Geographic view positioning method and apparatus, electronic device, and storage medium

Abstract

The application particularly relates to a geographic view positioning method and device, electronic equipment and a storage medium, wherein the method comprises the following steps: collecting a ground panoramic image of a region to be positioned and extracting first basic feature data of the ground panoramic image; determining a remote sensing image database of the region to be positioned and extracting second basic feature data of remote sensing images in the remote sensing image database; jointly fitting and predicting the first basic feature data and preset environment parameters; constructing a ground image environment variable coefficient model; performing spatial recovery after determining the environment coefficient of the region to be positioned; obtaining static and dynamic ground panoramic images satisfying preset consistency feature conditions; extracting third basic feature data of the static ground panoramic image; and performing near neighbor matching in combination with the second basic feature data to obtain the best remote sensing image to be positioned. Thus, the problem that the positioning system is affected in real environment use due to the deviation of image representation caused by the difference in image collection time or region is solved.

Description

Technical Field

[0001] This application relates to the field of computer vision, and in particular to a geographic view positioning method, apparatus, electronic device, and storage medium. Background Technology

[0002] Geographic view data is affected by factors such as illumination, visibility, and atmospheric scattering under different weather conditions. Geographic view data needs to be located using geographic image positioning technology. This involves extracting and matching features from different images taken on the ground and in the sky to obtain the best matching sky image. The latitude, longitude, and altitude information carried in the sky image are then used to achieve the location.

[0003] In related technologies, geographic image positioning technology involves characterizing and matching ground images and remote sensing images under the same lighting and meteorological conditions.

[0004] However, when images are acquired at different times and in different regions, there will be deviations in image representation, which will lead to errors in the matching results and thus affect the use of the positioning system in real-world environments. Summary of the Invention

[0005] This application provides a geographic view positioning method, apparatus, electronic device, and storage medium to solve the problem that when the image acquisition time or region is different, the image representation will be biased, which will cause errors in the matching results and thus affect the use of the positioning system in the real environment.

[0006] The first aspect of this application provides a geographic view positioning method, including the following steps:

[0007] Collect panoramic ground images of the area to be located and determine a remote sensing image database containing the area to be located;

[0008] First basic feature data is obtained by extracting features from the ground panoramic image, and second basic feature data is obtained by extracting features from all remote sensing images in the remote sensing image database.

[0009] By combining the first basic feature data of the ground panoramic image with preset environmental parameters, a ground image environmental variable coefficient model corresponding to the area to be located is constructed through fitting and prediction.

[0010] Based on the ground image environmental variable coefficient model, the environmental coefficients of the area to be located are determined, and spatial reconstruction based on environmental weights is performed according to the environmental coefficients to obtain static and dynamic ground panoramic images that meet preset consistency feature conditions; and

[0011] The static ground panoramic image is subjected to feature extraction to obtain third basic feature data, and Euclidean distance-based nearest neighbor matching is performed based on the third basic feature data and the second basic feature data to obtain the optimal remote sensing image to be located.

[0012] According to one embodiment of this application, the acquisition of the ground panoramic image of the area to be located includes:

[0013] Based on the area to be located, a preset environmental parameter model is constructed;

[0014] Based on multiple acquisition environments in the preset environmental parameter model, multiple panoramic ground images of the area to be located are acquired, wherein the acquisition environment includes multiple illumination parameters and / or multiple meteorological parameters.

[0015] According to one embodiment of this application, the step of fitting and predicting the first basic feature data of the ground panoramic image and preset environmental parameters to construct a ground image environmental variable coefficient model corresponding to the area to be located includes:

[0016] Based on the first basic feature data of multiple ground panoramic images of the area to be located, the average feature data of the first basic feature data of the multiple ground panoramic images is obtained.

[0017] Based on the difference between the first basic feature data and the average feature data, the dynamic features of each ground panoramic image based on environmental changes are obtained, and a ground image environmental variable coefficient model is constructed according to the response of the dynamic features in the ground panoramic image.

[0018] According to one embodiment of this application, the step of determining the environmental coefficient of the area to be located based on the ground image environmental variable coefficient model, and performing spatial reconstruction based on environmental weights on the fourth basic feature data according to the environmental coefficient to obtain a static ground panoramic image and a dynamic ground panoramic image that satisfy the preset consistency feature conditions includes:

[0019] Based on multiple panoramic ground images of the area to be located, and combined with the preset environmental parameters, feature extraction is performed on each panoramic ground image of the area to be located to obtain the fourth basic feature data.

[0020] Based on the environmental coefficient of the area to be located, the fourth basic feature data is subjected to environmental transformation to recover static and dynamic ground panoramic images that meet the preset consistency feature conditions.

[0021] According to one embodiment of this application, the step of extracting features from the static ground panoramic image to obtain third basic feature data, and performing nearest neighbor matching based on Euclidean distance based on the third basic feature data and the second basic feature data to obtain the optimal remote sensing image to be located, includes:

[0022] The third basic feature data of the static ground panoramic image is extracted and matched with the second basic feature data based on Euclidean distance to obtain the best positioning result;

[0023] Based on the static ground panoramic image that meets the preset consistency feature conditions, the static ground panoramic image and the second basic feature data are matched with nearest neighbors based on Euclidean distance to obtain the first distance data between each static ground panoramic image and all remote sensing images in the remote sensing image database.

[0024] Based on the dynamic ground panoramic image that meets the preset consistency feature conditions, the dynamic ground panoramic image and the second basic feature data are matched with the nearest neighbor based on Euclidean distance to obtain the second distance data between each dynamic ground panoramic image and all remote sensing images in the remote sensing image database.

[0025] By combining the first distance data and the second distance data, the comprehensive distance data between each dynamic ground panoramic image and all remote sensing images in the remote sensing image database is obtained, and the image with the smallest comprehensive distance is selected as the best remote sensing image to be located.

[0026] According to the geographic view positioning method of this application, a ground panoramic image of the area to be positioned is acquired and its first basic feature data is extracted. A remote sensing image database containing the area to be positioned is determined, and the second basic feature data of all remote sensing images is extracted. The first basic feature data and preset environmental parameters are combined for fitting and prediction to construct a ground image environmental variable coefficient model corresponding to the area to be positioned. Then, the environmental coefficient of the area to be positioned is determined for spatial restoration, resulting in a static ground panoramic image and a dynamic ground panoramic image that meet preset consistency feature conditions. The static ground panoramic image is used to extract features to obtain the third basic feature data. At the same time, the second basic feature data is combined with Euclidean distance-based nearest neighbor matching to obtain the optimal remote sensing image to be positioned. This solves the problem that when the image acquisition time or region is different, the image representation will be biased, resulting in errors in the matching results and affecting the use of the positioning system in real environment. By extracting and separating the representation of the ground panoramic image and the remote sensing image under any environmental variable conditions, the matching accuracy of the ground panoramic image and the remote sensing image is improved, thereby achieving consistency of the remote sensing image positioning results under any environmental variable conditions.

[0027] A second aspect of this application provides a geographic view positioning device, comprising:

[0028] The acquisition module is used to acquire panoramic ground images of the area to be located and to determine a remote sensing image database containing the area to be located.

[0029] The extraction module is used to extract features from the ground panoramic image to obtain first basic feature data, and to extract second basic feature data from all remote sensing images in the remote sensing image database.

[0030] The construction module is used to combine the first basic feature data of the ground panoramic image with preset environmental parameters to perform fitting and prediction, and construct the ground image environmental variable coefficient model corresponding to the area to be located.

[0031] The acquisition module is used to determine the environmental coefficients of the area to be located based on the ground image environmental variable coefficient model, and to perform spatial reconstruction based on environmental weights according to the environmental coefficients to obtain static and dynamic ground panoramic images that meet preset consistency feature conditions; and

[0032] The matching module is used to extract features from the static ground panoramic image to obtain third basic feature data, and to perform nearest neighbor matching based on Euclidean distance based on the third basic feature data and the second basic feature data to obtain the optimal remote sensing image to be located.

[0033] According to one embodiment of this application, the acquisition module is specifically used for:

[0034] Based on the area to be located, a preset environmental parameter model is constructed;

[0035] Based on multiple acquisition environments in the preset environmental parameter model, multiple panoramic ground images of the area to be located are acquired, wherein the acquisition environment includes multiple illumination parameters and / or multiple meteorological parameters.

[0036] According to one embodiment of this application, the construction module is specifically used for:

[0037] Based on the first basic feature data of multiple ground panoramic images of the area to be located, the average feature data of the first basic feature data of the multiple ground panoramic images is obtained.

[0038] Based on the difference between the first basic feature data and the average feature data, the dynamic features of each ground panoramic image based on environmental changes are obtained, and a ground image environmental variable coefficient model is constructed according to the response of the dynamic features in the ground panoramic image.

[0039] According to one embodiment of this application, the acquisition module is specifically used for:

[0040] Based on multiple panoramic ground images of the area to be located, and combined with the preset environmental parameters, feature extraction is performed on each panoramic ground image of the area to be located to obtain the fourth basic feature data.

[0041] Based on the environmental coefficient of the area to be located, the fourth basic feature data is subjected to environmental transformation to recover static and dynamic ground panoramic images that meet the preset consistency feature conditions.

[0042] According to one embodiment of this application, the matching module is specifically used for:

[0043] The third basic feature data of the static ground panoramic image is extracted and matched with the second basic feature data based on Euclidean distance to obtain the best positioning result;

[0044] Based on the static ground panoramic image that meets the preset consistency feature conditions, the static ground panoramic image and the second basic feature data are matched with nearest neighbors based on Euclidean distance to obtain the first distance data between each static ground panoramic image and all remote sensing images in the remote sensing image database.

[0045] Based on the dynamic ground panoramic image that meets the preset consistency feature conditions, the dynamic ground panoramic image and the second basic feature data are matched with the nearest neighbor based on Euclidean distance to obtain the second distance data between each dynamic ground panoramic image and all remote sensing images in the remote sensing image database.

[0046] By combining the first distance data and the second distance data, the comprehensive distance data between each dynamic ground panoramic image and all remote sensing images in the remote sensing image database is obtained, and the image with the smallest comprehensive distance is selected as the best remote sensing image to be located.

[0047] According to the geographic view positioning device of this application embodiment, a ground panoramic image of the area to be positioned is acquired and its first basic feature data is extracted. A remote sensing image database containing the area to be positioned is determined, and the second basic feature data of all remote sensing images is extracted. The first basic feature data and preset environmental parameters are combined for fitting and prediction to construct a ground image environmental variable coefficient model corresponding to the area to be positioned. Then, the environmental coefficient of the area to be positioned is determined for spatial restoration, resulting in a static ground panoramic image and a dynamic ground panoramic image that meet preset consistency feature conditions. The static ground panoramic image is used to extract features to obtain a third basic feature data. At the same time, the second basic feature data is combined with Euclidean distance-based nearest neighbor matching to obtain the optimal remote sensing image to be positioned. Thus, the problem of image representation deviation caused by different image acquisition times or regions, resulting in matching errors and affecting the use of the positioning system in real-world environments, is solved. By extracting and separating the representation of ground panoramic images and remote sensing images under arbitrary environmental variable conditions, the matching accuracy of ground panoramic images and remote sensing images is improved, thereby achieving consistency of remote sensing image positioning results under arbitrary environmental variable conditions.

[0048] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the geographic view positioning method as described in the above embodiments.

[0049] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the geographic view positioning method as described in the above embodiments.

[0050] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0051] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0052] Figure 1 This is a flowchart of a geographic view positioning method according to an embodiment of this application;

[0053] Figure 2 This is a schematic diagram of the overall process of a geographic view positioning method and system based on a different illumination and meteorological consistency according to an embodiment of this application;

[0054] Figure 3 This is a block diagram of a geographic view positioning device according to an embodiment of this application;

[0055] Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0056] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0057] The following describes a geographic view positioning method, apparatus, electronic device, and storage medium according to embodiments of this application with reference to the accompanying drawings. Addressing the issue mentioned in the background art where image representation deviates due to different image acquisition times or regions, leading to errors in matching results and affecting the use of the positioning system in real-world environments, this application provides a geographic view positioning method. In this method, a ground panoramic image of the area to be positioned is acquired and its first basic feature data is extracted. A remote sensing image database containing the area to be positioned is determined, and second basic feature data of all remote sensing images is extracted. The first basic feature data and preset environmental parameters are combined for fitting and prediction to construct a ground image environmental variable coefficient model corresponding to the area to be positioned. Then, the environmental coefficient of the area to be positioned is determined for spatial restoration, resulting in a static ground panoramic image and a dynamic ground panoramic image that meet preset consistency feature conditions. Features are extracted from the static ground panoramic image to obtain third basic feature data. Simultaneously, nearest neighbor matching based on Euclidean distance is performed using the second basic feature data to obtain the optimal remote sensing image to be positioned. This solves the problem that deviations in image representation due to different image acquisition times or regions lead to errors in matching results, thus affecting the use of the positioning system in real-world environments. By extracting and separating ground panoramic images and remote sensing images under arbitrary environmental variable conditions, the matching accuracy of ground panoramic images and remote sensing images is improved, thereby achieving consistency in remote sensing image positioning results under arbitrary environmental variable conditions.

[0058] Specifically, Figure 1 This is a flowchart illustrating a geographic view positioning method provided in an embodiment of this application.

[0059] like Figure 1 As shown, the geographic view positioning method includes the following steps:

[0060] In step S101, a panoramic ground image of the area to be located is acquired, and a remote sensing image database containing the area to be located is determined.

[0061] According to one embodiment of this application, acquiring ground panoramic images of a region to be located includes: constructing a preset environmental parameter model based on the region to be located; and acquiring multiple ground panoramic images of the region to be located based on multiple acquisition environments in the preset environmental parameter model, wherein the acquisition environment includes multiple illumination parameters and / or multiple meteorological parameters.

[0062] The preset environmental parameter model can be an environmental parameter model constructed by a person skilled in the art based on a variety of environmental parameters selected according to actual image positioning requirements, and is not specifically limited here.

[0063] Specifically, in the geographic view positioning process of this application embodiment, it is first necessary to select the area to be positioned based on the actual positioning requirements in order to fix the specific location of the acquired image; secondly, a preset environmental parameter model is constructed based on the area to be positioned, and multiple ground panoramic images of the area to be positioned are acquired based on multiple acquisition environments in the preset environmental parameter model, wherein the acquisition environment includes multiple illumination parameters and / or multiple meteorological parameters.

[0064] Furthermore, in this embodiment, illumination and weather conditions are used as environmental parameters for image localization. Therefore, this embodiment can acquire panoramic ground images of the area to be localized under multiple illumination parameters and / or multiple weather parameters, and the acquired panoramic ground images are denoted as I1, I2, ..., I... n Where n is the number of environmental parameters, and the light intensity under each sampling environment is recorded using an illuminometer, denoted as L1, L2, ..., L n Simultaneously, a remote sensing image database is identified and made public within the area to be located, so that all remote sensing images in the database can cover the ground points to be located in the area as completely and densely as possible.

[0065] In step S102, feature extraction is performed on the ground panoramic image to obtain the first basic feature data, and all remote sensing images in the remote sensing image database are extracted to obtain the second basic feature data.

[0066] Specifically, in this embodiment, based on multiple ground panoramic images of the area to be located, features are extracted from each ground panoramic image under multiple illumination parameters and / or multiple meteorological parameters to obtain a feature vector for each ground panoramic image. That is, feature extraction is performed on multiple ground panoramic images under multiple illumination parameters and / or multiple meteorological parameters to obtain first basic feature data of size 1×C for each ground panoramic image, denoted as F. G1 ,F G2 ,...,F GnWhere C is the dimension of the feature vector, which can be adjusted according to specific circumstances in actual use, and n is the number of ground panoramic images; simultaneously, feature extraction is performed on all remote sensing images in the remote sensing image database under multiple illumination parameters and / or multiple meteorological parameters to obtain the second basic feature data of size 1×C for each remote sensing image, denoted as F. S1 ,F S2 ,...,F Sm , where m is the total number of remote sensing images in the remote sensing database.

[0067] It should be noted that the feature extraction of ground panoramic images and remote sensing images in this application embodiment can adopt the feature extraction methods in related technologies, such as using a feature aggregation strategy of multiple spatial embeddings to extract features from ground panoramic images and remote sensing images, or other related feature extraction methods, which are not specifically limited here.

[0068] In step S103, the first basic feature data of the ground panoramic image and the preset environmental parameters are combined for fitting and prediction to construct the ground image environmental variable coefficient model corresponding to the area to be located.

[0069] According to one embodiment of this application, a ground image environmental variable coefficient model is constructed by fitting and predicting the first basic feature data of the ground panoramic image and preset environmental parameters, including: obtaining the average feature data of the first basic feature data of the multiple ground panoramic images based on the first basic feature data of the area to be located; obtaining the dynamic features of each ground panoramic image based on environmental changes based on the difference between the first basic feature data and the average feature data; and constructing the ground image environmental variable coefficient model according to the response of the dynamic features in the ground panoramic image.

[0070] The preset environmental parameters can be multiple light parameters, multiple meteorological parameters, or other environmental parameters, and no specific limitations are made here.

[0071] Specifically, based on the first basic feature data of multiple ground panoramic images of the area to be located obtained under multiple illumination parameters and / or multiple meteorological parameters according to the embodiments of this application, firstly, the average value of the first basic feature data is taken to obtain the average feature data of the first basic feature data of the multiple ground panoramic images, denoted as F. G =(F G1 +F G2 +...+F Gn ) / n, and based on the first basic feature data F of multiple ground panoramic images acquired under multiple illumination parameters and / or multiple meteorological parameters. G1 ,F G2 ,...,F Gn and average feature data FG The difference is used to obtain n dynamic features F to describe environmental changes. Di =F Gi -F G .

[0072] Secondly, this application's embodiments model the responses of two environmental changes—illuminance and meteorology—to ground panoramic image features, i.e., constructing a ground image environmental variable coefficient model, representing illumination changes as a proportionality coefficient k, and meteorological changes as an offset parameter b, based on the recorded illumination intensities L1, L2, ..., L... n The system of equations is optimized to solve for the illuminance coefficient k and the meteorological parameter b, where k is a real coefficient and b is a real vector of size 1×C. The specific formula of the system of equations is as follows:

[0073]

[0074] In step S104, the environmental coefficients of the area to be located are determined based on the ground image environmental variable coefficient model, and spatial restoration based on environmental weights is performed according to the environmental coefficients to obtain static ground panoramic images and dynamic ground panoramic images that meet the preset consistency feature conditions.

[0075] According to one embodiment of this application, the environmental coefficients of the area to be located are determined based on a ground image environmental variable coefficient model, and spatial restoration based on environmental weights is performed on the fourth basic feature data according to the environmental coefficients to obtain static and dynamic ground panoramic images that meet preset consistency feature conditions. The process includes: extracting features from each ground panoramic image of the area to be located based on multiple ground panoramic images of the area to be located, combined with preset environmental parameters, to obtain fourth basic feature data; and performing environmental transformation on the fourth basic feature data according to the environmental coefficients of the area to be located to restore static and dynamic ground panoramic images that meet preset consistency feature conditions.

[0076] The preset consistency feature conditions can be relevant conditions set by those skilled in the art based on actual image positioning needs, and are not specifically limited here.

[0077] Specifically, in this embodiment, multiple ground panoramic images I of the area to be located are selected, their illumination intensity L is recorded, and feature extraction is performed on each ground panoramic image of the area to be located to obtain fourth basic feature data with a size of 1×C for each ground panoramic image; based on the illumination coefficient k and meteorological parameter b obtained above, environmental transformation is performed on the fourth basic feature data to recover the consistent static features corresponding to the static ground panoramic image that meets the preset consistency feature conditions. The consistent dynamic features corresponding to dynamic ground panoramic images are

[0078] In step S105, feature extraction is performed on the static ground panoramic image to obtain the third basic feature data, and nearest neighbor matching based on Euclidean distance is performed based on the third basic feature data and the second basic feature data to obtain the optimal remote sensing image to be located.

[0079] According to one embodiment of this application, feature extraction is performed on a static ground panoramic image to obtain third basic feature data, and nearest neighbor matching based on Euclidean distance is performed on the third basic feature data and the second basic feature data to obtain the optimal remote sensing image to be located. This includes: extracting the third basic feature data of the static ground panoramic image and performing nearest neighbor matching based on Euclidean distance with the second basic feature data to obtain the optimal positioning result; based on a static ground panoramic image that meets a preset consistency feature condition, performing nearest neighbor matching based on Euclidean distance between the static ground panoramic image and the second basic feature data to obtain first distance data between each static ground panoramic image and all remote sensing images in the remote sensing image database; based on a dynamic ground panoramic image that meets a preset consistency feature condition, performing nearest neighbor matching based on Euclidean distance between the dynamic ground panoramic image and the second basic feature data to obtain second distance data between each dynamic ground panoramic image and all remote sensing images in the remote sensing image database; combining the first distance data and the second distance data to obtain comprehensive distance data between each dynamic ground panoramic image and all remote sensing images in the remote sensing image database, and selecting the image with the smallest comprehensive distance as the optimal remote sensing image to be located.

[0080] Specifically, based on the consistent static features obtained above The consistent dynamic characteristics are The embodiments of this application respectively address the consistent static features. The consistent dynamic characteristics are The second basic feature data of all remote sensing images in the remote sensing image database are matched with nearest neighbors based on Euclidean distance to obtain a comprehensive distance D. The remote sensing image with the smallest comprehensive distance to the ground panoramic image is selected as the best positioning result, and thus this remote sensing image is regarded as the best remote sensing image to be located. The specific calculation method of the comprehensive distance is as follows:

[0081]

[0082] Here, Dist() calculates the Euclidean distance, and sigmoid(x) represents the Logistic function.

[0083] In summary, to facilitate a more intuitive understanding of the embodiments of this application by those skilled in the art, the following will be based on... Figure 2 To provide further details, the specific steps are as follows:

[0084] In step S201, a single panoramic image of the ground is acquired under any weather and lighting conditions, and a candidate remote sensing image database is selected.

[0085] In step S202, basic feature blocks are extracted from the ground panoramic image and the remote sensing image.

[0086] In step S203, the ground panoramic image is combined to perform fitting and prediction of external environmental parameters such as illumination and meteorology, and an environmental variable coefficient model of the ground image is constructed.

[0087] In step S204, spatial restoration based on environmental weights is performed on the feature blocks of the panoramic image to be located based on environmental coefficients to obtain a unified static and dynamic representation of the ground panoramic image.

[0088] In step S205, features are extracted from the statically represented ground panoramic image, and nearest neighbor matching based on Euclidean distance is performed on the remote sensing image features in the remote sensing database to obtain the best-matching remote sensing image as the localization result.

[0089] According to the geographic view positioning method of this application, a ground panoramic image of the area to be positioned is acquired and its first basic feature data is extracted. A remote sensing image database containing the area to be positioned is determined, and the second basic feature data of all remote sensing images is extracted. The first basic feature data and preset environmental parameters are combined for fitting and prediction to construct a ground image environmental variable coefficient model corresponding to the area to be positioned. Then, the environmental coefficient of the area to be positioned is determined for spatial restoration, resulting in a static ground panoramic image and a dynamic ground panoramic image that meet preset consistency feature conditions. The static ground panoramic image is used to extract features to obtain the third basic feature data. At the same time, the second basic feature data is combined with Euclidean distance-based nearest neighbor matching to obtain the optimal remote sensing image to be positioned. This solves the problem that when the image acquisition time or region is different, the image representation will be biased, resulting in errors in the matching results and affecting the use of the positioning system in real environment. By extracting and separating the representation of the ground panoramic image and the remote sensing image under any environmental variable conditions, the matching accuracy of the ground panoramic image and the remote sensing image is improved, thereby achieving consistency of the remote sensing image positioning results under any environmental variable conditions.

[0090] Next, the geographic view positioning device proposed according to the embodiments of this application is described with reference to the accompanying drawings.

[0091] Figure 3 This is a block diagram of a geographic view positioning device according to an embodiment of this application.

[0092] like Figure 3 As shown, the geographic view-based positioning device 10 includes: a data acquisition module 100, an extraction module 200, a construction module 300, an acquisition module 400, and a matching module 500.

[0093] The acquisition module 100 is used to acquire panoramic ground images of the area to be located and to determine a remote sensing image database containing the area to be located.

[0094] The extraction module 200 is used to extract features from the ground panoramic image to obtain the first basic feature data, and to extract the second basic feature data from all remote sensing images in the remote sensing image database.

[0095] Module 300 is used to combine the first basic feature data of the ground panoramic image with preset environmental parameters to perform fitting and prediction, and to construct the ground image environmental variable coefficient model corresponding to the area to be located.

[0096] The acquisition module 400 is used to determine the environmental coefficients of the area to be located based on the environmental variable coefficient model of the ground image, and to perform spatial reconstruction based on environmental weights according to the environmental coefficients to obtain static and dynamic ground panoramic images that meet preset consistency feature conditions; and

[0097] The matching module 500 is used to extract features from the static ground panoramic image to obtain the third basic feature data, and to perform nearest neighbor matching based on Euclidean distance based on the third basic feature data and the second basic feature data to obtain the best remote sensing image to be located.

[0098] According to one embodiment of this application, the acquisition module 100 is specifically used for:

[0099] Based on the area to be located, construct a model with preset environmental parameters;

[0100] Based on multiple acquisition environments in the preset environmental parameter model, multiple panoramic ground images of the area to be located are acquired. The acquisition environment includes multiple illumination parameters and / or multiple meteorological parameters.

[0101] According to one embodiment of this application, the construction module 300 is specifically used for:

[0102] Based on the first basic feature data of multiple ground panoramic images of the area to be located, the average feature data of the first basic feature data of multiple ground panoramic images is obtained.

[0103] Based on the difference between the first basic feature data and the average feature data, the dynamic features of each ground panoramic image based on environmental changes are obtained, and a ground image environmental variable coefficient model is constructed according to the response of the dynamic features in the ground panoramic image.

[0104] According to one embodiment of this application, the acquisition module 400 is specifically used for:

[0105] Based on multiple panoramic ground images of the area to be located, combined with preset environmental parameters, features are extracted from each panoramic ground image of the area to be located to obtain the fourth basic feature data.

[0106] Based on the environmental coefficient of the area to be located, the fourth basic feature data is transformed to recover static and dynamic ground panoramic images that meet the preset consistency feature conditions.

[0107] According to one embodiment of this application, the matching module 500 is specifically used for:

[0108] The third basic feature data of the static ground panoramic image is extracted, and the second basic feature data is matched with the nearest neighbor based on Euclidean distance to obtain the best localization result;

[0109] Based on static ground panoramic images that meet the preset consistency feature conditions, the static ground panoramic images and the second basic feature data are matched with nearest neighbors based on Euclidean distance to obtain the first distance data between each static ground panoramic image and all remote sensing images in the remote sensing image database.

[0110] Based on dynamic ground panoramic images that meet preset consistency feature conditions, the dynamic ground panoramic images and the second basic feature data are matched with nearest neighbors based on Euclidean distance to obtain the second distance data between each dynamic ground panoramic image and all remote sensing images in the remote sensing image database.

[0111] By combining the first distance data and the second distance data, the comprehensive distance data of each dynamic ground panoramic image and all remote sensing images in the remote sensing image database is obtained, and the image with the smallest comprehensive distance is selected as the best remote sensing image to be located.

[0112] According to the geographic view positioning device of this application embodiment, a ground panoramic image of the area to be positioned is acquired and its first basic feature data is extracted. A remote sensing image database containing the area to be positioned is determined, and the second basic feature data of all remote sensing images is extracted. The first basic feature data and preset environmental parameters are combined for fitting and prediction to construct a ground image environmental variable coefficient model corresponding to the area to be positioned. Then, the environmental coefficient of the area to be positioned is determined for spatial restoration, resulting in a static ground panoramic image and a dynamic ground panoramic image that meet preset consistency feature conditions. The static ground panoramic image is used to extract features to obtain a third basic feature data. At the same time, the second basic feature data is combined with Euclidean distance-based nearest neighbor matching to obtain the optimal remote sensing image to be positioned. Thus, the problem of image representation deviation caused by different image acquisition times or regions, resulting in matching errors and affecting the use of the positioning system in real-world environments, is solved. By extracting and separating the representation of ground panoramic images and remote sensing images under arbitrary environmental variable conditions, the matching accuracy of ground panoramic images and remote sensing images is improved, thereby achieving consistency of remote sensing image positioning results under arbitrary environmental variable conditions.

[0113] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include:

[0114] The memory 401, the processor 402, and the computer program stored on the memory 401 and capable of running on the processor 402.

[0115] When the processor 402 executes the program, it implements the geographic view positioning method provided in the above embodiments.

[0116] Furthermore, electronic devices also include:

[0117] Communication interface 403 is used for communication between memory 401 and processor 402.

[0118] The memory 401 is used to store computer programs that can run on the processor 402.

[0119] The memory 401 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0120] If the memory 401, processor 402, and communication interface 403 are implemented independently, then the communication interface 403, memory 401, and processor 402 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0121] Optionally, in a specific implementation, if the memory 401, processor 402, and communication interface 403 are integrated on a single chip, then the memory 401, processor 402, and communication interface 403 can communicate with each other through an internal interface.

[0122] Processor 402 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0123] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described geographic view positioning method.

[0124] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0125] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0126] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A geographic view positioning method, characterized in that, Includes the following steps: Collect panoramic ground images of the area to be located and determine a remote sensing image database containing the area to be located; First basic feature data is obtained by extracting features from the ground panoramic image, and second basic feature data is obtained by extracting features from all remote sensing images in the remote sensing image database. By combining the first basic feature data of the ground panoramic image with preset environmental parameters, a ground image environmental variable coefficient model corresponding to the area to be located is constructed through fitting and prediction. The environmental coefficients of the area to be located are determined based on the environmental variable coefficient model of the ground image, and spatial restoration based on environmental weights is performed according to the environmental coefficients to obtain static and dynamic ground panoramic images that meet the preset consistency feature conditions. as well as The static ground panoramic image is subjected to feature extraction to obtain third basic feature data, and Euclidean distance-based nearest neighbor matching is performed based on the third basic feature data and the second basic feature data to obtain the optimal remote sensing image to be located.

2. The method according to claim 1, characterized in that, The acquisition of panoramic ground images of the area to be located includes: Based on the area to be located, a preset environmental parameter model is constructed; Based on multiple acquisition environments in the preset environmental parameter model, multiple panoramic ground images of the area to be located are acquired, wherein the acquisition environment includes multiple illumination parameters and / or multiple meteorological parameters.

3. The method according to claim 1, characterized in that, The method of fitting and predicting the first basic feature data of the combined ground panoramic image and preset environmental parameters to construct a ground image environmental variable coefficient model corresponding to the area to be located includes: Based on the first basic feature data of multiple ground panoramic images of the area to be located, the average feature data of the first basic feature data of the multiple ground panoramic images is obtained. Based on the difference between the first basic feature data and the average feature data, the dynamic features of each ground panoramic image based on environmental changes are obtained, and a ground image environmental variable coefficient model is constructed according to the response of the dynamic features in the ground panoramic image.

4. The method according to claim 1, characterized in that, The process of determining the environmental coefficients of the area to be located based on the ground image environmental variable coefficient model, and performing spatial reconstruction of the fourth basic feature data based on environmental weights according to the environmental coefficients to obtain static and dynamic ground panoramic images that satisfy the preset consistency feature conditions, includes: Based on multiple panoramic ground images of the area to be located, and combined with the preset environmental parameters, feature extraction is performed on each panoramic ground image of the area to be located to obtain the fourth basic feature data. Based on the environmental coefficient of the area to be located, the fourth basic feature data is subjected to environmental transformation to recover static and dynamic ground panoramic images that meet the preset consistency feature conditions.

5. The method according to claim 1, characterized in that, The process of extracting features from the static panoramic ground image to obtain third basic feature data, and performing nearest neighbor matching based on Euclidean distance based on the third basic feature data and the second basic feature data to obtain the optimal remote sensing image to be located, includes: The third basic feature data of the static ground panoramic image is extracted and matched with the second basic feature data based on Euclidean distance to obtain the best positioning result; Based on the static ground panoramic image that meets the preset consistency feature conditions, the static ground panoramic image and the second basic feature data are matched with nearest neighbors based on Euclidean distance to obtain the first distance data between each static ground panoramic image and all remote sensing images in the remote sensing image database. Based on the dynamic ground panoramic image that meets the preset consistency feature conditions, the dynamic ground panoramic image and the second basic feature data are matched with the nearest neighbor based on Euclidean distance to obtain the second distance data between each dynamic ground panoramic image and all remote sensing images in the remote sensing image database. By combining the first distance data and the second distance data, the comprehensive distance data between each dynamic ground panoramic image and all remote sensing images in the remote sensing image database is obtained, and the image with the smallest comprehensive distance is selected as the best remote sensing image to be located.

6. A geographic view positioning device, characterized in that, include: The acquisition module is used to acquire panoramic ground images of the area to be located and to determine a remote sensing image database containing the area to be located. The extraction module is used to extract features from the ground panoramic image to obtain first basic feature data, and to extract second basic feature data from all remote sensing images in the remote sensing image database. The construction module is used to combine the first basic feature data of the ground panoramic image with preset environmental parameters to perform fitting and prediction, and construct the ground image environmental variable coefficient model corresponding to the area to be located. The acquisition module is used to determine the environmental coefficient of the area to be located based on the environmental variable coefficient model of the ground image, and to perform spatial restoration based on environmental weights according to the environmental coefficient to obtain static ground panoramic image and dynamic ground panoramic image that meet the preset consistency feature conditions. as well as The matching module is used to extract features from the static ground panoramic image to obtain third basic feature data, and to perform nearest neighbor matching based on Euclidean distance based on the third basic feature data and the second basic feature data to obtain the optimal remote sensing image to be located.

7. The apparatus according to claim 6, characterized in that, The acquisition module is specifically used for: Based on the area to be located, a preset environmental parameter model is constructed; Based on multiple acquisition environments in the preset environmental parameter model, multiple panoramic ground images of the area to be located are acquired, wherein the acquisition environment includes multiple illumination parameters and / or multiple meteorological parameters.

8. The apparatus according to claim 6, characterized in that, The building module is specifically used for: Based on the first basic feature data of multiple ground panoramic images of the area to be located, the average feature data of the first basic feature data of the multiple ground panoramic images is obtained. Based on the difference between the first basic feature data and the average feature data, the dynamic features of each ground panoramic image based on environmental changes are obtained, and a ground image environmental variable coefficient model is constructed according to the response of the dynamic features in the ground panoramic image.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the geographic view positioning method as described in any one of claims 1-5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the geographic view positioning method as described in any one of claims 1-5.

Images