A method, apparatus, electronic device and storage medium for determining a target service

Abstract

The present invention discloses a method, apparatus, electronic device, and storage medium for determining a target service. The method includes: receiving a target selection area, and determining a target classification image corresponding to the target selection area according to a pre-established corresponding relationship; the corresponding relationship includes the mapping relationship between the area identifier, at least one remote sensing image compression file, the image to be used, and the image classification result; determining a target image to be used corresponding to the target selection area, and determining at least one closed sub-region in the target image to be used; determining the target pixel points corresponding to each closed sub-region in the target classification image, and determining the service attribute information corresponding to the target selection area according to each target pixel point, so as to determine the target service according to the service attribute information. The problem that risks occur in handling the corresponding service due to the inability to verify the authenticity of the service attribute information is solved, and the effect of providing a true and reliable basis for the development of the corresponding service is achieved.

Description

Technical Field

[0001] Embodiments of the present invention relate to the technical field of remote sensing satellites, and in particular, to a method, device, electronic device, and storage medium for determining a target service. Background Art

[0002] For the development of credit services, the business attribute information corresponding to each classification result can be determined according to the image classification result, and then the corresponding service can be determined according to the business attribute information.

[0003] Currently, when classifying different regions in an image, the method of manual marking can be used. However, such a statistical method usually requires a large number of personnel to survey the field site and obtain the corresponding area in the image, which not only consumes manpower and material resources, but also may have a large error in the actual statistical process, or the problem that the statistical update of the actual area corresponding to the image is not timely, resulting in a security risk problem when determining relevant services according to the business attribute information.

[0004] To solve the above problems, the actual area corresponding to different regions in the image can be statistically completed by intelligent means, and then the corresponding service can be determined according to the business attribute information corresponding to the classification result. Summary of the Invention

[0005] The present invention provides a method, device, electronic device, and storage medium for determining a target service, so as to provide a true and reliable basis for the development of corresponding services.

[0006] In a first aspect, an embodiment of the present invention provides a method for determining a target service, which is characterized by including:

[0007] Receiving a target boxed area, and determining a target classification image corresponding to the target boxed area according to a pre-established corresponding relationship; wherein, the corresponding relationship includes the mapping relationship between a region identifier, at least one remote sensing image compression file, an image to be used, and an image classification result;

[0008] Determining a target image to be used corresponding to the target boxed area, and determining at least one closed sub-region in the target image to be used;

[0009] Determining the target pixel points corresponding to each closed sub-region in the target classification image, and determining the business attribute information corresponding to the target boxed area according to each target pixel point, so as to determine the target service according to the business attribute information.

[0010] In a second aspect, an embodiment of the present invention further provides a device for determining a target service, which is characterized by including:

[0011] A target classification image determination module, configured to receive a target box selection area, and determine a target classification image corresponding to the target box selection area according to a pre-established corresponding relationship; wherein, the corresponding relationship includes a mapping relationship between a region identifier, at least one remote sensing image compression file, an image to be used, and an image classification result;

[0012] A closed sub-region determination module, configured to determine a target image to be used corresponding to the target box selection area, and determine at least one closed sub-region in the target image to be used;

[0013] A target service determination module, configured to determine target pixel points corresponding to each closed sub-region in the target classification image, and determine service attribute information corresponding to the target box selection area according to each target pixel point, so as to determine a target service according to the service attribute information.

[0014] In a third aspect, an embodiment of the present invention further provides an electronic device, where the electronic device includes:

[0015] One or more processors;

[0016] A storage device, configured to store one or more programs,

[0017] When the one or more programs are executed by the one or more processors, the one or more processors implement the method for determining a target service according to any one of the embodiments of the present invention.

[0018] In a fourth aspect, an embodiment of the present invention further provides a storage medium containing computer-executable instructions, where the computer-executable instructions are used to execute the method for determining a target service according to any one of the embodiments of the present invention when executed by a computer processor.

[0019] The technical solution of this embodiment receives a target boxed area, and determines a target classified image corresponding to the target boxed area according to a pre-established corresponding relationship; wherein, the corresponding relationship includes the mapping relationship between the area identifier, at least one remote sensing image compression file, the image to be used, and the image classification result. By dividing the obtained target classified image, the image to be used, that is, the agricultural land area and the non-agricultural land area, can be obtained, and the image to be used can be pixelated through a specific algorithm to obtain at least one closed sub-region. Determine the target image to be used corresponding to the target boxed area, and determine at least one closed sub-region in the target image to be used, and then determine the target classification result of the target pixel points corresponding to the closed sub-region, and the business attribute information corresponding to the target pixel points. Determine the target pixel points corresponding to each closed sub-region in the target classified image, and determine the business attribute information corresponding to the target boxed area according to each target pixel point, so as to determine the target business according to the business attribute information, and then when handling the business, the business attribute information can be used to complete the relevant verification work. It solves the problem that when handling business for customers, the authenticity of business attribute information cannot be accurately verified, resulting in the risk of handling corresponding business, and realizes providing a true and reliable basis for the development of corresponding business. Brief Description of the Drawings

[0020] In order to more clearly illustrate the technical solution of the exemplary embodiment of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the introduced drawings are only the drawings of a part of the embodiments to be described by the present invention, rather than all the drawings. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

[0021] Figure 1 It is a flowchart of a method for determining a target business provided by Embodiment 1 of the present invention;

[0022] Figure 2 It is a schematic flowchart of a method for determining a target business provided by Embodiment 2 of the present invention;

[0023] Figure 3 It is a schematic flowchart of a specific implementation process for determining a target business provided by Embodiment 3 of the present invention;

[0024] Figure 4 It is a satellite visualization image provided by Embodiment 3 of the present invention;

[0025] Figure 5 It is an agricultural land classification image corresponding to the satellite visualization image provided by Embodiment 3 of the present invention;

[0026] Figure 6Schematic structural diagram of an apparatus for determining a target service according to Embodiment 4 of the present invention.

[0027] Figure 7 Schematic structural diagram of an electronic device according to Embodiment 5 of the present invention. Detailed implementation manners

[0028] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the present invention, rather than limiting the present invention. In addition, it should be noted that, for the sake of description, only parts related to the present invention rather than all structures are shown in the accompanying drawings.

[0029] Embodiment 1

[0030] Figure 1 Schematic flow chart of a method for determining a target service according to Embodiment 1 of the present invention. This embodiment is applicable to the situation of providing a true and reliable basis for the development of corresponding services when handling credit-related services for customers. This method can be executed by an apparatus for determining a target service, and the apparatus can be implemented in the form of software and / or hardware. The hardware can be an electronic device, such as a mobile terminal or a PC, etc.

[0031] As Figure 1 shown, the method of this embodiment includes:

[0032] S110. Receive a target bounding box area, and determine a target classification image corresponding to the target bounding box area according to a pre-established corresponding relationship; wherein, the corresponding relationship includes a mapping relationship between a region identifier, at least one remote sensing image compression file, an image to be used, and an image classification result.

[0033] Among them, the target box selection area can be understood as the area selected by the user in the image and designated as the target. For example, when the user is handling a business and needs to use a certain area as collateral, the relevant business personnel can provide the user with a display device on which images of each area are displayed. The user can determine the area to be used as collateral by triggering the corresponding selection area on the display device, and the area selected by the user is used as the target box selection area. The classified image can be understood as marking the image areas with the same characteristics in the image, and then dividing the image according to different marking information to obtain multiple marked images; the image classification model used to obtain the classified image can include a convolutional neural network model, a residual shrinkage network model, an SVM neural network model, a small convolutional kernel neural network model, etc. The target classified image can be understood as the image corresponding to the target box selection area. For example, if a classified image of multiple cities across the country is included in an image, and each city's classified image contains classified images of multiple sub-city areas, at this time, if the area selected by the user's target box selection area is Beijing, the target classified image is the image corresponding to the Beijing area. The area identifier can be understood as the marking information corresponding to each city area. Exemplarily, by dividing each province and city in the country, multiple provinces and cities can be obtained, and each province and city can include multiple urban areas or multiple towns, etc. That is, the classified image of each province and city can include multiple classified images, each area identifier can include multiple sub-area identifiers, each area identifier can correspond to a classified image, and each classified image is marked, then each area can correspond to an area identifier. According to the area identifier, the corresponding area can be found. For example, the area identifier of Beijing can be set to "Jing", the area identifier of Tianjin can be set to "Jin", and the area identifier of Hebei can be set to "Ji", etc. In this embodiment, the remote sensing image can be understood as a satellite image that records the electromagnetic wave magnitudes of various ground objects, and the satellite image contains electromagnetic signal information of multiple bands collected by the sensor in the corresponding time and space. The image to be used can be understood as the image determined by the normalized difference vegetation index of each pixel point.

[0034] It should also be noted that it can also be a classified image corresponding to the entire province or city. According to the area identifier of the province or city, the classified image of the entire province or city corresponding to the area identifier can be obtained. For the convenience of subsequent processing, a region can contain multiple sub-area identifiers and the classified images corresponding to each area identifier.

[0035] Specifically, through model training, images can be classified into different categories, and according to actual needs, the images corresponding to the area identifiers are determined as target classification images. For example, the target classification image corresponding to the area identifier of agricultural land is an agricultural land image. Then, a correspondence between the target bounding area and the target classification image is established in advance. Exemplarily, the correspondence may include the mapping relationship between the area identifier, at least one remote sensing image compression file, the image to be used, and the image classification result. The user can draw a vector circle on the image to enclose the target bounding area. After the terminal receives the target bounding area, according to the pre-established correspondence between the target bounding area and the target classification image, the target classification image corresponding to the target bounding area is determined.

[0036] Optionally, determining the target classification image corresponding to the target bounding area according to the pre-established correspondence includes: determining the area identifier corresponding to the target bounding area, and retrieving the target classification image corresponding to the area identifier.

[0037] Specifically, taking a remote sensing image as an example, when the user selects the target bounding area, a visual remote sensing image can be used as the drawing background. Each area in the remote sensing image is marked in advance to determine the area identifier corresponding to each area, such as the area identifiers of each province, city or each administrative region, etc. The user circles the target bounding area in the remote sensing image, then determines the area identifier corresponding to the target bounding area, and according to the correspondence between the area identifier and each area in the target classification image, the target classification image corresponding to the area identifier can be retrieved.

[0038] S120. Determine the target image to be used corresponding to the target bounding area, and determine at least one closed sub-area in the target image to be used.

[0039] Among them, the target image to be used can be understood as the image after pixelization of the target bounding area. For example, on a display device, the remote sensing image corresponding to each area can be stored, and the image after pixelization of the remote sensing image can be used as the target image to be used. After the target image to be used is divided according to the characteristics of the objects in the image, each image corresponds to a sub-area in the target bounding area, and the characteristics of the objects included in each sub-area are the same or similar. The closed sub-area can be understood as the sub-area obtained by dividing the target image to be used according to the object characteristics, and the area whose boundary can form a complete closed line.

[0040] Specifically, according to the target bounding box area selected by the user on the terminal, after performing actual pixelization processing on the remote sensing image corresponding to the target bounding box area, a target image to be used corresponding to the target bounding box area can be obtained. Based on the target image to be used after pixelization processing, regional division is performed, and each pixel point is divided into different sub-regions according to similarity. Among them, a region with a complete closed line at the boundary of the divided image can be regarded as a closed sub-region, and at least one closed sub-region in the target image to be used is determined.

[0041] Exemplarily, after the user determines the target bounding box area on the terminal, pixelization processing is performed on the image corresponding to the target bounding box area to determine the target image to be used corresponding to the target bounding box area. The target image to be used may contain various different types of classified images. For example, in a remote sensing image, the characteristics of agricultural land can be characterized by flat land, and with the sowing season of crops, the area shows changes in crop replacement, etc.; the characteristics of urban land can be characterized by dense houses, a large number of vehicles, and extensive road distribution, etc.; the characteristics of forest land can be characterized by thick vegetation or a large green area, etc. Through different characteristics, the target image to be used can be divided into multiple images, including regional images such as agricultural land areas, urban land areas, and forest land areas. If the boundary in the divided image can form a complete closed line, it can be regarded as a closed sub-region. However, the characteristics contained in the divided image may not be unique. For example, in the image of urban land, there may be part of the image of agricultural land, or when dividing the target image to be used, due to the unclear characteristics of the objects in the image, the boundary of the divided image is not complete enough, then the sub-region corresponding to this image needs to be further processed to form a closed sub-region, and at least one closed sub-region in the target image to be used is determined.

[0042] Optionally, the determining the target image to be used corresponding to the target bounding box area and determining at least one closed sub-region in the target image to be used includes: retrieving the target remote sensing image file corresponding to the target bounding box area, and based on the target remote sensing image file, determining the target image to be used for the target bounding box area; filling the non-closed areas in the target image to be used based on the flood fill algorithm to obtain the target image for use; wherein, the target image for use includes at least one closed sub-region.

[0043] Among them, there can be many remote sensing image files collected by satellites. The target remote sensing image file can be understood as the remote sensing file corresponding to the target boxed area. Exemplarily, if the target boxed area of the user is a certain province or a certain city, the target remote sensing file is the remote sensing image file of the province or the city collected by the satellite. The flood filling algorithm can be understood as an image processing algorithm. For each marked pixel point in the image, according to the pre-set constraint conditions, through iterative calculation, the unmarked pixel points that meet the constraint conditions around the marked pixel points are marked until no new pixel points are marked. The non-closed area can be understood as the area corresponding to the image after the target image to be used is divided is not a complete area. The target image to be used can be understood as the image corresponding to the closed sub-area. The target image to be used includes at least one closed sub-area.

[0044] Specifically, after the user determines the target boxed area, the terminal can retrieve the target remote sensing image file corresponding to the target boxed area, and based on the target remote sensing image file, use the image corresponding to the target boxed area as the target image to be used. Then, specify pixel points in the target image to be used and set constraint conditions, and process the target image to be used through the flood filling algorithm to mark the pixel points with similar characteristics around the specified pixel points. Mark the marked pixel points in the same way, continue to find pixel points with similar characteristics around the pixel points, and after repeated iterative calculations, mark all pixel points that meet the constraint conditions until no new pixel points are marked. Based on the flood filling algorithm, the non-closed area in the target image to be used can be filled to obtain a closed sub-area, and the image in the target remote sensing image corresponding to the closed sub-area is the target image to be used. Among them, the target image to be used includes at least one closed sub-area.

[0045] S130. Determine the target pixel points corresponding to each closed sub-area in the target classification image, and determine the service attribute information corresponding to the target boxed area according to each target pixel point, so as to determine the target service according to the service attribute information.

[0046] Among them, the target pixel points can be understood as the pixel points that coincide between the closed sub-area and the target classification image; the service attribute information can be understood as the credit service information that can be handled in the area mortgaged by the user or corresponding service level information, etc.; the target service can be understood as the service that can be carried out corresponding to the service attribute information.

[0047] Specifically, by counting each pixel point in the image one by one, if the pixel point in the closed sub-region can find a corresponding pixel point in the target classification image, then this pixel point can be used as the target pixel point. Further, according to the target pixel points, the area corresponding to each closed region can be determined, and then according to the area of the image corresponding to each closed region, the business attribute information corresponding to the target box selection region can be determined, such as information like business credit rating. According to the business attribute information, the target business can be determined.

[0048] Optionally, the determining the target pixel points corresponding to each closed sub-region in the target classification image and determining the business attribute information corresponding to the target box selection region according to each target pixel point includes: determining the target classification results corresponding to each pixel point in the closed sub-region in the target classification image; using the pixel points with the target classification results consistent with the preset classification results as the target pixel points, determining the occupied area corresponding to each target pixel point, and determining the target occupied area according to the occupied area; determining the business attribute information of the target box selection region according to the target occupied area, and determining the target business based on the business attribute information.

[0049] Specifically, each pixel point in each closed region can be matched with the pixel points in the target classification map. If a pixel point belongs to both the closed sub-region and the target classification image at the same time, then this pixel point can be used as the target pixel point. According to the target classification image corresponding to the target pixel point, the target classification result corresponding to the target pixel point can be further determined. If the target classification result is consistent with the preset classification result, then this pixel point can be used as the target pixel point. Exemplarily, when the coordinates of a pixel point are located in the closed sub-region and at the same time a corresponding pixel point can be found in the target classification image, if the classification result of the target classification image is agricultural land and the preset classification result is agricultural land, then this pixel point can be determined as the target pixel point. Further, in the image, each pixel point can be used as the smallest element, and multiple pixel points connected together can aggregate into a connected region. By counting the number of pixel points in the connected region, the area of the connected region composed of pixel points can be determined. Then, according to the area composed of each target pixel point, the area occupied by the target box selection region can be obtained. Then, according to the corresponding relationship between the target box selection region and the target classification image, the area in the target classification image corresponding to the target box selection region can be determined, and the area corresponding to the obtained target classification image is used as the target occupied area. Then, according to the target occupied area, the business attribute information corresponding to the target box selection region is determined, and the target business is determined based on the business attribute information.

[0050] The technical solution of this embodiment is to receive a target boxed area and determine a target classified image corresponding to the target boxed area according to a pre-established corresponding relationship. Among them, the corresponding relationship includes the mapping relationship between the area identifier, at least one remote sensing image compression file, the image to be used, and the image classification result. By dividing the obtained target classified image, the images to be used can be obtained, that is, the agricultural land area and the non-agricultural land area. And through a specific algorithm, the images to be used can be pixelated to obtain at least one closed sub-region. Determine the target image to be used corresponding to the target boxed area, and determine at least one closed sub-region in the target image to be used, and then determine the target classification result of the target pixel point corresponding to the closed sub-region and the business attribute information corresponding to the target pixel point. Determine the target pixel points corresponding to each closed sub-region in the target classified image, and determine the business attribute information corresponding to the target boxed area according to each target pixel point, so as to determine the target business according to the business attribute information. Furthermore, when handling the business, the business attribute information can be used to complete the relevant verification work. This solves the problem that when handling business for customers, the authenticity of the business attribute information cannot be accurately verified, resulting in the risk of handling corresponding business, and realizes providing a true and reliable basis for the development of corresponding business.

[0051] Embodiment 2

[0052] As an optional embodiment of the above embodiment, Figure 2 It is a schematic flowchart of a method for determining a target business provided by Embodiment 2 of the present invention. Optionally, the method for determining a target business further includes: obtaining at least one remote sensing image compression file corresponding to each region to obtain a remote sensing image compression file package corresponding to each region; through cropping, splicing, and projection transformation processing of each remote sensing image compression file in the remote sensing image compression package of each region, obtaining the images to be processed corresponding to each region; through band component processing of each of the images to be processed, obtaining each image to be used; classifying and processing the images to be used in each region based on an image classification model to obtain the image classification result of the images to be used; establishing the corresponding relationship between the region identifier, at least one remote sensing image compression file, the image to be used, and the image classification result of each region, so as to determine the target business corresponding to the target boxed area based on the corresponding relationship.

[0053] As Figure 2 shown, the specific method includes:

[0054] S210. Obtain at least one remote sensing image compression file corresponding to each region to obtain a remote sensing image compression file package corresponding to each region.

[0055] Specifically, satellite remote sensing image compression files within a certain time range are obtained through specific remote sensing image acquisition channels. Among them, each region can correspond to one or more remote sensing image compression files. If a region corresponds to multiple remote sensing image compression files, they may be remote sensing image compression files of the same region collected at different times. If one wants to obtain remote sensing images of the region at different times, the region can correspond to multiple remote sensing compression files. Exemplarily, according to the harvest season of crops, remote sensing image compression files of crops corresponding to different harvest seasons can be collected. For example, in July of each year when wheat is ripe, remote sensing image compression files when wheat is ripe can be collected in July. Or, in different months, the crops planted in agricultural land in the same region may be different, and the growth conditions of the crops are also different. Remote sensing image compression files of agricultural land in each month can be collected. By establishing the region identifier of each region and the corresponding relationship with each obtained remote sensing image compression file, the remote sensing image compression files are screened to obtain at least one remote sensing image compression file. Further, at least one remote sensing image compression file is compressed to obtain a remote sensing image compression file package corresponding to each region.

[0056] S220. Through the processes of cropping, splicing, and projection transformation of each remote sensing image compression file in the remote sensing image compression packages of each region, the to-be-processed images corresponding to each region are obtained.

[0057] Among them, the projection transformation process can be understood as re-projecting the images in the satellite remote sensing images to align the satellite remote sensing images with the longitude and latitude directions. The methods of the projection transformation process can include rotation, resampling, and aligning the longitude and latitude directions, etc.

[0058] Specifically, after obtaining the remote sensing image compression file packages corresponding to each region and decompressing them locally, remote sensing image files can be obtained. At this time, the remote sensing image files may contain remote sensing images corresponding to at least one region. Since there may be positioning deviations when the remote sensing image device captures the remote sensing images corresponding to each region, the remote sensing images corresponding to each obtained region may cover too large an area or fail to cover the entire region. Therefore, the decompressed remote sensing image files can be cropped or spliced to obtain accurate remote sensing image files corresponding to the region. Exemplarily, if the region selected by the user at this time is Region A, but the remote sensing image file corresponding to Region A may include Region A, Region B, and Region C. Therefore, it is necessary to crop the remote sensing image file to remove the regions outside Region A and only retain the remote sensing image corresponding to Region A. Or when the region to be obtained is large, for example, when obtaining the remote sensing image file of a province and a remote sensing image file cannot cover the entire province, the remote sensing image files corresponding to the province can be spliced to obtain the remote sensing image file corresponding to the entire province. Further, perform projection transformation processing on the remote sensing images corresponding to the target region after cropping or splicing, including but not limited to operations such as rotation, resampling, and aligning the longitude and latitude directions, to obtain the images to be processed corresponding to each region. At this time, the length and width directions of the obtained images to be processed are consistent with the longitude and latitude directions.

[0059] S230. Obtain each image to be used by processing the band components of each of the images to be processed.

[0060] Among them, when collecting satellite remote sensing images, it is due to the different light intensities reflected by different parts of the ground objects, resulting in different degrees of photosensitivity on the photosensitive material and different tones of each part. And some bands in the light cannot penetrate the clouds. Therefore, it is necessary to perform component processing on the bands in the light. It can be known that different colors in the remote sensing image correspond to different bands. For example, the band corresponding to blue is about 493nm, the band corresponding to green is about 560nm, the band corresponding to red is about 665nm, and the band corresponding to near-infrared is about 833nm, including a total of 4 bands of light intensity information. By synthesizing different band components, a visual remote sensing image can be obtained.

[0061] Specifically, for the light intensity information of 4 bands, namely blue (band about 493nm), green (band about 560nm), red (band about 665nm), and near-infrared (band about 833nm) in each image to be processed, then process each band component, and then synthesize the red, green, and blue band components into a visual remote sensing image for storage to obtain each image to be used.

[0062] Optionally, obtaining each image to be used by processing each band component of the images to be processed includes: for each image to be processed, extracting the band components of each pixel in the current image to be processed in each band; determining the normalized difference vegetation index (NDVI) of each pixel according to the near-infrared band illumination intensity and the red band illumination intensity in the band components; and determining the image to be used according to the NDVI of each pixel.

[0063] Among them, the normalized difference vegetation index (NDVI) can be understood as one of the important parameters used to characterize the ground vegetation in satellite remote sensing images and reflect the growth and nutritional information of crops. The higher the value of NDVI, the higher the vegetation density and the better the health status.

[0064] Specifically, for each image to be processed, extract the band components of each pixel in the current image to be processed in each band. Further, the NDVI of each pixel can be calculated by dividing the difference between the reflectance value in the near-infrared band and the reflectance value in the red band of each pixel by the sum of the two, that is, NDVI = (NIR - R) / (NIR + R), where NIR represents the near-infrared band illumination intensity and R represents the red band illumination intensity. Calculate the NDVI of each pixel one by one, and save the corresponding NDVI value of each pixel to form a pixel-by-pixel index set. According to the NDVI of each pixel, the image to be used can be determined.

[0065] S240. Classify and process the images to be used in each region based on an image classification model to obtain the image classification result of the images to be used.

[0066] Among them, the image classification model can be understood as a model for classifying images, which can divide images with the same characteristics into the same category according to the characteristics in the images.

[0067] Specifically, input the images to be used in each region into the image classification model, and divide the images to be used with the same characteristics into one category according to the characteristics in the images to be used, and determine the image classification result of each image to be used.

[0068] Optionally, the classifying and processing the images to be used in each region based on an image classification model to obtain the image classification result of the images to be used includes: determining the to-be-determined image classification result corresponding to each pixel in each image to be used and the to-be-determined probability value corresponding to the to-be-determined image classification result based on the image classification model.

[0069] Among them, the image classification result to be determined can be understood as the image classification result obtained by the image classification model for the image to be processed by image processing. The image classification result to be determined may include the target image classification result and other image classification results. The probability value to be determined can be understood as the probability value that each pixel is determined to be the corresponding image classification result.

[0070] Specifically, different image classification results can be obtained by classifying each image to be used based on the image classification model, including the target image classification result and the non-target image classification result. Then, the value of each pixel in the image is analyzed, such as the value of the band corresponding to each pixel or the NDVI value corresponding to each pixel, so as to determine the image classification result to be determined corresponding to each pixel. The probability value to be determined corresponding to each pixel and its corresponding image classification result to be determined can also be determined through the image classification model.

[0071] Optionally, the classifying and processing each image to be used in each region based on the image classification model to obtain the image classification result of the image to be used includes: for each pixel, determining the probability value to be used for the current pixel according to the probability value to be determined of the current pixel, the average classification probability corresponding to the same image classification result, and the average total probability of the image to be used to which the current pixel belongs.

[0072] Among them, the average classification probability can be understood as the probability value corresponding to each pixel on average when the same image classification result corresponds to multiple pixels; the average total probability can be understood as the average value of each pixel of the image to be used; the probability value to be used can be understood as the probability value that each pixel can be used as a pixel in the image constituting the image classification result to be determined.

[0073] Specifically, the values of the band components corresponding to the pixels of agricultural land and non-agricultural land are different. After obtaining the NDVI image, for each pixel in the NDVI image, calculations are performed to obtain the probability value to be determined of the current pixel, the average classification probability corresponding to the same image classification result, and the average total probability of the image to be used to which the current pixel belongs. Rules are preset. For example, when the probability value to be determined of the current pixel is greater than the average classification probability and the average probability in the image to be used to which the current pixel belongs, the probability value to be determined of the current pixel can be determined as the probability value to be used.

[0074] Exemplarily, if the average total probability of the image to be used to which the current pixel belongs is 1 and the average classification probability corresponding to the same image classification result is 0.7, then for the pixel whose probability value to be determined of the current pixel is less than 0.7, and for the pixel whose probability value to be determined of the current pixel is greater than 0.7, it is used as the probability value to be used.

[0075] Optionally, the classification processing of the to-be-used images in each region by the image classification model to obtain the image classification result of the to-be-used images includes: updating the to-be-determined image classification result of each pixel point in the corresponding to-be-used image according to the to-be-used probability value of each pixel point to obtain the image classification result, and determining each sub-region in the to-be-used image based on the updated to-be-determined image classification result of each pixel point; wherein, the image classification result includes an agricultural land category and a non-agricultural land category.

[0076] Specifically, the to-be-used probability value can represent the value of the pixel point that can be used as the target image classification result. By judging whether the to-be-used probability value is greater than the classification probability mean value corresponding to the same image classification result and updating the pixel point value in the to-be-determined image according to the to-be-used probability value, the image classification result can be obtained. Based on the to-be-determined image classification result, each sub-region in the to-be-used image can be determined.

[0077] Exemplarily, the maximum inter-class variance method is used to automatically calculate the threshold of the NDVI image, and the NDVI image is segmented with the maximum segmentation threshold, that is, the classification probability mean value corresponding to the same image classification result, to divide the to-be-used image into each sub-region. If the classification probability mean value corresponding to the same image classification result is 0.7, then the pixel points in the NDVI image with pixel point values greater than 0.7 are replaced with 255 (white), and the pixel point values corresponding to the to-be-used probability values less than or equal to 0.7 are replaced with 0 (black) to obtain the binary agricultural land classification image. Among them, the pixel points with a value of 0 in the to-be-used image represent the non-agricultural land category, and the pixel points with a value of 255 represent the non-agricultural land category.

[0078] S250. Establish the correspondence relationship between the region identifier of each region, at least one remote sensing image compression file, the to-be-used image, and the image classification result, so as to determine the target service corresponding to the target boxed area based on the correspondence relationship.

[0079] Specifically, establishing the correspondence relationship between the region identifier of each region, at least one remote sensing image compression file, the to-be-used image, and the image classification result means that, in advance, a region identifier is set for each region, and each region can correspond to at least one remote sensing image compression file. Then, at least one remote sensing image compression file can be obtained according to the pre-set region identifier. After decompressing the selected remote sensing image compression file, through operations such as cropping, stitching, projection transformation processing, and band component processing, the to-be-used image can be obtained. Then, according to the image classification result corresponding to the target pixel point in the to-be-used image and the service attribute information corresponding to the target pixel point, after the user determines the target boxed area, based on the correspondence relationship, the target service corresponding to the target boxed area can be determined.

[0080] The technical solution of this embodiment is to obtain at least one remote sensing image compression file corresponding to each region, and obtain a remote sensing image compression file package corresponding to each region. By decompressing the obtained remote sensing image compression file package, the remote sensing compression files corresponding to each region can be obtained, which is convenient for processing the obtained remote sensing compression files. By performing cropping, splicing, and projection transformation processing on each remote sensing image compression file in the remote sensing image compression packages of each region, the images to be processed corresponding to each region are obtained. Further, the band components of each pixel point in each band of each image to be processed can be extracted. Based on the processing of the band components, the normalized difference vegetation index of each pixel point in each image to be processed can be determined, and the corresponding images to be used can be determined. By processing the band components of each of the images to be processed, the images to be used can be obtained, and the image classification results corresponding to each pixel point in each image to be used can be determined. Then, based on the corresponding relationship between each pixel point and the image to be used, the images to be used in each region are classified and processed based on the image classification model to obtain the image classification results of the images to be used. A corresponding relationship is established among the region identifiers of each region, at least one remote sensing image compression file, the images to be used, and the image classification results, so as to determine the target service corresponding to the target boxed area based on the corresponding relationship. Furthermore, when handling the service, the service attribute information can be used to complete relevant verification work. This solves the problem of being unable to accurately verify the authenticity of the service attribute information when handling services for customers, resulting in the risk problem of handling corresponding services, and realizes providing a true and reliable basis for the development of corresponding services.

[0081] Embodiment III

[0082] In a specific example, this technical solution can be processed by a satellite image preprocessing module and a box selection calculation module. See Figure 3 . Taking agricultural land as an example, the satellite image preprocessing module mainly includes a method for inferring the situation of agricultural land using satellite images. First, a satellite remote sensing image data source with the advantages of high spatial resolution and timely updated image data is obtained. See Figure 4, select samples with cloud cover less than 10% and appropriate time range, download and decompress them. Then, based on the downloaded satellite remote sensing images, perform preprocessing operations such as cropping, stitching, projection transformation, band extraction, and calculation of NDVI values after local decompression to obtain the remote sensing image file corresponding to the target selected area. Among them, cropping and stitching operations can make the corresponding area of the remote sensing image more accurate, and the projection transformation operation can adjust the length and width directions of the remote sensing image to be consistent with the longitude and latitude directions. Further, extract the band components corresponding to different colors in the remote sensing image. For example, extract the illumination intensity information of 4 bands, namely blue (band about 493nm), green (band about 560nm), red (band about 665nm), and near-infrared (band about 833nm) for the remote sensing image of the target area, calculate the normalized difference vegetation index value (NDVI), form a per-pixel index set, generate an NDVI image, and save the NDVI image in the database. It can be understood that the resolution of the visualized remote sensing image at this time is the same as that of the NDVI image. Use an artificial intelligence classification algorithm on the NDVI image to calculate the segmentation threshold that maximizes the between-class variance, and binarize the NDVI image with this threshold to form an agricultural land classification image, participating in Figure 5 , mainly based on NDVI and supplemented by manual confirmation and adjustment, use an image classification model, and based on the automatically generated maximum threshold as the classification threshold for agricultural land, and accordingly divide the satellite remote sensing image into agricultural land and non-agricultural land. The part with a value of 0 (black) in the image is the non-agricultural land part, and the part with a value of 255 (white) is the agricultural land part.

[0083] The circle selection calculation module mainly includes a calculation method for statistically calculating the agricultural land area in the circle selection graphic. Based on the remote sensing image as the drawing background, the user draws a VG format vector circle selection graphic in the remote sensing image of the terminal to circle out the target selected area in the remote sensing image, that is, the agricultural land circle selection data. Then use a specific algorithm to pixelize the image corresponding to the target selected area, and use the flood-fill algorithm to label the circled part. Statistically calculate the number of pixels that are both marked as agricultural land (that is, the target selected area) and marked as the circled part (that is, the enclosed sub-region) in the pixel image, and obtain the agricultural land area in the circle selection area by multiplying the number of pixels by the ground area occupied by a single pixel.

[0084] After obtaining the vector selection image drawn by the user on the terminal, the selected area is used as the target selection area, and the target classification image corresponding to the target selection area is divided into multiple small images. Then, a specific algorithm is used to pixelate each image, that is, the target classification image corresponding to the target selection area is drawn into the pixel image. The selection image uses the visual remote sensing image as the drawing background, and it is determined that there is a corresponding relationship between the target selection area, the visual remote sensing image, the NDVI image, and the agricultural land classification image. The resolution after pixelation is also consistent with the resolution of the target classification image. Then, the flood filling algorithm is used to fill the non-closed area after pixelation until all sub-areas are filled into closed sub-areas. The target classification result corresponding to each pixel point in the closed sub-area is determined, and then the pixel points consistent with the preset classification result are used as target pixel points. Then, a specific algorithm is used to statistically determine whether two conditions are met in the agricultural land classification image for each pixel: (1) the coordinates of the target pixel point are within the closed sub-area; (2) the occupied area corresponding to the target pixel point is the product of the number of pixel points and the area occupied by a single pixel. For example, if the length and width of the ground part corresponding to the pixel are both 10 meters, the area is 100 square meters = 0.15 mu. Then, the area of the agricultural land part within the target selection area is obtained.

[0085] Finally, the target service is determined according to the service attribute information when handling the service.

[0086] The technical solution of this embodiment receives the target selection area and determines the target classification image corresponding to the target selection area according to the pre-established corresponding relationship; wherein, the corresponding relationship includes the mapping relationship between the area identifier, at least one remote sensing image compression file, the image to be used, and the image classification result. By dividing the obtained target classification image, the image to be used can be obtained, that is, the agricultural land area and the non-agricultural land area, and the image to be used can be pixelated through a specific algorithm to obtain at least one closed sub-area. Determine the target image to be used corresponding to the target selection area, and determine at least one closed sub-area in the target image to be used, and then determine the target classification result of the target pixel point corresponding to the closed sub-area, as well as the service attribute information corresponding to the target pixel point. Determine the target pixel points corresponding to each closed sub-area in the target classification image, and determine the service attribute information corresponding to the target selection area according to each target pixel point, so as to determine the target service according to the service attribute information, and then when handling the service, the service attribute information can be used to complete the relevant verification work. It solves the problem that when handling services for customers, it is impossible to accurately check the authenticity of the service attribute information, resulting in the risk of handling corresponding services, and realizes providing a true and reliable basis for the development of corresponding services.

[0087] Embodiment 4

[0088] Figure 6 A device for determining a target service provided in Embodiment 4 of the present invention, the device includes: a target classification image determination module 410, a closed sub-region determination module 420, and a target service determination module 430.

[0089] The target classification image determination module 410 is configured to receive a target boxed area and determine a target classification image corresponding to the target boxed area according to a pre-established corresponding relationship; wherein, the corresponding relationship includes a mapping relationship between a region identifier, at least one remote sensing image compression file, an image to be used, and an image classification result;

[0090] The closed sub-region determination module 420 is configured to determine a target image to be used corresponding to the target boxed area and determine at least one closed sub-region in the target image to be used;

[0091] The target service determination module 430 is configured to determine target pixel points corresponding to each closed sub-region in the target classification image, and determine service attribute information corresponding to the target boxed area according to each target pixel point, so as to determine the target service according to the service attribute information.

[0092] The technical solution of this embodiment receives a target boxed area and determines a target classification image corresponding to the target boxed area according to a pre-established corresponding relationship; wherein, the corresponding relationship includes a mapping relationship between a region identifier, at least one remote sensing image compression file, an image to be used, and an image classification result. By dividing the obtained target classification image, an image to be used can be obtained, that is, an agricultural use area and a non-agricultural use area, and the image to be used can be pixelated through a specific algorithm to obtain at least one closed sub-region. Determine a target image to be used corresponding to the target boxed area and determine at least one closed sub-region in the target image to be used, and then determine the target classification result of the target pixel points corresponding to the closed sub-region and the service attribute information corresponding to the target pixel points. Determine the target pixel points corresponding to each closed sub-region in the target classification image, and determine the service attribute information corresponding to the target boxed area according to each target pixel point, so as to determine the target service according to the service attribute information. Furthermore, when handling a service, the service attribute information can be used to complete relevant verification work. It solves the problem that when handling a service for a customer, the authenticity of the service attribute information cannot be accurately verified, resulting in the risk problem of handling the corresponding service, and realizes providing a true and reliable basis for the development of the corresponding service.

[0093] Based on any optional technical solution in the embodiments of the present invention, optionally, the device for determining a target service further includes:

[0094] The compressed file package determination module is used to obtain at least one remote sensing image compressed file corresponding to each region, and obtain a remote sensing image compressed file package corresponding to each region;

[0095] The image to be processed determination module is used to obtain the images to be processed corresponding to each region by performing cropping, splicing, and projection transformation on each remote sensing image compressed file in the remote sensing image compressed packages of each region;

[0096] Each image to be used determination module is used to obtain each image to be used by performing band component processing on each of the images to be processed;

[0097] The image classification result determination module is used to perform classification processing on the images to be used in each region based on an image classification model, and obtain the image classification result of the images to be used;

[0098] The second target service determination module is used to establish the corresponding relationship among the region identifiers of each region, at least one remote sensing image compressed file, the images to be used, and the image classification results, and determine the target service corresponding to the target boxed region based on the corresponding relationship.

[0099] Based on any optional technical solution in the embodiment of the present invention, optionally, the each image to be used determination module specifically includes:

[0100] The band component extraction sub-module is used to extract the band components of each pixel point in each band of the current image to be processed for each image to be processed;

[0101] The vegetation index determination sub-module is used to determine the normalized vegetation index of each pixel point according to the illumination intensity of the near-infrared band and the illumination intensity of the red band in the band components;

[0102] The image to be used determination sub-module is used to determine the image to be used according to the normalized vegetation index of each pixel point.

[0103] Based on any optional technical solution in the embodiment of the present invention, optionally, the image classification result determination module specifically includes:

[0104] The to-be-determined probability value determination sub-module is used to determine the to-be-determined image classification result corresponding to each pixel point in each image to be used and the to-be-determined probability value corresponding to the to-be-determined image classification result based on the image classification model;

[0105] The to-be-used probability value determination sub-module is used to determine the to-be-used probability value of the current pixel point for each pixel point according to the to-be-determined probability value of the current pixel point, the classification probability mean value corresponding to the same image classification result, and the total probability mean value of the to-be-used image to which the current pixel point belongs;

[0106] A sub - region determining sub - module, configured to update the to - be - determined image classification results of each pixel in the corresponding to - be - used image according to the to - be - used probability values of each pixel, obtain the image classification results, and determine each sub - region in the to - be - used image based on the updated to - be - determined image classification results of each pixel; wherein, the image classification results include agricultural land categories and non - agricultural land categories.

[0107] Based on any optional technical solution in the embodiments of the present invention, optionally, the target classification image determining module is configured to:

[0108] Determine the region identifier corresponding to the target box - selected region, and retrieve the target classification image corresponding to the region identifier

[0109] Based on any optional technical solution in the embodiments of the present invention, optionally, the closed sub - region determining module specifically includes:

[0110] A target to - be - used image determining sub - module, configured to retrieve the target remote sensing image file corresponding to the target box - selected region, and determine the target to - be - used image of the target box - selected region based on the target remote sensing image file;

[0111] A target used - image determining sub - module, configured to fill the non - closed regions in the target to - be - used image based on the flood - fill algorithm to obtain the target used - image; wherein, the target used - image includes at least one closed sub - region.

[0112] Based on any optional technical solution in the embodiments of the present invention, optionally, the first target service determining module specifically includes:

[0113] A target classification result determining sub - module, configured to determine the target classification results corresponding to each pixel in the closed sub - region in the target classification image;

[0114] A target occupied area determining sub - module, configured to use the pixels whose target classification results are consistent with the preset classification results as target pixels, determine the occupied areas corresponding to each target pixel, and determine the target occupied area according to the occupied areas;

[0115] A target service determining sub - module, configured to determine the service attribute information of the target box - selected region according to the target occupied area, and determine the target service based on the service attribute information.

[0116] The device for determining the target service provided by the embodiments of the present invention can execute the method for determining the target service provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0117] It should be noted that the various units and modules included in the above-mentioned device are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the embodiments of the present invention.

[0118] Embodiment Five

[0119] Figure 7 FIG. is a schematic structural diagram of an electronic device provided in Embodiment Five of the present invention. Figure 7 FIG. shows a block diagram of an exemplary electronic device 40 suitable for use in implementing the embodiments of the present invention. Figure 7 The shown electronic device 40 is merely an example and should not impose any limitation on the functions and usage scope of the embodiments of the present invention.

[0120] As Figure 7 shown, the electronic device 40 is presented in the form of a general-purpose computing device. The components of the electronic device 40 may include, but are not limited to: one or more processors or processing units 401, a system memory 402, and a bus 403 connecting different system components (including the system memory 402 and the processing unit 401).

[0121] The bus 403 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus structures. By way of example, these architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

[0122] The electronic device 40 typically includes a variety of computer system-readable media. These media can be any available media accessible by the electronic device 40, including volatile and non-volatile media, removable and non-removable media.

[0123] The system memory 402 may include computer system-readable media in the form of volatile memory, such as random access memory (RAM) 404 and / or cache memory 405. The electronic device 40 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, a storage system 406 may be used for reading and writing non-removable, non-volatile magnetic media ( Figure 7 not shown, typically referred to as a "hard disk drive"). Although Figure 7Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (such as a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (such as a CD-ROM, DVD-ROM or other optical medium) can be provided. In these cases, each drive can be connected to the bus 403 through one or more data medium interfaces. The memory 402 can include at least one program product having a set (such as at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0124] A program / utility 408 having a set (at least one) of program modules 407 can be stored in, for example, the memory 402. Such program modules 407 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. The implementation of a network environment may be included in each or some combination of these examples. The program modules 407 generally perform the functions and / or methods in the embodiments described in the present invention.

[0125] The electronic device 40 can also communicate with one or more external devices 409 (such as a keyboard, a pointing device, a display 410, etc.), and can also communicate with one or more devices that enable a user to interact with the electronic device 40, and / or communicate with any device that enables the electronic device 40 to communicate with one or more other computing devices (such as a network card, a modem, etc.). Such communication can be carried out through an input / output (I / O) interface 411. Moreover, the electronic device 40 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 412. As shown in the figure, the network adapter 412 communicates with other modules of the electronic device 40 through the bus 403. It should be understood that although Figure 7 not shown in the figure, other hardware and / or software modules can be used in combination with the electronic device 40, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0126] The processing unit 401 executes various functional applications and data processing by running the programs stored in the system memory 402, such as implementing the method for determining a target service provided by the embodiments of the present invention.

[0127] Embodiment Six

[0128] Embodiment 6 of the present invention further provides a storage medium containing computer-executable instructions, and the computer-executable instructions are used to execute a method for determining a target service when executed by a computer processor. The method includes: receiving a target box selection area, and determining a target classification image corresponding to the target box selection area according to a pre-established correspondence; wherein, the correspondence includes a mapping relationship between a region identifier, at least one remote sensing image compression file, an image to be used, and an image classification result; determining a target image to be used corresponding to the target box selection area, and determining at least one closed sub-region in the target image to be used; determining target pixel points corresponding to each closed sub-region in the target classification image, and determining service attribute information corresponding to the target box selection area according to each target pixel point, so as to determine a target service according to the service attribute information.

[0129] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer-readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device.

[0130] The computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which computer-readable program code is carried. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than the computer-readable storage medium, and the computer-readable medium may send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device.

[0131] The program code contained on the computer-readable medium may be transmitted by any appropriate medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0132] Computer program code for performing the operations of the embodiments of the present invention can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0133] Note that the above is only the preferred embodiment of the present invention and the applied technical principles. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein. Various obvious changes, re-adjustments, and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments. Without departing from the concept of the present invention, more other equivalent embodiments can be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for determining a target service, characterized in that, including: Receiving a target box selection area, and determining a target classification image corresponding to the target box selection area according to a pre-established corresponding relationship; wherein, the corresponding relationship includes a mapping relationship between a region identifier, at least one remote sensing image compression file, an image to be used, and an image classification result; Determining a target image to be used corresponding to the target box selection area, and determining at least one enclosed sub-region in the target image to be used; Determining target pixel points corresponding to each enclosed sub-region in the target classification image, and determining service attribute information corresponding to the target box selection area according to each target pixel point, so as to determine a target service according to the service attribute information; obtaining at least one remote sensing image compression file corresponding to each region, and obtaining a remote sensing image compression file package corresponding to each region; Through processes of cropping, splicing, and projection transformation on each remote sensing image compression file in the remote sensing image compression packages of each region, obtaining a to-be-processed image corresponding to each region; Through processing the band components of each to-be-processed image, obtaining each image to be used; Based on an image classification model, classifying and processing the images to be used in each region, and obtaining an image classification result of the images to be used; Establishing a corresponding relationship between the region identifier, at least one remote sensing image compression file, the image to be used, and the image classification result of each region, so as to determine a target service corresponding to the target box selection area based on the corresponding relationship; wherein, the classifying and processing the images to be used in each region based on the image classification model, and obtaining the image classification result of the images to be used includes: Based on the image classification model, determining a to-be-determined image classification result corresponding to each pixel point in each image to be used, and a to-be-determined probability value corresponding to the to-be-determined image classification result; For each pixel point, determining a probability value to be used for the current pixel point according to the to-be-determined probability value of the current pixel point, the average classification probability corresponding to the same image classification result, and the total probability average value of the image to be used to which the current pixel point belongs; According to the probability values to be used for each pixel point, updating the to-be-determined image classification results of each pixel point in the corresponding image to be used, obtaining the image classification result, and determining each sub-region in the image to be used based on the updated to-be-determined image classification results of each pixel point; wherein, the image classification result includes an agricultural land category and a non-agricultural land category; wherein, the determining target pixel points corresponding to each enclosed sub-region in the target classification image, and determining service attribute information corresponding to the target box selection area according to each target pixel point includes: Determining a target classification result corresponding to each pixel point in the enclosed sub-region in the target classification image; Taking the pixel points whose target classification results are consistent with the preset classification results as target pixel points, determining the occupied area corresponding to each target pixel point, and determining a target occupied area according to the occupied area; Determining service attribute information corresponding to the target box selection area according to the target occupied area, and determining a target service based on the service attribute information.

2. The method according to claim 1, characterized in that, The obtaining of each image to be used by processing the band components of each image to be processed includes: For each image to be processed, extracting the band components of each pixel in each band of the current image to be processed; Determining the normalized difference vegetation index of each pixel according to the near-infrared band illumination intensity and the red band illumination intensity in the band components; Determining the image to be used according to the normalized difference vegetation index of each pixel.

3. The method according to claim 1, characterized in that, The determining of the target classification image corresponding to the target boxed area according to the pre-established corresponding relationship includes: Determining the area identifier corresponding to the target boxed area and retrieving the target classification image corresponding to the area identifier.

4. The method according to claim 1, wherein The determining of the target image to be used corresponding to the target boxed area and determining at least one enclosed sub-area in the target image to be used includes: Retrieving the target remote sensing image file corresponding to the target boxed area and determining the target image to be used for the target boxed area based on the target remote sensing image file; Filling the non-enclosed areas in the target image to be used based on the flood fill algorithm to obtain the target image for use; wherein, the target image for use includes at least one enclosed sub-area.

5. An apparatus for determining a target service, characterized in that, It includes: A target classification image determination module, configured to receive a target boxed area and determine a target classification image corresponding to the target boxed area according to a pre-established corresponding relationship; wherein, the corresponding relationship includes the mapping relationship between the area identifier, at least one remote sensing image compression file, the image to be used, and the image classification result; An enclosed sub-area determination module, configured to determine a target image to be used corresponding to the target boxed area and determine at least one enclosed sub-area in the target image to be used; A first target service determination module, configured to determine the target pixels corresponding to each enclosed sub-area in the target classification image and determine the service attribute information corresponding to the target boxed area according to each target pixel, so as to determine the target service according to the service attribute information; The apparatus for determining the target service further includes: A compression file package determination module, configured to obtain at least one remote sensing image compression file corresponding to each area and obtain a remote sensing image compression file package corresponding to each area; An image to be processed determination module, configured to obtain the images to be processed corresponding to each area through cropping, splicing, and projection transformation processing of each remote sensing image compression file in the remote sensing image compression packages of each area; Each image to be used determination module, configured to obtain each image to be used by processing the band components of each of the images to be processed; An image classification result determination module, configured to perform classification processing on the images to be used in each area based on an image classification model to obtain the image classification result of the images to be used; A second target service determination module, configured to establish the corresponding relationship between the area identifier, at least one remote sensing image compression file, the image to be used, and the image classification result of each area, so as to determine the target service corresponding to the target boxed area based on the corresponding relationship; The image classification result determination module specifically includes: A to-be-determined probability value determination sub-module, configured to determine, based on the image classification model, a to-be-determined image classification result corresponding to each pixel in each to-be-used image, and a to-be-determined probability value corresponding to the to-be-determined image classification result; A to-be-used probability value determination sub-module, configured to, for each pixel, determine the to-be-used probability value of the current pixel according to the to-be-determined probability value of the current pixel, the classification probability mean value corresponding to the same image classification result, and the total probability mean value of the to-be-used image to which the current pixel belongs; A sub-region determination sub-module, configured to update the to-be-determined image classification result of each pixel in the corresponding to-be-used image according to the to-be-used probability value of each pixel, obtain the image classification result, and determine each sub-region in the to-be-used image based on the updated to-be-determined image classification result of each pixel; wherein, the image classification result includes an agricultural land category and a non-agricultural land category; The first target service determination module specifically includes: A target classification result determination sub-module, configured to determine a target classification result corresponding to each pixel in the closed sub-region in the target classification image; A target occupied area determination sub-module, configured to use the pixels whose target classification results are consistent with the preset classification result as target pixels, determine the occupied area corresponding to each target pixel, and determine the target occupied area according to the occupied area; A target service determination sub-module, configured to determine the service attribute information of the target bounding box area according to the target occupied area, and determine the target service based on the service attribute information; 6. An electronic device, characterized in that, The electronic device includes: One or more processors; A storage device, configured to store one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method for determining a target service as described in any one of claims 1-4.

7. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions are used to execute the method for determining a target service as described in any one of claims 1-4 when executed by a computer processor.

Images