Method and device for acquiring remote sensing clear sky data set, electronic equipment and storage medium

Abstract

The present disclosure relates to the technical field of remote sensing data processing, and particularly relates to a method and device for acquiring remote sensing clear sky data set, electronic equipment and storage medium, the method comprises the following steps: preprocessing remote sensing image data collected in a predetermined time period to obtain a global remote sensing image time sequence; selecting a target remote sensing image time sequence of a target region from the global remote sensing image time sequence, and then extracting the reflectivity spectral characteristics of each pixel in the target waveband and the NDVI of each pixel; for a plurality of pixels with the same target position but different time sequences in the target remote sensing image time sequence, using a K-means algorithm to cluster the plurality of pixels to obtain a plurality of clustering clusters; identifying a clear sky clustering cluster in the clustering cluster; selecting a clear sky pixel of a target position from the clear sky clustering cluster; and synthesizing a clear sky image of the target region based on the clear sky pixels of each target position in the target region. The technical scheme can simply and quickly acquire clear sky data.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of remote sensing data processing, in particular to a method and device for obtaining a remote sensing clear sky dataset, an electronic device and a storage medium. BACKGROUND

[0002] Remote sensing clear sky dataset processing is a process of distinguishing clear sky areas from multi-temporal data and synthesizing clear sky through specific algorithms and analysis methods from image data obtained from satellite or aerial platforms using remote sensing technology. The purpose is to improve the usability and quality of remote sensing data, ensure that subsequent ground feature extraction, environmental monitoring, resource assessment and other applications can be based on accurate and clear data, and thus improve the practical application effect and scientific research value of remote sensing technology.

[0003] Traditional remote sensing clear sky dataset processing methods mainly rely on the identification and removal of cloud information in remote sensing images to achieve the extraction of clear sky areas. Clouds usually have high reflectivity and low transmissivity, which is particularly evident in the visible and infrared bands. Traditional schemes are mostly based on the analysis of the spectral characteristics (such as reflectivity, brightness temperature, etc.) of a large number of bands of remote sensing images such as visible, infrared and microwave bands, as well as related data of various external data sources (such as weather data, terrain data, etc.), to identify and calculate the physical properties of clouds such as temperature, height and thickness, and then distinguish clear sky areas to obtain clear sky data. However, this kind of method usually needs to combine a large number of band information of remote sensing images and external data sources for comprehensive analysis and judgment, and also needs to consider seasonal factors in the analysis process, which is a relatively complex scheme. SUMMARY

[0004] To solve the problems in the related art, the embodiments of the present disclosure provide a method and device for obtaining a remote sensing clear sky dataset, an electronic device and a storage medium.

[0005] In a first aspect, the embodiments of the present disclosure provide a method for obtaining a remote sensing clear sky dataset, comprising:

[0006] obtaining remote sensing image data collected within a predetermined time period;

[0007] preprocessing the remote sensing image data to obtain a global remote sensing image time sequence within the predetermined time period, the global remote sensing image time sequence being a sequence of global remote sensing images arranged in chronological order;

[0008] selecting a target remote sensing image time sequence of a target area from the global remote sensing image time sequence;

[0009] extracting reflectivity spectral features of each pixel in a target band and normalized difference vegetation index (NDVI) of each pixel from the target remote sensing image time sequence;

[0010] using a K-means algorithm to cluster the multiple pixels with the same target position but different time sequences in the target remote sensing image time sequence, to obtain multiple clustering clusters;

[0011] identifying a clear sky clustering cluster in the clustering clusters;

[0012] selecting a clear sky pixel of the target position from the clear sky clustering cluster;

[0013] synthesizing a clear sky image of the target region based on the clear sky pixels of each target position in the target region.

[0014] In a possible implementation manner,

[0015] The remote sensing image data is L1-level data of a medium resolution spectral imager (MERSI) load of a Fengyun-3 satellite; the preprocessing of the remote sensing image data set to obtain a global remote sensing image time sequence in a predetermined time period comprises:

[0016] performing atmospheric correction processing, equi-latitude-longitude projection processing and data splicing processing on the collected remote sensing image data to obtain multiple global remote sensing images;

[0017] sequencing the multiple global remote sensing images in the predetermined time period in time order to form the global remote sensing image time sequence.

[0018] In a possible implementation manner, the atmospheric correction processing comprises:

[0019] acquiring remote sensing image data in a day from the remote sensing image data;

[0020] acquiring original reflectivity of each pixel in each target band and angle information including a solar elevation angle, a solar zenith angle, a satellite elevation angle, a satellite zenith angle and elevation data based on the remote sensing image data in the day.

[0021] when the solar zenith angle meets a correction condition, correcting the original reflectivity of each pixel in the remote sensing image data in the day based on the angle information to obtain atmospheric corrected remote sensing image data.

[0022] In a possible implementation manner, the equi-latitude-longitude projection processing comprises:

[0023] performing equi-latitude-longitude projection on the atmospheric corrected remote sensing image data to obtain initial map projection data with a predetermined resolution;

[0024] interpolating the initial map projection data to obtain map projection data.

[0025] In a possible implementation, the data splicing processing includes:

[0026] acquiring a target waveband of the map projection data;

[0027] For each target waveband, acquiring equal latitude and longitude information of the map projection data of the target waveband;

[0028] Based on the equal latitude and longitude information of the map projection data of the target waveband, merging a plurality of pieces of map projection data of the target waveband into block data of a predetermined latitude and longitude span;

[0029] Based on the satellite zenith angle of the block data of the predetermined latitude and longitude span, merging the block data of the predetermined latitude and longitude span into integral block data, and then splicing the integral block data to obtain a global remote sensing image of the target waveband.

[0030] In a possible implementation, the target waveband includes a blue waveband, a green waveband, a red waveband, and a near-infrared waveband.

[0031] In a possible implementation, the clustering of the plurality of pixels using the K-means algorithm to obtain a plurality of clustering clusters includes:

[0032] Clustering the plurality of pixels using the K-means algorithm to obtain two clustering clusters;

[0033] Identifying a suspected clear-sky clustering cluster in the two clustering clusters;

[0034] Re-clustering the pixels in the suspected clear-sky clustering cluster using the K-means algorithm to obtain three clustering clusters;

[0035] Identifying a clear-sky clustering cluster in the three clustering clusters.

[0036] In a possible implementation, the selecting of the clear-sky pixel of the target position from the clear-sky clustering cluster includes:

[0037] Selecting, as the clear-sky pixel of the target position, a pixel with the maximum NDVI from the clear-sky clustering cluster.

[0038] In a second aspect, the embodiments of the present disclosure provide an acquisition device of remote sensing clear-sky data set, including:

[0039] An acquisition module configured to acquire remote sensing image data collected in a predetermined time period;

[0040] a preprocessing module configured to preprocess the remote sensing image data to obtain a global remote sensing image time sequence in a predetermined time period, the global remote sensing image time sequence being a sequence of global remote sensing images arranged in time sequence;

[0041] a region selection module configured to select a target remote sensing image time sequence of a target region from the global remote sensing image time sequence;

[0042] a feature extraction module configured to extract reflectivity spectral features of each pixel in a target waveband and normalized difference vegetation index (NDVI) of each pixel from the target remote sensing image time sequence;

[0043] a clustering module configured to use a K-means algorithm to cluster a plurality of pixels at the same target position but different time sequences in the target remote sensing image time sequence, to obtain a plurality of clustering clusters;

[0044] an identification module configured to identify a clear sky clustering cluster in the clustering clusters;

[0045] a pixel selection module configured to select a clear sky pixel of the target position from the clear sky clustering cluster;

[0046] an image synthesis module configured to synthesize a clear sky image of the target region based on the clear sky pixels of each target position in the target region.

[0047] In a third aspect, an electronic device is provided, including a memory and a processor, wherein the memory is configured to store one or more computer instructions, and the one or more computer instructions are executed by the processor to implement the method in any one of the first aspect.

[0048] In a fourth aspect, a computer readable storage medium is provided, having stored thereon computer instructions, which, when executed by a processor, implement the method in any one of the first aspect.

[0049] According to the technical scheme provided by the embodiment of the present disclosure, the remote sensing image data collected in a predetermined time period can be preprocessed to construct a global remote sensing image with high quality in a global range, and then a target remote sensing image time sequence of a target region can be selected from the global remote sensing image time sequence. The reflectivity spectral characteristics of each pixel in a target band and the NDVI of each pixel are extracted from the target remote sensing image time sequence. Then, for a plurality of pixels at the same target position but with different time sequences in the target remote sensing image time sequence, the plurality of pixels are clustered by using a K-means algorithm to obtain a plurality of clustering clusters. The clear sky clustering cluster in the clustering cluster is identified. The clear sky pixel of the target position is selected from the clear sky clustering cluster. The clear sky image of the target region is synthesized based on the clear sky pixels of each target position in the target region, so that a high-quality clear sky image can be obtained. The K-means algorithm is used in the scheme, and only the reflectivity spectral characteristics of a few target bands and the NDVI of each pixel are needed to cluster the clear sky pixels of the target position. The clear sky image of the target region can be synthesized based on the clear sky pixels of each target position in the target region. The data required for identifying the clear sky pixels is less, and is not affected by other external factors. The analysis scheme is simple and easy to implement, and maintains high efficiency and stability when processing large-scale remote sensing image data.

[0050] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0051] Other features, objects, and advantages of the present disclosure will become more apparent from the following detailed description of the non-limiting embodiments, taken in conjunction with the accompanying drawings. In the drawings:

[0052] Figure 1 A flowchart of a method for acquiring a remote sensing clear sky dataset according to an embodiment of the present disclosure is shown.

[0053] Figure 2 A schematic diagram of the reflectivity spectral characteristics of a pixel at a target position in a target band and the NDVI extracted from a target remote sensing image time sequence in a predetermined time period according to an embodiment of the present disclosure is shown.

[0054] Figure 3 A structural block diagram of an acquisition device for a remote sensing clear sky dataset according to an embodiment of the present disclosure is shown.

[0055] Figure 4 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown.

[0056] Figure 5 A structural schematic diagram of a computer system suitable for implementing the method of the embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0057] Hereinafter, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so as to be easily carried out by one of ordinary skill in the art. Also, portions unrelated to the description of the exemplary embodiments are omitted in the drawings for the sake of clarity.

[0058] In the present disclosure, it should be understood that terms such as "include" or "have" or the like are intended to indicate that there are features, numbers, steps, actions, components, parts or combinations thereof disclosed in the specification, and do not exclude the possibility that one or more other features, numbers, steps, actions, components, parts or combinations thereof exist or are added.

[0059] It should also be noted that the embodiments in the present disclosure and the features in the embodiments can be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0060] Figure 1 A flowchart of a method for acquiring a remote sensing clear sky dataset according to an embodiment of the present disclosure is shown. As shown in the figure, the method comprises the following steps S101-S108: Figure 1

[0061] In step S101, remote sensing image data collected in a predetermined time period is acquired;

[0062] In step S102, the remote sensing image data is preprocessed to obtain a global remote sensing image time series sequence in the predetermined time period, the global remote sensing image time series sequence being a sequence in which global remote sensing images are arranged in chronological order;

[0063] In step S103, a target remote sensing image time series sequence of a target region is selected from the global remote sensing image time series sequence;

[0064] In step S104, reflectance spectral features of each pixel in a target band and normalized difference vegetation index (NDVI) of each pixel are extracted from the target remote sensing image time series sequence;

[0065] In step S105, for a plurality of pixels at the same target position but with different time series sequences in the target remote sensing image time series sequence, the plurality of pixels are clustered using a K-means algorithm to obtain a plurality of clustering clusters;

[0066] In step S106, a clear sky clustering cluster is identified in the clustering clusters;

[0067] In step S107, a clear sky pixel of the target position is selected from the clear sky clustering cluster;

[0068] ​In step S108, a clear sky image of the target region is synthesized based on the clear sky pixels of each target position in the target region.

[0069] In a possible implementation, the method for acquiring the remote sensing clear sky dataset is applicable to a computer, a computing device, a server, a server cluster, or the like, which can execute the acquisition of the remote sensing clear sky dataset.

[0070] In a possible implementation, the remote sensing image data acquired by the remote sensing device of the satellite payload in a predetermined time period can be acquired, and the predetermined time period is a continuous time period, such as a month.

[0071] In a possible implementation, the remote sensing image data can be preprocessed, and different remote sensing image data can be preprocessed differently, as long as a global remote sensing image time sequence in a predetermined time period can be acquired, and the global remote sensing image time sequence is a sequence in which global remote sensing images are arranged in time sequence. For example, the preprocessing can include correction processing, projection processing, and splicing processing. The correction processing is mainly to restore the ground true reflectivity of each pixel in the remote sensing image data. The projection processing is mainly to project the remote sensing image data into the same map coordinate system, so as to be spliced subsequently. The splicing processing is mainly to splice the remote sensing image data in the same time period together to form a complete image in a global range. Of course, in other implementations, some remote sensing image data can not need to be corrected, and in this case, the preprocessing can include the projection processing and the splicing processing.

[0072] In a possible implementation, a target region can be selected in the global range as needed, and for each global remote sensing image in the global remote sensing image time sequence, a target remote sensing image corresponding to the target region in the global remote sensing image can be selected, and these target remote sensing images can form a target remote sensing image time sequence in time sequence.

[0073] In a possible implementation, in order to obtain the clear-sky pixels in the target remote sensing image, a feature that can represent whether the pixel is a clear-sky pixel or a non-clear-sky pixel needs to be extracted from the target remote sensing image, which can be a reflectance spectral feature of each pixel in the target band. Clouds and other ground objects can be distinguished by using reflectance spectral feature analysis, which is a method based on the reflectance difference of ground objects in different spectral bands. The reflectance spectral feature refers to the reflectance variation characteristics of ground objects in different wavelength ranges. These characteristics are mainly determined by the physical, chemical, and biological characteristics of the ground objects, and are affected by multiple factors such as the wavelength of incident light, the incidence angle, the observation angle, and the atmospheric conditions. Therefore, a small number of target bands that can best distinguish the clear-sky feature and the non-clear-sky feature (such as the cloud feature) of the pixel can be selected, and the reflectance spectral feature of the target band is used for subsequent clustering analysis. The reflectance spectral feature includes reflectance and / or brightness temperature and the like.

[0074] In a possible implementation, the information can further include the NDVI (Normalized Difference Vegetation Index) of each pixel. The NDVI is a commonly used vegetation index, which is used to evaluate and quantify the vegetation condition in a remote sensing image. The calculation formula is: NDVI=(NIR-Red) / (NIR+Red), where NIR represents the reflectance value of the near-infrared band, and Red represents the reflectance value of the visible red band. This index estimates the lushness of vegetation by comparing the reflectance difference between the near-infrared band and the red band, and the inverse use of the NDVI can indicate non-vegetation areas, which may indirectly reflect cloud information.

[0075] In a possible implementation, the K-means algorithm is an unsupervised learning method in machine learning, which is a distance-based clustering algorithm. The core idea is to divide the data into K clusters through iteration, so that the data points in each cluster are as close to the cluster center as possible, and the data points between different clusters are as far away as possible. The K-means algorithm is mainly used to iteratively cluster multiple pixels of the same target position but different time sequences in the target remote sensing image time sequence, and divide the pixels into different cluster centers based on the reflectance spectral characteristics and NDVI of the pixels in several target bands. During clustering, the value of K can be determined according to the research purpose and characteristics of the target remote sensing image. If only two types of pixels, clear sky and non-clear sky, need to be distinguished, K can be set to 2. If further classification of the region is needed, a larger K value needs to be set. For multiple pixels of the same target position in the target remote sensing image time sequence, the distance of each pixel to the K cluster centers can be calculated, and the pixel is assigned to the cluster corresponding to the nearest cluster center. For each cluster, the cluster center is recalculated (usually the mean of all pixels in the cluster). Repeat the above steps until a certain stopping condition is met (such as the change of the cluster center being less than a specified threshold, reaching the maximum number of iterations, etc.). After the iteration ends, K cluster centers can be obtained.

[0076] In a possible implementation, for the obtained K cluster centers, the type of the cluster center, such as a cloud cluster center or a clear sky cluster center, can be identified by analyzing the reflectance spectral characteristics of the cluster center or analyzing the reflectance spectral characteristics of each pixel in the cluster center.

[0077] In a possible implementation, the pixels in the clear sky cluster center are all clear sky pixels of the target position, so a pixel can be randomly selected as the clear sky pixel of the target position. Of course, to obtain better clear sky data, the pixel with the most significant clear sky features can be selected from the cluster center as the clear sky pixel of the target position.

[0078] In a possible implementation, the above scheme can be used to obtain the clear sky pixels of each target position in the target remote sensing image. By synthesizing the clear sky pixels of each target position together, the clear sky image of the target region can be obtained.

[0079] The embodiment can construct a global remote sensing image with high quality in a global range by preprocessing remote sensing image data collected in a predetermined time period, and then selecting a target remote sensing image time sequence of a target region from a global remote sensing image time sequence; extracting reflectivity spectral characteristics of each pixel in a target waveband and NDVI of each pixel from the target remote sensing image time sequence; then using a K-means algorithm to cluster a plurality of pixels at the same target position but with different time sequences in the target remote sensing image time sequence, to obtain a plurality of clustering clusters; identifying a clear sky clustering cluster in the clustering clusters; selecting a clear sky pixel of the target position from the clear sky clustering cluster; and synthesizing a clear sky image of the target region based on the clear sky pixels of each target position in the target region, so that a high-quality clear sky image can be obtained. The scheme adopts the K-means algorithm, and only needs reflectivity spectral characteristics of a few target wavebands and NDVI of each pixel to cluster the clear sky pixels of the target position. The clear sky image of the target region can be synthesized based on the clear sky pixels of each target position in the target region. The data required for identifying the clear sky pixels is less, and is not affected by other external factors. The analysis scheme is simple and easy to implement, and maintains high efficiency and stability when processing large-scale remote sensing image data.

[0080] In a possible implementation, the remote sensing image data is L1 data of a MERSI (Medium Resolution Spectral Imager) payload of the Fengyun-3 satellite; the preprocessing of the remote sensing image data set to obtain a global remote sensing image time sequence in a predetermined time period comprises:

[0081] Performing atmospheric correction processing, equi-latitude and equi-longitude projection processing, and data splicing processing on the remote sensing image data collected each day to obtain a global remote sensing image each day;

[0082] Sorting the global remote sensing image each day in the predetermined time period to form a global remote sensing image time sequence.

[0083] In this embodiment, the remote sensing image data is L1 data of a MERSI (Medium Resolution Spectral Imager) payload of the Fengyun-3 satellite, which records multi-waveband and multi-spectrum remote sensing data, providing a rich data source for obtaining clear sky data of the present disclosure. These data sources not only contain basic information such as surface reflectivity and brightness temperature, but also contain rich spectral characteristics, providing an important basis for obtaining clear sky data.

[0084] In this embodiment, for the remote sensing image data of the FY-3 satellite MERSI load, the remote sensing image data needs to be processed by atmospheric correction, equirectangular projection and data splicing to obtain global remote sensing images of each day. The atmospheric correction is mainly to eliminate the influence of the atmosphere on the radiation transmission and restore the real reflectivity of the ground. The equirectangular projection is a kind of azimuthal projection, in which the meridians are projected as straight lines from the poles, and the latitudes are shown as concentric arcs with equal intervals. The equirectangular projection technology can be used to project the remote sensing image data into the same latitude and longitude coordinate system. The data splicing processing refers to splicing the remote sensing image data of the same time period together to form a complete global remote sensing image in the global range according to the latitude and longitude coordinates. After splicing, the global remote sensing images in the predetermined time period can be sorted in time sequence to form a time sequence of global remote sensing images.

[0085] In a possible implementation, the atmospheric correction processing includes:

[0086] Obtaining the remote sensing image data of the day from the remote sensing image data;

[0087] Based on the remote sensing image data of the day, obtaining the original reflectivity of each pixel in each target band and the angle information of each pixel in the remote sensing image data of the day, the angle information including the solar elevation angle, the solar zenith angle, the satellite elevation angle, the satellite zenith angle and the elevation data;

[0088] When the solar zenith angle meets the correction condition, correcting the original reflectivity of each pixel in the remote sensing image data of the day according to the angle information to obtain the atmospheric corrected remote sensing image data.

[0089] In this embodiment, when performing atmospheric correction, it is first judged whether the remote sensing image data is remote sensing image data of the day. If not, it is directly rejected. If yes, for each remote sensing image data of the day, the original reflectivity of each pixel in each target band and the angle information of each pixel in the remote sensing image data of the day are obtained. Then for each pixel, it is judged whether the solar zenith angle of the pixel meets the correction condition. If not, the remote sensing image data in which the pixel is located is directly rejected. If yes, the original reflectivity of the pixel is corrected according to the angle information of the pixel to obtain the real reflectivity of the ground of the pixel. The specific atmospheric correction calculation is clear to those skilled in the art and will not be described here.

[0090] In a possible implementation, the equirectangular projection processing includes:

[0091] Performing equirectangular projection on the atmospheric corrected remote sensing image data to obtain initial map projection data of a predetermined resolution;

[0092] The initial map projection data is interpolated to obtain map projection data.

[0093] In this embodiment, for the atmospheric corrected remote sensing image data of each day, 5 minutes of atmospheric corrected remote sensing image data can be obtained, and then the atmospheric corrected remote sensing image data is projected to equal latitude and longitude to obtain initial map projection data of a predetermined resolution, which is generally 1 km.

[0094] In this embodiment, some pixels in the initial map projection data obtained after projection can be blank, and in this case, an interpolation algorithm can be used to interpolate the initial map projection data to obtain map projection data. For example, 1 km of initial map projection data can be read, and for the 1 km of initial map projection data, block interpolation calculation can be performed based on the target band and satellite zenith angle thereof, and interpolation can be performed along the X / Y axis first, and then data corner point interpolation is performed, so that the interpolation is completed to obtain map projection data.

[0095] In a possible embodiment, the data splicing processing includes:

[0096] A target band of the map projection data is obtained.

[0097] For each target band, equal latitude and longitude information of the map projection data of the target band is obtained.

[0098] Based on the equal latitude and longitude information of the map projection data of the target band, a plurality of pieces of map projection data of the target band are combined into block data of a predetermined latitude and longitude span.

[0099] Based on the satellite zenith angle of the block data of the predetermined latitude and longitude span, the block data of the predetermined latitude and longitude span is combined into whole block data, and then the whole block data is spliced to obtain a global remote sensing image of the target band.

[0100] In this embodiment, when data stitching is performed, for the map projection data of each day, the target band of the map projection data can be determined first, and the map projection data of the same target band is stitched. For each target band, the equirectangular information (such as latitude and longitude information, resolution information, etc.) of the map projection data of the target band is obtained, and then, based on the equirectangular information of the map projection data of the target band and the satellite zenith angle, the map projection data of the target band is first combined into a plurality of pieces of data with a predetermined latitude and longitude span (such as 10 degrees or 11.25 degrees), and these pieces of data are located at different directions of the predetermined latitude and longitude span. The pieces of data with the predetermined latitude and longitude span can be combined into a whole piece of data with the predetermined latitude and longitude span, and then all the whole pieces of data are seamlessly stitched to obtain a global remote sensing image of the target band. In this way, the global remote sensing images of each target band can be obtained.

[0101] In a possible implementation, the target band has four bands, which are a blue band, a green band, a red band, and a near-infrared band.

[0102] In this embodiment, clouds usually have a high reflectivity in the visible light band (such as 0.4-0.76 μm) because they reflect most of the incident visible light. In the infrared band, especially the short-wave infrared and the medium-wave infrared (such as 1-3 μm), the water vapor and ice crystals inside the cloud will absorb and scatter light of these wavelengths, resulting in a significant decrease in the reflectivity of the cloud. The spectral curve of the cloud is usually flat in the visible light band, and the reflectivity is high; while in the infrared band, the reflectivity decreases sharply. This difference makes the cloud have a unique shape on the spectral curve graph, which is convenient for identification. The reflectivity spectrum of vegetation has obvious double-peak and double-valley characteristics in the visible light band. There is a reflection peak in the green band (about 0.55 μm), while there are two absorption bands in the blue band (about 0.45 μm) and the red band (about 0.67 μm). In the near-infrared band (such as 0.7-1.1 μm), the reflectivity of vegetation increases sharply, forming the so-called “red edge” feature. This is a unique spectral feature of vegetation and can be used to distinguish vegetation from other ground objects. Through research, the present disclosure adopts the reflectivity spectrum characteristics of the blue band, the green band, the red band, and the near-infrared band in the L1 data and the NDVI as the feature data for identifying clear skies.

[0103] Preferably, the target band can be band1 (0.47 μm), band2 (0.55 μm), band3 (0.65 μm), and band4 (0.865 μm). Of course, the value of the target band can be 0.01 μm left and right of the value of the above four bands, which is not limited herein.

[0104] In a possible implementation, the clustering of the plurality of pixels using the K-means algorithm comprises:

[0105] The clustering of the plurality of pixels using the K-means algorithm comprises two clustering clusters;

[0106] Identifying a suspected clear sky clustering cluster in the two clustering clusters;

[0107] The re-clustering of the pixels in the suspected clear sky clustering cluster using the K-means algorithm comprises three clustering clusters;

[0108] Identifying a clear sky clustering cluster in the three clustering clusters.

[0109] In this implementation, when the plurality of pixels are clustered using the K-means algorithm, the present disclosure can be clustered twice. In the first clustering, the K value is 2, two initial clustering centers are selected first, and then the plurality of pixels are iteratively assigned to the nearest clustering center according to their reflectance spectral characteristics and NDVI, and the clustering center is updated to the mean of the reflectance spectral characteristics and NDVI of the pixels in each category after each iteration. This process is repeated until the clustering center is stable or the preset number of iterations is reached. After the clustering is completed, by analyzing the reflectance spectral characteristics of each clustering center, one of the two clustering clusters is identified as a cloud area clustering cluster, and the other is identified as a suspected clear sky clustering cluster. Then, the pixels in the suspected clear sky clustering cluster with low reflectivity are clustered for the second time, the K value is 3, and three clustering clusters are obtained, and the clear sky clustering cluster in the three clustering clusters can be identified. For example, the mean of the NDVI of each pixel in each clustering cluster can be calculated, and the clustering cluster with the largest mean is taken as the clear sky clustering cluster.

[0110] The present disclosure can more accurately determine the clear sky pixels through two clustering, and further obtain a high-quality clear sky image.

[0111] In a possible implementation, the selecting of the clear sky pixel of the target position from the clear sky clustering cluster comprises:

[0112] The maximum NDVI pixel in the clear sky clustering cluster is selected as the clear sky pixel of the target position.

[0113] In this implementation, the pixels in the clear sky clustering cluster are all pixels at the same target position, and the maximum NDVI pixel in the clear sky clustering cluster is the pixel with the most prominent clear sky characteristics at the target position. This pixel can be selected as the clear sky pixel of the target position.

[0114] In order to prove the effectiveness and reliability of the present disclosure, the following verification can be performed:

[0115] Figure 2 Fig. 2 shows a schematic diagram of reflectance spectral characteristics and NDVI of the same target location pixel extracted from a target remote sensing image time series within a predetermined time period, Figure 2 For a target location pixel in a target remote sensing image of each day within the range of January 1st to January 31st, 2023, the values of reflectance of four different target bands (band1 (0.47 pm), band2 (0.55 pm), band3 (0.65 pm), band4 (0.865 pm)) and NDVI, i.e. 30 pixels of reflectance of four different target bands and NDVI, in order to distinguish different weather or cloud cover conditions from these data, the k-means clustering algorithm is used to cluster the 30 pixels twice, which roughly divides the 30 pixels into four categories: cloud, thin cloud, suspected cloud, and clear sky. As shown in Figure 2 Fig. 3, label = 1 is cloud, label = 3 is thin cloud, label = 4 is clear sky, and label = 5 is suspected cloud. Finally, the second day pixel with the maximum NDVI in the clear sky cluster, i.e. the label = 4 cluster, is selected as the clear sky pixel of the target location. In order to further verify its accuracy, the visual interpretation method is used, experts can directly identify and interpret the remote sensing image, by viewing the classification results of the target location pixel day by day, and comparing with the actual true color image, it is found that the classification results of the k-means method are basically consistent with the visual interpretation results, which shows the effectiveness and reliability of the k-means algorithm in the present disclosure.

[0116] The present disclosure also provides a remote sensing clear sky dataset acquisition device, Figure 3 Fig. 4 shows a structural block diagram of the remote sensing clear sky dataset acquisition device according to an embodiment of the present disclosure. The device can be realized as part or all of an electronic device through software, hardware, or a combination of the two. As shown in Figure 3 The remote sensing clear sky dataset acquisition device comprises:

[0117] An acquisition module 301 configured to acquire remote sensing image data collected within a predetermined time period;

[0118] A preprocessing module 302 configured to preprocess the remote sensing image data to obtain a global remote sensing image time series within a predetermined time period, wherein the global remote sensing image time series is a sequence of global remote sensing images arranged in chronological order;

[0119] A region selection module 303 configured to select a target remote sensing image time series of a target region from the global remote sensing image time series;

[0120] The feature extraction module 304 is configured to extract reflectivity spectral features of each pixel in a target waveband and normalized difference vegetation index (NDVI) of each pixel from the target remote sensing image time sequence.

[0121] The clustering module 305 is configured to cluster a plurality of pixels at the same target position but different time sequences in the target remote sensing image time sequence using a K-means algorithm to obtain a plurality of clustering clusters.

[0122] The identification module 306 is configured to identify a clear sky clustering cluster in the clustering clusters.

[0123] The pixel selection module 307 is configured to select a clear sky pixel of the target position from the clear sky clustering cluster.

[0124] The image synthesis module 308 is configured to synthesize a clear sky image of the target region based on the clear sky pixels of each target position in the target region.

[0125] In a possible implementation, the remote sensing image data is L1-level data of a medium resolution spectral imager (MERSI) payload of the Fengyun-3 satellite; and the preprocessing module is configured to:

[0126] perform atmospheric correction processing, equi-latitude-longitude projection processing, and data splicing processing on the collected remote sensing image data to obtain a plurality of global remote sensing images;

[0127] sort the plurality of global remote sensing images in the predetermined time period in time sequence to form a global remote sensing image time sequence.

[0128] In a possible implementation, the atmospheric correction processing in the preprocessing module is configured to:

[0129] obtain daytime remote sensing image data from the remote sensing image data;

[0130] obtain original reflectivity of each pixel in each target waveband and angle information of each pixel in the daytime remote sensing image data based on the daytime remote sensing image data, the angle information including a solar elevation angle, a solar zenith angle, a satellite elevation angle, a satellite zenith angle, and elevation data;

[0131] when the solar zenith angle meets a correction condition, correct the original reflectivity of each pixel in the daytime remote sensing image data according to the angle information to obtain atmospheric corrected remote sensing image data.

[0132] In a possible implementation, the equi-latitude-longitude projection processing in the preprocessing module is configured to:

[0133] Project the atmospheric corrected remote sensing image data to equal latitude and longitude to obtain initial map projection data of a predetermined resolution;

[0134] Interpolate the initial map projection data to obtain map projection data.

[0135] In a possible implementation, the data splicing processing in the preprocessing module is configured to:

[0136] Obtain target wavebands of the map projection data;

[0137] For each target waveband, obtain equal latitude and longitude information of the map projection data of the target waveband;

[0138] Integrate the map projection data of the target waveband into block data of a predetermined latitude and longitude span;

[0139] Based on the satellite zenith angle of the block data of the predetermined latitude and longitude span, integrate the block data of the predetermined latitude and longitude span into integral block data, and then splice the integral block data to obtain a global remote sensing image of the target waveband.

[0140] In a possible implementation, the target wavebands include a blue waveband, a green waveband, a red waveband, and a near-infrared waveband.

[0141] In a possible implementation, the clustering module is configured to:

[0142] Cluster the plurality of pixels using a K-means algorithm to obtain two cluster clusters;

[0143] Identify a suspected clear sky cluster cluster from the two cluster clusters;

[0144] Re-cluster the pixels in the suspected clear sky cluster cluster using a K-means algorithm to obtain three cluster clusters;

[0145] Identify a clear sky cluster cluster from the three cluster clusters.

[0146] In a possible implementation, the pixel selection module is configured to:

[0147] Select a pixel with the maximum NDVI from the clear sky cluster cluster as a clear sky pixel of the target position.

[0148] The technical terms and technical features mentioned in the device embodiments are the same as or similar to those mentioned in the above method embodiments, and the explanation and description of the technical terms and technical features involved in the device can refer to the explanation and description of the above method embodiments, which will not be repeated here.

[0149] The present disclosure also discloses an electronic device,Figure 4 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown.

[0150] As shown in Figure 4 the electronic device 400 includes a memory 401 and a processor 402, wherein the memory 401 is configured to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor 402 to implement a method according to an embodiment of the present disclosure.

[0151] Figure 5 A structural schematic diagram of a computer system suitable for implementing a method of an embodiment of the present disclosure is shown.

[0152] As shown in Figure 5 the computer system 500 includes a processing unit 501 which can perform various processes in the above-described embodiments according to programs stored in a read-only memory (ROM) 502 or loaded from a storage section 508 into a random access memory (RAM) 503. Various programs and data required for the operation of the computer system 500 are also stored in the RAM 503. The processing unit 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0153] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as necessary. A removable media 511 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 510 as necessary, so that a computer program read therefrom is installed into the storage section 508 as necessary. The processing unit 501 can be implemented as a CPU, a GPU, a TPU, a FPGA, a NPU, etc.

[0154] In particular, according to an embodiment of the present disclosure, the method described above can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product including computer instructions which, when executed by a processor, implement the method steps described above. In such an embodiment, the computer program product can be downloaded and installed from a network by the communication section 509, and / or installed from the removable media 511.

[0155] The flow and block diagrams in the drawings represent possible architectural, functional, and operational architectures of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block can represent a module, a segment, or a portion of code that comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustrations, and combinations thereof, can be implemented by special purpose hardware-based systems that perform the specified functions or operations, or combinations of special purpose hardware and computer instructions.

[0156] The units or modules described in the embodiments of the present disclosure can be implemented by software, or by programmable hardware. The described units or modules can also be arranged in a processor, and the names of the units or modules do not constitute a limitation on the units or modules themselves in some cases.

[0157] As another aspect, the present disclosure also provides a computer readable storage medium, which can be the computer readable storage medium included in the electronic device or the computer system in the above embodiments, or can exist separately from the device and not be assembled into the device. The computer readable storage medium stores one or more programs for execution by one or more processors to perform the methods described in the present disclosure.

[0158] The above description is merely preferred embodiments of the present disclosure and a description of principles of applied technologies. It should be understood by those skilled in the art that the scope of the present disclosure is not limited to the technical solutions formed by the specific combinations of the above technical features, and also includes other technical solutions formed by the combinations of the above technical features or equivalent features without departing from the inventive concept. For example, the above technical features can be replaced with the technical features disclosed in the present disclosure (but not limited to) having similar functions to form technical solutions.

Claims

1. A method for obtaining a remote sensing clear sky dataset, characterized in that, The method comprises the following steps: acquiring remote sensing image data collected in a predetermined time period; preprocessing the remote sensing image data to obtain a global remote sensing image time sequence in the predetermined time period, the global remote sensing image time sequence being a sequence of global remote sensing images arranged in time sequence; the preprocessing comprises atmospheric correction processing; the atmospheric correction processing comprises: acquiring daytime remote sensing image data from the remote sensing image data; based on the daytime remote sensing image data, acquiring the original reflectivity of each pixel in the daytime remote sensing image data in each target waveband and angle information, the angle information comprising the solar elevation angle, the solar zenith angle, the satellite elevation angle, the satellite zenith angle, and the elevation data; when the solar zenith angle satisfies a correction condition, correcting the original reflectivity of each pixel in the daytime remote sensing image data according to the angle information to obtain atmospheric corrected remote sensing image data; selecting a target remote sensing image time sequence of a target region from the global remote sensing image time sequence; extracting the reflectivity spectral characteristics of each pixel in a target waveband and the normalized difference vegetation index (NDVI) of each pixel from the target remote sensing image time sequence, the target waveband comprising a blue waveband, a green waveband, a red waveband, and a near-infrared waveband; for a plurality of pixels at the same target position but with different time sequences in the target remote sensing image time sequence, clustering the plurality of pixels based on the reflectivity spectral characteristics of each pixel in the target waveband and the normalized difference vegetation index (NDVI) of each pixel using a K-means algorithm to obtain a plurality of clustering clusters; identifying a clear sky clustering cluster in the clustering clusters; the clustering of the plurality of pixels using the K-means algorithm to obtain a plurality of clustering clusters and the identification of the clear sky clustering cluster in the clustering clusters comprise: clustering the plurality of pixels using the K-means algorithm to obtain two clustering clusters; identifying a suspected clear sky clustering cluster in the two clustering clusters; re-clustering the pixels in the suspected clear sky clustering cluster using the K-means algorithm to obtain three clustering clusters; identifying a clear sky clustering cluster in the three clustering clusters; selecting a clear sky pixel of the target position from the clear sky clustering cluster; synthesizing a clear sky image of the target region based on the clear sky pixels of each target position in the target region.

2. The acquisition method of claim 1, wherein, The remote sensing image data is L1-level data of a medium resolution spectral imager (MERSI) payload of a Fengyun-3 satellite; the preprocessing of the remote sensing image data to obtain a global remote sensing image time sequence in a predetermined time period comprises: performing atmospheric correction processing, equi-latitude and equi-longitude projection processing, and data splicing processing on the collected remote sensing image data to obtain a plurality of global remote sensing images; sequencing the plurality of global remote sensing images in a predetermined time period in time sequence to form a global remote sensing image time sequence.

3. The method of claim 2, wherein, The equi-latitude and equi-longitude projection processing comprises: performing equi-latitude and equi-longitude projection on the atmospheric corrected remote sensing image data to obtain initial map projection data of a predetermined resolution; performing interpolation on the initial map projection data to obtain map projection data.

4. The method of claim 3, wherein, The data splicing processing comprises: acquiring a target waveband of the map projection data; For each target waveband, obtain equal latitude and longitude information of map projection data of the target waveband; Based on the equal latitude and longitude information of the map projection data of the target waveband, integrate the multiple pieces of map projection data of the target waveband into piece data of a predetermined latitude and longitude span; Based on the satellite zenith angle of the piece data of the predetermined latitude and longitude span, integrate the piece data of the predetermined latitude and longitude span into whole piece data, and then stitch the whole piece data to obtain a global remote sensing image of the target waveband.

5. The method of claim 1, wherein, The selecting of the clear-sky pixel of the target position from the clear-sky cluster includes: Selecting a pixel with the maximum NDVI from the clear-sky cluster as the clear-sky pixel of the target position.

6. A device for acquiring remote sensing clear sky datasets, characterized in that, It includes: An acquisition module configured to acquire remote sensing image data collected in a predetermined time period; A preprocessing module configured to preprocess the remote sensing image data to obtain a global remote sensing image time sequence in the predetermined time period, the global remote sensing image time sequence being a sequence in which global remote sensing images are arranged in time sequence; The preprocessing includes atmospheric correction processing; the atmospheric correction processing includes: acquiring daytime remote sensing image data from the remote sensing image data; based on the daytime remote sensing image data, acquiring original reflectivity of each pixel in the daytime remote sensing image data in each target waveband and angle information, the angle information including solar elevation angle, solar zenith angle, satellite elevation angle, satellite zenith angle, and elevation data; when the solar zenith angle meets a correction condition, correcting the original reflectivity of each pixel in the daytime remote sensing image data according to the angle information to obtain atmospheric-corrected remote sensing image data; A region selection module configured to select a target remote sensing image time sequence of a target region from the global remote sensing image time sequence; A feature extraction module configured to extract reflectivity spectral features of each pixel in a target waveband and normalized difference vegetation index (NDVI) of each pixel from the target remote sensing image time sequence, the target waveband including blue light waveband, green light waveband, red light waveband, and near-infrared waveband; A clustering module configured to, for multiple pixels of the same target position but different time sequences in the target remote sensing image time sequence, cluster the multiple pixels based on reflectivity spectral features of each pixel in a target waveband and normalized difference vegetation index (NDVI) of each pixel using a K-means algorithm to obtain multiple clusters; the clustering of the multiple pixels using the K-means algorithm to obtain the multiple clusters includes: clustering the multiple pixels using the K-means algorithm to obtain two clusters; identifying a suspected clear-sky cluster in the two clusters; re-clustering the pixels in the suspected clear-sky cluster using the K-means algorithm to obtain three clusters; An identification module configured to identify a clear-sky cluster in the clusters; A pixel selection module configured to select a clear-sky pixel of the target position from the clear-sky cluster; An image synthesis module configured to synthesize a clear-sky image of the target region based on clear-sky pixels of each target position in the target region.

7. An electronic device, comprising: A computer program product comprising a memory for storing one or more computer instructions, wherein the one or more computer instructions are executed by a processor to implement the method of any one of claims 1 to 5.

8. A readable storage medium, characterized by, A computer program product comprising a memory for storing one or more computer instructions, wherein the one or more computer instructions are executed by a processor to implement the method of any one of claims 1 to 5.

Images