such as residential areas, streets, and public facilities together constitute cities. In geography, cities are defined as irregular areas with a certain population and space. With the acceleration of urbanization and the rapid growth of population, the rapid expansion of impervious layers has become a significant sign of urbanization, and has also caused a series of environmental problems, including urban heat island effect, environmental degradation, floods and traffic congestion. The growth of urban impervious layers has resulted in the inability of surface water to penetrate downward, which in turn has increased the duration, intensity and flow of urban surface runoff. This change has a direct impact on the spread of non-point source pollutants and poses severe challenges to the land and water environment. In addition, the area and spatial distribution of impervious layers will have a significant impact on urban climate, which may change the sensible heat flux and latent heat flux in the urban canopy and boundary layer, thereby causing problems such as the urban heat island effect. At the same time, the increase in impervious layers is often accompanied by drastic changes in land cover and land use, which has a profound impact on the regional environment. Therefore, monitoring and analysis of impermeable layers are extremely important for environmental management, such as regulation of surface runoff and groundwater circulation, and urban planning, such as infrastructure construction and sustainable urban growth.

[0003] The traditional method of mapping urban impervious layers mainly relies on artificial ground measurements, especially ground surveys based on the Global Positioning System (GPS). Although this method can provide relatively accurate impervious layer information, it is not suitable for large-scale urban impervious layer mapping due to its long time consumption and high cost. Remote sensing technology has unique advantages such as all-weather, all-round, multi-platform and multi-temporal. These characteristics provide efficient and convenient data resources for urban problem research, especially the study of the spatial morphology and changes of urban impervious layers, thereby significantly reducing the investment of time, manpower and material resources. This technology uses sensors to remotely and non-contactly detect the electromagnetic radiation or reflection characteristics of the ground object, and then obtain the spectral reflectance curve of the ground object. This is based on the difference in the reflection ability (reflectivity) of different ground objects to the incident wavelength electromagnetic radiation to distinguish and identify various ground object types.

[0004] Although the above-mentioned ground survey methods are accurate, they are time-consuming and labor-intensive, and it is difficult to achieve rapid monitoring over a large area. However, light satellite data has great potential in extracting impervious layers due to its unique nighttime light detection capabilities. By capturing the distribution of urban lights at night, light satellites can indirectly reflect the spatial distribution and density of urban impervious layers, providing a new and efficient monitoring method for urban planners, ecological and environmental researchers, and policy makers. Therefore, light satellite data is used to achieve rapid extraction of large-scale urban impervious layers to meet the urgent needs of modern urban management and research.

[0005] The commonly used method for extracting urban impervious layers from light satellites is image classification. Image classification is mainly based on the differences in spectral and geometric characteristics between urban impervious layers and other objects in remote sensing images. This classification can be performed at two different scales: pixel scale and object scale. At the pixel scale, classification is based on the spectral information of a single pixel; at the object scale, classification is based on the overall characteristics of a larger object composed of multiple pixels. Image classification accurately distinguishes and identifies urban impervious layers at different scales by identifying and analyzing the unique characteristics of different objects in remote sensing images.

[0006] Image classification methods are mainly based on the differences in spectral and geometric characteristics between urban impervious layers and other types of land objects in remote sensing images. According to the different spatial resolutions, classification is mainly carried out at the pixel scale and object scale. With the rise of machine learning, deep learning and other algorithms, more and more algorithms are being used to extract urban impervious layers. However, the light satellite spillover effect is a common problem in night light remote sensing data. When satellites capture night lights, halos or light pollution may appear in the area around the strong light source. This is the so-called spillover effect. The light satellite spillover effect often leads to over-extraction problems in the process of directly using image classification methods to extract impervious layers.

[0007] In addition, many current models often require thresholds to complete the final extraction of urban impermeable layers, and the choice of thresholds will also introduce human interference. Summary of the invention

[0008] In view of the shortcomings of the prior art mentioned above, the purpose of the present invention is to provide a method and system for extracting urban impermeable layers, a storage medium and an electronic device, and to construct an Urban Impermeable Layer Extraction Index model (UILIE) based on multi-source and multi-modal remote sensing images, so as to quickly complete the extraction of urban impermeable layers around the world.

[0009] In a first aspect, the present invention provides a method for extracting an urban impervious layer, the method comprising the following steps: acquiring a multi-source multi-modal remote sensing image of a city; preprocessing the multi-source multi-modal remote sensing image; removing stray light from the preprocessed multi-source multi-modal remote sensing image; removing invalid values ​​from the multi-source multi-modal remote sensing image after the stray light is removed; performing data normalization on the multi-source multi-modal remote sensing image after the invalid values ​​are removed; constructing an urban impervious layer extraction index model based on the multi-source multi-modal remote sensing image after the data normalization, so as to realize urban impervious layer extraction based on the urban impervious layer extraction index model.

[0010] In an implementation of the first aspect, the multi-source multi-modal remote sensing image includes DMSP-OLS stable night light data, NPP-VIIRS light satellite data, multispectral red band data, and multispectral near-infrared band data.

[0011] In an implementation of the first aspect, preprocessing the multi-source multi-modal remote sensing image comprises the following steps:

[0012] orthorectifying the multi-source multi-modal remote sensing images;

[0013] Perform radiation correction on multi-source and multi-modal remote sensing images after orthorectification;

[0014] Perform atmospheric correction on the multi-source and multi-modal remote sensing images after radiation correction.

[0015] In an implementation of the first aspect, stray light removal is performed on the preprocessed multi-source multi-modal remote sensing image.

[0016] P i =(∫ρ(λ i )dλ i <8)×0+(∫ρ(λ i )dλ i ≥8)×∫ρ(λ)dλ i

[0017] Among them, P i represents the multi-source and multi-modal remote sensing image with stray light removed, i represents one of the DMSP-OLS stable night light data and NPP-VIIRS light satellite data, ∫ρ(λ i )dλ i represents all pixels in the i-th image, ρ(λ i ) represents the λth pixel in the i-th image, ρ(λ i )dλ i Represents the brightness value of the λth pixel in the i-th image.

[0018] In an implementation of the first aspect, according to After removing invalid values, the multi-source multi-modal remote sensing images are normalized, where P Normalization represents the normalized image, P represents the input image, P min Indicates the minimum value of the current image, P max Indicates the maximum value in the current image.

[0019] In an implementation of the first aspect, the urban impervious layer extraction index model is expressed as

[0020]

[0021] Among them, DMSP represents the data-normalized DMSP-OLS data with stray light removed, VIIRS represents the data-normalized NPP-VIIRS data with stray light removed, R represents the data-normalized preprocessed multispectral red band data, and NIR represents the data-normalized preprocessed multispectral near-infrared band data.

[0022] In an implementation of the first aspect, realizing urban impervious layer extraction based on the urban impervious layer extraction index model includes assigning non-urban impervious layer pixels to empty according to the urban impervious layer extraction index model to realize rapid extraction of urban impervious layers.

[0023] In a second aspect, the present invention provides an urban impermeable layer extraction system, the system comprising an acquisition module, a preprocessing module, a first removal module, a second removal module, a normalization module and an extraction module;

[0024] The acquisition module is used to acquire multi-source and multi-modal remote sensing images of the city;

[0025] The preprocessing module is used to preprocess the multi-source multi-modal remote sensing images;

[0026] The first removal module is used to remove stray light from the pre-processed multi-source multi-modal remote sensing image;

[0027] The second removal module is used to remove invalid values ​​from the multi-source multi-modal remote sensing image after the stray light is removed;

[0028] The normalization module is used to perform data normalization processing on the multi-source multi-modal remote sensing images after invalid values ​​are removed;

[0029] The extraction module is used to construct an urban impervious layer extraction index model based on the multi-source multi-modal remote sensing images after data normalization, so as to realize urban impervious layer extraction based on the urban impervious layer extraction index model.

[0030] In a third aspect, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned urban impermeable layer extraction method.

[0031] In a fourth aspect, the present invention provides an electronic device, comprising: a processor and a memory;

[0032] The memory is used to store computer programs;

[0033] The processor is used to execute the computer program stored in the memory so that the electronic device performs the above-mentioned urban impermeable layer extraction method.

[0034] As described above, the urban impermeable layer extraction method and system, storage medium and electronic device of the present invention have the following beneficial effects:

[0035] (1) An urban impervious layer extraction index model is constructed based on multi-source and multi-modal remote sensing images, which can quickly complete the extraction of urban impervious layers around the world;

[0036] (2) There is no need to set a threshold, which avoids human interference and effectively improves the accuracy and reliability of urban impermeable layer extraction. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 Shown is a flow chart of an urban impermeable layer extraction method in one embodiment of the present invention;

[0038] Figure 2 Shown is a schematic cross-sectional curve diagram of NPP-VIIRS data in one embodiment;

[0039] Figure 3 Shown is a schematic cross-sectional curve diagram of DMSP-OLS data in one embodiment;

[0040] FIG. 4( a ) is a schematic diagram showing a reclassification result of an image to be verified in one embodiment;

[0041] FIG4( b ) is a schematic diagram showing a reclassification result of an image to be verified in another embodiment;

[0042] Figure 5 Shown is a schematic diagram of a data feature matrix in one embodiment;

[0043] Figure 6 Shown is a schematic structural diagram of an urban impermeable layer extraction system in one embodiment of the present invention;

[0044] Figure 7 It is a schematic structural diagram of an electronic device of the present invention in one embodiment. DETAILED DESCRIPTION

[0045] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0046] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0047] The following embodiments of the present invention provide a method for extracting an urban impermeable layer, which can be applied to electronic devices. The electronic devices described in the present invention may include mobile phones, tablet computers, laptop computers, wearable devices, vehicle-mounted devices, augmented reality (AR) / virtual reality (VR) devices, ultra-mobile personal computers (UMPC), netbooks, personal digital assistants (PDA), etc., which have wireless charging functions. The embodiments of the present invention do not impose any restrictions on the specific types of electronic devices.

[0048] For example, the electronic device may be a station (STAION, ST) in a WLAN with a wireless charging function, a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with a wireless charging function, a computing device or other processing device, a computer, a laptop computer, a handheld communication device, a handheld computing device, and / or other devices for communicating on a wireless system and a next generation communication system, such as a mobile terminal in a 5G network, a mobile terminal in a future evolved Public Land Mobile Network (PLMN), or a mobile terminal in a future evolved Non-terrestrial Network (NTN).

[0049] For example, the electronic device can communicate with the network and other devices through wireless communication. The above wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobilecommunication (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), email, Short Messaging Service (SMS), BT, GNSS, WLAN, NFC, FM, and / or IR technology. The GNSS can include Global Positioning System (GPS), Global Navigation Satellite System (GLONASS), BeiDou navigation Satellite System (BDS), Quasi-Zenith Satellite System (QZSS) and / or Satellite Based Augmentation Systems (SBAS).

[0050] The technical solutions in the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0051] like Figure 1 As shown, in one embodiment, the urban impermeable layer extraction method of the present invention includes steps S1 to S6.

[0052] Step S1: Acquire multi-source and multi-modal remote sensing images of a city.

[0053] Specifically, the multi-source multi-modal remote sensing images include DMSP-OLS stable night light data, NPP-VIIRS light satellite data, multi-spectral red band data, and multi-spectral near-infrared band data. Among them, the light satellite mainly uses DMSP-OLS stable night light data and NPP-VIIRS light satellite data. Multispectral satellite data can flexibly use multispectral images, and specifically requires that the images need to have a red band with a band range of 600-750nm and a near-infrared band of 760nm-1000nm. At present, the testing of Sunflower 9 satellite and MODIS satellite data has been completed.

[0054] The DMSP-OLS stable nighttime light data provides global nighttime light information, covering the area from 75°N to 65°S, basically covering all human-inhabited areas. DMSP-OLS is equipped with sensors with advanced optoelectronic technology, which can capture weak electromagnetic waves on the earth's surface at night, such as urban lighting, lights in small settlements, signal lights of fishing boats, and fire sources. So far, there are 34 global stable nighttime light data, covering the time span from 1992 to 2013, covering 6 different satellite observation data.

[0055] The NPP-VIIRS sensor is carried on the Suomi-NPP satellite launched in October 2011. The NPP-VIIRS sensor has a total of 22 bands, and the sensor's spectral range is 400-1250nm. The shooting mode is similar to the DMSP-OLS sensor. The sensor can obtain global images of both daytime and nighttime scenes every day, but the two sensors have different transit times at night. The Day / Night Band (DNB) band of the NPP-VIIRS sensor is the main band used to detect nighttime light information. The range of the DNB band is 500-900nm. Compared with the stable nighttime light data of DMSP-OLS, this data will not have the so-called "oversaturation" phenomenon. However, precisely because of the high sensitivity of the sensor to nighttime lights, "noise" will appear in the data, such as the lights of fishing boats and numerical anomalies caused by high reflections of ice and snow.

[0056] Step S2: preprocessing the multi-source multi-modal remote sensing images.

[0057] Specifically, after obtaining the multi-source multi-spectral remote sensing images of the target area, in order to ensure the accuracy and reliability of the data, a series of pre-processing operations need to be performed on these remote sensing image data. In one embodiment, orthorectification, radiation correction, and atmospheric correction are sequentially performed on the multi-source multi-modal remote sensing images. Among them, the purpose of orthorectification is to eliminate the influence of terrain or deformation caused by camera orientation, etc., and generate a planar orthophoto. Radiation correction refers to the process of eliminating various distortions (noise) attached to the radiation brightness in the image data. Atmospheric correction converts the radiation brightness or surface reflectivity into the actual reflectivity of the surface.

[0058] Through the above preprocessing, the pixel brightness value of the original image will be converted into surface reflectance. This is because the surface reflectance image can more truly reflect the spectral characteristics of the ground object than the simple pixel brightness value image, so the corrected surface reflectance data can be used to accurately extract the impervious surface information in the target area.

[0059] Step S3: removing stray light from the preprocessed multi-source multi-modal remote sensing image.

[0060] Specifically, the radiometric resolution of the DMSP-OLS stable nighttime light data is relatively low, with a radiometric resolution of 10 -10 ~10 -8 W -2 sr -1 um -1 Therefore, the pixel brightness values ​​in the city center are often too high, and the surrounding non-urban pixels are also affected by the overflow phenomenon and are judged as urban pixels. Compared with the stable night light data of DMSP-OLS, the data obtained by the NPP-VIIRS sensor will not have the so-called "oversaturation" phenomenon. However, precisely because of the high sensitivity of the sensor to night lights, its data is more susceptible to the images of factors such as stray light caused by scattering, lights of fishing boats, and high reflection of ice and snow, resulting in numerical anomalies and noise in the data. After analysis, the specific reasons for the above problems can be summarized as follows:

[0061] ① Atmospheric scattering: When light passes through the atmosphere, it will be scattered by particles in the atmosphere, causing the light to diffuse in all directions, thereby generating stray light and expanding the range of light on the remote sensing image.

[0062] ② Response of the photomultiplier tube: The photomultiplier tube is an important component in remote sensing image receiving equipment. Its response to bright light sources may lead to signal amplification. At the same time, the spillover effect of bright light sources can also be understood as a type of stray light, which will have a great impact on the accuracy of light satellite monitoring.

[0063] ③ Limitation of sensor radiation resolution: Due to the limitation of the sensor’s radiation resolution, the radiation information of high-brightness areas in the city center may not be fully recorded, which will also lead to the expansion of the lighting range.

[0064] Therefore, it can be concluded that the presence of stray light and light spillover effects will reduce the contrast and signal-to-noise ratio of the image plane. In severe cases, the target signal may be completely annihilated by the stray light, interfering with data application. For light satellite stray light, it will greatly affect the accuracy of urban impermeable layer extraction, so it is necessary to remove the stray light. The data characteristics of NPP-VIIRS and DMSP-OLS are analyzed, and two types of data from the same year are selected for cross-sectional analysis.

[0065] like Figure 2 As shown in the figure, a profile analysis of the NPP-VIIRS data shows that the stray light pixels in the NPP-VIIRS data are pixels with a brightness value less than 8. Figure 3 As shown in the figure, a profile analysis of the DMSP-OLS data was performed, and it can be found from the data that the stray light pixels of the DMSP-OLS data are pixels with brightness values ​​less than 8.

[0066] Therefore, it is necessary to remove stray light from the DMSP-OLS stable nighttime light data and the NPP-VIIRS light satellite data and complete a preliminary correction of the image brightness. In one embodiment, the stray light removal of the pre-processed multi-source multi-modal remote sensing image is performed by

[0067] P i =(∫ρ(λ i )dλ i <8)×0+(∫ρ(λ i )dλ i ≥8)×∫ρ(λ)dλ i

[0068] Among them, P i represents the multi-source and multi-modal remote sensing image with stray light removed, i represents one of the DMSP-OLS stable night light data and NPP-VIIRS light satellite data, ∫ρ(λ i )dλ i represents all pixels in the i-th image, ρ(λ i ) represents the λth pixel in the i-th image, ρ(λ i )dλ iIndicates the brightness value of the λth pixel in the i-th image. In other words, the pixels with pixel values ​​less than 8 are defined as stray light pixels, and the stray light pixels to be removed are assigned a value of 0. In the stray light removal, the DMSP-OLS stable night light data and the NPP-VIIRS light satellite data are processed in turn to obtain the corrected images after removing the stray light of the two scenes.

[0069] Step S4: removing invalid values ​​from the multi-source multi-modal remote sensing image after the stray light is removed.

[0070] Specifically, in order to ensure the accuracy of urban impervious surface extraction and avoid the interference of some abnormal pixels on the extraction of impervious surfaces from DMSP-OLS stable night light data and NPP-VIIRS light satellite data, it is necessary to sample the target area vector data to crop the multi-source and multi-modal remote sensing images and assign all non-target areas to empty, that is, Nodata.

[0071] Step S5: performing data normalization processing on the multi-source multi-modal remote sensing images after invalid values ​​are removed.

[0072] Specifically, the multi-source multi-modal remote sensing image of the present invention integrates DMSP-OLS stable night light data, NPP-VIIRS light satellite data, multi-spectral red band data, and multi-spectral near-infrared band data. In image data processing, because the numerical range and distribution of image data from different sources are different, in order to avoid errors caused by inconsistent data dimensions, the remote sensing image needs to be normalized so that different features are comparable. Through normalization, it can be ensured that the data of all bands are in the same numerical range, which facilitates subsequent analysis and processing. The maximum and minimum sampling normalization method of the present invention completes the normalization of different data, and the calculation method is as follows:

[0073]

[0074] Where P Normalization represents the normalized image, P represents the input image, P min Indicates the minimum value of the current image, P max Indicates the maximum value in the current image. Therefore, the data normalization of DMSP-OLS stable night light data after removing stray light, NPP-VIIRS light satellite data after removing stray light, multispectral red band data after atmospheric correction, and multispectral near-infrared band data after atmospheric correction is completed in sequence.

[0075] Step S6: constructing an urban impervious layer extraction index model based on the multi-source multi-modal remote sensing images after data normalization, so as to realize urban impervious layer extraction based on the urban impervious layer extraction index model.

[0076] Specifically, the characteristics of DMSP-OLS stable night light data and NPP-VIIRS light satellite data in urban areas are: showing higher night light brightness values, the light distribution is clustered and the boundaries of the light area are relatively clear. As the size of the city increases, the night light area also increases accordingly; the characteristics of non-urban areas are: showing lower or close to 0 night light intensity values, the light distribution is relatively dispersed, and smaller discrete light patches may appear in rural residential areas, long and thin light strips may appear along roads, and some factories and other facilities may form small light patches.

[0077] The near-infrared band has a very good recognition ability for cities and vegetation, which is mainly manifested in the following aspects: ① Vegetation area characteristics: In the near-infrared band, green vegetation shows a higher reflectivity, forming a bright pixel value. When the vegetation grows well, the near-infrared reflectivity is high. Different types of vegetation have different reflection characteristics in the near-infrared band, such as crops, woodlands, grasslands, etc. ② Urban area characteristics: In the near-infrared band, urban areas show a lower reflectivity due to the lack of green vegetation, forming a dark pixel value. Impervious surfaces (such as buildings, roads, etc.) have a very low reflectivity in the near-infrared band. Vegetation areas such as green spaces and parks in cities will show a higher near-infrared reflectivity. ③ Other characteristics: Water bodies have strong absorption in the near-infrared band, very low reflectivity, and present dark pixel values. The reflectivity of bare land surfaces in the near-infrared band is between vegetation and cities. The near-infrared band has little impact on the atmosphere and the data quality is good.

[0078] The identification characteristics of cities and vegetation in the infrared band are as follows: ① Vegetation area characteristics: In the infrared band (especially the short-wave infrared band), green vegetation has a strong reflection due to the large number of intercellular spaces, showing a brighter pixel value. Different types of vegetation have different reflection characteristics in the infrared band, which can be used to distinguish vegetation types. The better the vegetation growth condition, the higher the infrared band reflectivity. ② Urban area characteristics: In the infrared band, urban artificial objects (such as buildings, roads, etc.) have a low reflectivity due to their dense materials, showing darker pixel values. Different building materials (such as cement, bricks and tiles, etc.) have different reflection characteristics in the infrared band, which can be used to classify buildings. Vegetation areas such as urban green spaces and parks will show a higher infrared band reflectivity. ③ Other characteristics: Water bodies have strong absorption in the infrared band, and the reflectivity is very low, showing dark pixel values. The reflectivity of bare land in the infrared band is between vegetation and cities. The infrared band is less affected by the atmosphere, but it will be affected by temperature and humidity.

[0079] Therefore, based on the data characteristics of different images, the present invention establishes an urban impervious layer extraction index model based on multi-source multi-modal remote sensing images of light satellite data and multi-spectral satellite data, which is specifically as follows:

[0080]

[0081] Among them, DMSP represents the data normalized DMSP-OLS data with stray light removed, VIIRS represents the data normalized NPP-VIIRS data with stray light removed, R represents the data normalized preprocessed multispectral red band data, and NIR represents the data normalized preprocessed multispectral near-infrared band data. After the urban impervious layer extraction index model is calculated, the non-urban impervious layer pixels can be automatically assigned to null, and the rapid extraction of urban impervious layers can be directly realized.

[0082] In order to verify the extraction effect of the urban impervious layer extraction index model on the urban impervious layer, the verification was carried out according to the following steps, namely, urban impervious layer extraction from reference images, urban impervious layer extraction from images to be verified, standardization of extraction results, construction of data feature matrix, selection of typical cities and calculation of data feature matrix, calculation of total accuracy and RMSE, and final accuracy result evaluation.

[0083] Step 1: Extract urban impervious layers from reference images.

[0084] The Wuhan University CLCD data is an annual land cover dataset for China based on Landsat satellite images. The spatial resolution of the data is 30 meters, which can provide more detailed surface cover information. The CLCD data divides land objects into 9 categories, namely: cropland, forest, shrub, grassland, water, snow / ice, bare land, impervious, and wetland. Therefore, the impervious surface in the CLCD image is extracted. In order to ensure data verification, the CLCD data is compared with other images to be verified, and the CLCD data is resampled to reduce its resolution to 1000m.

[0085] Step 2: Extract the impervious layer of the city in the image to be verified.

[0086] The images to be verified include the urban impervious layer extraction results of the urban impervious layer extraction index model, the urban impervious layer extraction results of DMSP-OLS data, and the urban impervious layer extraction results of NPP-VIIRS data. The urban impervious layer extraction results of CLCD data are the reference images. The urban impervious layer extraction results of DMSP-OLS data and the urban impervious layer extraction results of NPP-VIIRS data are used as the control group to evaluate the accuracy of the urban impervious layer extraction results of the urban impervious layer extraction index model. For the conventional threshold method for sampling DMSP-OLS data and NPP-VIIRS data, the threshold is set to 8 to extract the urban impervious layer. At the same time, the urban impervious layer extraction index model is called to fuse the DMSP data, NPP-VIIRS data, and multispectral data of the same period to achieve the extraction of urban impervious layers.

[0087] Step 3: Standardization of extraction results.

[0088] In order to better compare the extraction accuracy of the images to be verified, all data are standardized and classified, and the urban impervious layer extraction results of DMSP-OLS data, NPP-VIIRS data, and CLCD data are reclassified one by one, and the impervious layer pixels and non-impervious layers are assigned values, so that the urban impervious layer pixel value of the image to be verified is 3, and the non-urban impervious layer pixel is 1; the urban impervious layer pixel of the reference image is 8, and the non-impervious layer pixel is 1. As shown in Figure 4(a), the images to be verified include the urban impervious layer extraction results of the UILIE model, the urban impervious layer extraction results of the DMSP-OLS data, and the urban impervious layer extraction results of the NPP-VIIRS data. As shown in Figure 4(b), the urban impervious layer extraction results of the CLCD data are the reference image.

[0089] Step 4: Construct data feature matrix.

[0090] The CLCD results, the urban impervious layer extraction results of the urban impervious layer extraction index model, the urban impervious layer extraction results of the DMSP-OLS data, and the urban impervious layer extraction results of the NPP-VIIRS data are added one by one to obtain the added verification data feature matrix. In the matrix, when the pixel value is 2, the reference image and the image to be verified are both non-urban impervious layer pixels; when the pixel value is 12, the reference image and the image to be verified are both urban impervious layer pixels; when the pixel value is 9, the reference image is an urban impervious layer pixel, but the image to be verified is a non-urban impervious layer pixel; when the pixel value is 4, the reference image is a non-urban impervious layer pixel, but the image to be verified is an urban impervious layer pixel; data examples are as follows Figure 5 shown.

[0091] Step 5: Select typical cities and obtain data feature matrix.

[0092] According to step 4, the image to be verified and the reference image are added to obtain the verification data feature image. It can be seen that the pixel value of 4 accounts for the largest proportion, which means that the DMSP-OLS data and NPP-VIIRS data misclassify many non-urban impervious layer pixels as urban impervious layer pixels. The urban impervious layer area extracted by the UILIE model is closest to the CLCD data result of 30m.

[0093] For further verification, major cities in the east, south, north, central, southwest and northwest of China were selected for total accuracy and root mean square error verification. Among them, the Yangtze River Delta, Nanning, Beijing, Wuhan, Chengdu and Urumqi were selected for verification.

[0094] Step 6: Calculate the overall accuracy and RMSE.

[0095] Overall Accuracy (OA) refers to the ratio of correctly classified samples to all samples. In classification problems, it represents the ratio of all correctly classified samples (including true positive and true negative samples) to the total number of samples. The calculation method is Wherein, C represents all categories, including urban impervious layer category and non-urban impervious layer category in the present invention, i=1 indicates that the current category is urban impervious layer category, and i=2 indicates that the current category is non-urban impervious layer category. i represents the number of correctly classified samples of the i-th category, and n represents the total number of samples.

[0096] On the basis of calculating the total accuracy, the root mean square error is further calculated to evaluate the accuracy of the model. The root mean square error RMSE (Root Mean Square Error) is a commonly used indicator to measure the accuracy of the prediction model. RMSE can well reflect the accuracy of the prediction. The smaller the RMSE value, the better the accuracy of the prediction model in describing the experimental data. The calculation method is Wherein, C represents all categories, including urban impervious layer category and non-urban impervious layer category in the present invention, i=1 indicates that the current category is urban impervious layer category, and i=2 indicates that the current category is non-urban impervious layer category. n represents the total number of samples. i Indicates that the reference image is the number of pixels of the i-th category, b i Indicates the number of pixels of the i-th category in the image to be verified.

[0097] Step 7: Final accuracy result evaluation.

[0098] Calculate the total accuracy according to step six. The total accuracy results are shown in Table 1. The total accuracy OA of typical cities. According to the data in Table 1, the overall accuracy of the three urban impervious layer extraction methods in different cities is compared. ① UILIE model extraction results: accuracy range: 0.8063-0.9418, the highest accuracy appears in Chengdu (0.9418), and the accuracy in other cities is also relatively high. ② NPP-VIIRS threshold method extraction results: accuracy range: 0.4389-0.641, the highest accuracy appears in Chengdu (0.641), and the accuracy in most cities is relatively low. ③ DMSP-OLS threshold method extraction results: accuracy range: 0.428-0.6761, the highest accuracy appears in Chengdu (0.6761), and the accuracy in most cities is relatively low. In summary, the overall accuracy of the UILIE model is significantly higher than the other two methods. It has relatively high accuracy in all cities, especially in Chengdu, Nanning and other cities.

[0099] Table 1. Total accuracy OA of typical cities

[0100]

[0101] Calculate the root mean square error according to step six. The root mean square error results are shown in the root mean square error RMSE of typical cities in Table 2. The root mean square error of the UILIE model in all cities is much lower than that of the NPP-VIIRS and DMS-OLSP models, especially in the Yangtze River Delta region, where its root mean square error is only 37.818, while the other two models are more than 200 in this region. This shows that the UILIE model is significantly better than the other two models in terms of prediction accuracy. The root mean square error values ​​of the NPP-VIIRS and DMS-OLSP models are similar, exceeding 80 in most cities, and are not as good as the UILIE model. Among the two, the error of the DMS-OLSP model in Wuhan, Nanning and the Yangtze River Delta region is slightly lower than that of the NPP-VIIRS, but slightly higher in other cities. In summary, the UILIE model is the best performing model among the three, and its prediction accuracy is significantly higher than the other two models, especially in first-tier cities such as the Yangtze River Delta, where the advantages of the UILIE model are more obvious.

[0102] Table 2. RMSE of typical cities

[0103]

[0104] The protection scope of the urban impermeable layer extraction method described in the embodiment of the present invention is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing or replacing steps in the prior art based on the principles of the present invention are included in the protection scope of the present invention.

[0105] An embodiment of the present invention also provides an urban impermeable layer extraction system, which can implement the urban impermeable layer extraction method described in the present invention. However, the implementation device of the urban impermeable layer extraction system described in the present invention includes but is not limited to the structure of the urban impermeable layer extraction system listed in this embodiment. All structural deformations and replacements of the prior art made according to the principles of the present invention are included in the protection scope of the present invention.

[0106] like Figure 6 As shown, in one embodiment, the urban impermeable layer extraction system of the present invention includes an acquisition module 61, a preprocessing module 62, a first removal module 63, a second removal module 64, a normalization module 65 and an extraction module 66.

[0107] The acquisition module 61 is used to acquire multi-source and multi-modal remote sensing images of the city.

[0108] The preprocessing module 62 is connected to the acquisition module 61 and is used to preprocess the multi-source multi-modal remote sensing images.

[0109] The first removal module 63 is connected to the preprocessing module 62 and is used to remove stray light from the preprocessed multi-source multi-modal remote sensing image.

[0110] The second removal module 64 is connected to the first removal module 63 and is used to remove invalid values ​​from the multi-source multi-modal remote sensing image after the stray light is removed.

[0111] The normalization module 65 is connected to the second removal module 64 and is used to perform data normalization processing on the multi-source multi-modal remote sensing images after invalid values ​​are removed.

[0112] The extraction module 66 is connected to the normalization module 65, and is used to construct an urban impervious layer extraction index model based on the multi-source multi-modal remote sensing images after data normalization, so as to realize urban impervious layer extraction based on the urban impervious layer extraction index model.

[0113] Among them, the structures and principles of the acquisition module 61, the preprocessing module 62, the first removal module 63, the second removal module 64, the normalization module 65 and the extraction module 66 correspond one by one to the steps in the above-mentioned urban impermeable layer extraction method, so they are not repeated here.

[0114] In the several embodiments provided by the present invention, it should be understood that the disclosed system, device or method can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules / units is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules or units, which can be electrical, mechanical or other forms.

[0115] The modules / units described as separate components may or may not be physically separated, and the components displayed as modules / units may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules / units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention. For example, the functional modules / units in the various embodiments of the present invention may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.

[0116] Those of ordinary skill in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0117] The embodiment of the present invention also provides a computer-readable storage medium. A person skilled in the art can understand that all or part of the steps in the urban impermeable layer extraction method of the above embodiment can be completed by a program to instruct a processor, and the program can be stored in a computer-readable storage medium, and the storage medium is a non-transitory medium, such as a random access memory, a read-only memory, a flash memory, a hard disk, a solid-state hard disk, a magnetic tape, a floppy disk, an optical disc, and any combination thereof. The above storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a digital video disc (DVD)), or a semiconductor medium (for example, a solid-state disk (SSD)), etc.

[0118] An embodiment of the present invention further provides an electronic device, which includes a processor and a memory.

[0119] The memory is used to store computer programs.

[0120] The memory includes: ROM, RAM, disk, USB flash drive, memory card or CD and other media that can store program codes.

[0121] The processor is connected to the memory and is used to execute the computer program stored in the memory so that the electronic device executes the above-mentioned urban impermeable layer extraction method.

[0122] Preferably, the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0123] like Figure 7As shown, the electronic device of the present invention is in the form of a general computing device. The components of the electronic device may include but are not limited to: one or more processors or processing units 71, a memory 72, and a bus 73 connecting different system components (including the memory 72 and the processing unit 71).

[0124] Bus 73 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0125] Electronic devices typically include a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device, including volatile and non-volatile media, removable and non-removable media.

[0126] The memory 72 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 721 and / or cache memory 722. The electronic device may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 723 may be used to read and write non-removable, non-volatile magnetic media ( Figure 7 not shown, usually called a "hard drive"). Although Figure 7 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 73 via one or more data medium interfaces. The memory 72 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0127] A program / utility 724 having a set (at least one) of program modules 7241 may be stored, for example, in the memory 72, such program modules 7241 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 7241 generally perform the functions and / or methods of the embodiments described herein.

[0128] The electronic device may also communicate with one or more external devices (e.g., keyboards, pointing devices, displays, etc.), one or more devices that enable a user to interact with the electronic device, and / or any device that enables the electronic device to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via input / output (I / O) interface 74. Furthermore, the electronic device may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via network adapter 75. Figure 7 As shown, the network adapter 75 communicates with other modules of the electronic device via the bus 73. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0129] In summary, the urban impervious layer extraction method and system, storage medium and electronic device of the present invention construct an urban impervious layer extraction index model based on multi-source multi-modal remote sensing images, thereby being able to quickly complete the rapid extraction of global urban impervious layers; there is no need to set a threshold, human interference is avoided, and the accuracy and reliability of urban impervious layer extraction are effectively improved. Therefore, the present invention effectively overcomes various shortcomings in the prior art and has a high industrial utilization value.

[0130] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A method for extracting urban impermeable layers, characterized in that: The method comprises the following steps: Obtain multi-source and multi-modal remote sensing images of the city; Preprocessing the multi-source multi-modal remote sensing images; Remove stray light from preprocessed multi-source and multi-modal remote sensing images; Remove invalid values ​​from multi-source and multi-modal remote sensing images after removing stray light; Perform data normalization on multi-source and multi-modal remote sensing images after invalid values ​​are removed; An urban impervious layer extraction index model is constructed based on the multi-source multi-modal remote sensing images after data normalization, so as to realize urban impervious layer extraction based on the urban impervious layer extraction index model.

2. The method for extracting urban impermeable layers according to claim 1, characterized in that: The multi-source and multi-modal remote sensing images include DMSP-OLS stable night light data, NPP-VIIRS light satellite data, multispectral red band data, and multispectral near-infrared band data.

3. The method for extracting urban impermeable layer according to claim 1, characterized in that: Preprocessing the multi-source multi-modal remote sensing images includes the following steps: orthorectifying the multi-source multi-modal remote sensing images; Perform radiation correction on multi-source and multi-modal remote sensing images after orthorectification; Perform atmospheric correction on the multi-source and multi-modal remote sensing images after radiation correction.

4. The method for extracting urban impermeable layer according to claim 2, characterized in that: The stray light removal of preprocessed multi-source multi-modal remote sensing images is carried out by P i =(∫ρ(λ i )dλ i <8)×0+(∫ρ(λ i )dλ i ≥8)×∫ρ(λ)dλ i Among them, P i represents the multi-source and multi-modal remote sensing image with stray light removed, i represents one of the DMSP-OLS stable night light data and NPP-VIIRS light satellite data, ∫ρ(λ i )dλ i represents all pixels in the i-th image, ρ(λ i ) represents the λth pixel in the i-th image, ρ(λ i )dλ i Represents the brightness value of the λth pixel in the i-th image.

5. The method for extracting urban impermeable layer according to claim 1, characterized in that: according to After removing invalid values, the multi-source multi-modal remote sensing images are normalized, where P Normalization represents the normalized image, P represents the input image, P min Indicates the minimum value of the current image, P max Indicates the maximum value in the current image.

6. The method for extracting urban impermeable layers according to claim 2, characterized in that: The urban impervious layer extraction index model is expressed as Among them, DMSP represents the data-normalized DMSP-OLS data with stray light removed, VIIRS represents the data-normalized NPP-VIIRS data with stray light removed, R represents the data-normalized preprocessed multispectral red band data, and NIR represents the data-normalized preprocessed multispectral near-infrared band data.

7. The method for extracting urban impermeable layers according to claim 1, characterized in that: The urban impervious layer extraction based on the urban impervious layer extraction index model includes assigning non-urban impervious layer pixels to be empty according to the urban impervious layer extraction index model to achieve rapid extraction of urban impervious layers.

8. An urban impermeable layer extraction system, characterized in that: The system comprises an acquisition module, a preprocessing module, a first removal module, a second removal module, a normalization module and an extraction module; The acquisition module is used to acquire multi-source and multi-modal remote sensing images of the city; The preprocessing module is used to preprocess the multi-source multi-modal remote sensing images; The first removal module is used to remove stray light from the pre-processed multi-source multi-modal remote sensing image; The second removal module is used to remove invalid values ​​from the multi-source multi-modal remote sensing image after the stray light is removed; The normalization module is used to perform data normalization processing on the multi-source multi-modal remote sensing images after invalid values ​​are removed; The extraction module is used to construct an urban impervious layer extraction index model based on the multi-source multi-modal remote sensing images after data normalization, so as to realize urban impervious layer extraction based on the urban impervious layer extraction index model.

9. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the urban impermeable layer extraction method described in any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: include: Processor and memory; The memory is used to store computer programs; The processor is used to execute the computer program stored in the memory so that the electronic device executes the urban impermeable layer extraction method according to any one of claims 1 to 7.

Images