All-day GNSS NLOS recognition method and device based on fisheye camera

Abstract

In order to solve the problem of receiving non-line-of-sight (NLOS) signals broadcast by satellites blocked by tall buildings or trees during GNSS positioning, a method of obtaining a projection of a satellite sky map by shooting toward the sky with a fisheye camera is used to realize GNSS NLOS recognition. When performing data processing position calculation, its weight is reduced, which can effectively improve the GNSS positioning accuracy. The present invention discloses a GNSS NLOS recognition method and device based on a fisheye camera all day long. The method of shooting toward the sky with a fisheye camera is used to obtain a sky map of the surrounding environment. The GNSS NLOS recognition is realized by means of image segmentation and satellite projection, which satisfies both daytime and nighttime environments, thereby improving the GNSS positioning accuracy according to certain data processing means.

Description

Technical Field

[0001] The present invention belongs to the field of surveying and mapping science and technology, and relates to a method for processing abnormal values ​​in global navigation satellite system (GNSS) data, and in particular to a method and device for identifying GNSS NLOS all day long based on a fisheye camera. Background Art

[0002] At present, many cities in my country are piloting autonomous vehicles. In autonomous driving, high-precision navigation plays an important role in precise lateral / longitudinal positioning of vehicles, intelligent speed control, path planning and behavioral decision-making. Studies have shown that for highway environments, the lateral error required for autonomous driving should not exceed 0.57m (0.20m, 95%), and the longitudinal error should be less than 1.40m (0.48m, 95%); on urban streets, due to narrower roads and more complex geometric shapes, the lateral and longitudinal errors should be less than 0.29m (0.10m, 95%). It can be seen that in complex urban environments, high-precision navigation has become a key technology for the safe driving of autonomous vehicles.

[0003] However, due to the defects of the commonly used navigation technologies, no single technology can meet the accuracy and update frequency requirements of autonomous driving, such as non-line-of-sight (NLOS) signals of satellite positioning technology. NLOS signals will cause a significant decrease in GNSS positioning accuracy, especially in complex urban scenes. Therefore, processing GNSS NLOS signals is a key link in the study of high-precision satellite positioning technology and the realization of high-precision navigation in autonomous driving. Researchers often use fisheye cameras to shoot at the sky to obtain a sky map of the surrounding environment, and use image segmentation and satellite projection methods to realize GNSS NLOS recognition. At present, the existing methods have technical problems such as low image segmentation accuracy and can only be used in daytime with good lighting conditions. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a 24-hour GNSS NLOS recognition method based on a fisheye camera.

[0005] The technical solution adopted by the present invention is as follows:

[0006] The first aspect discloses a 24-hour GNSS NLOS recognition method based on a fisheye camera, the method comprising:

[0007] S1: Use a fisheye lens pointing to the sky to obtain a sky image of the current positioning point, and determine whether the current environment is a night environment. If not, preprocess the sky image of the current positioning point to segment the sky area and the non-sky area. If yes, use cycleGAN to perform style conversion on the sky image of the current positioning point and convert it into a daytime style sky image. Then preprocess the daytime style sky image to segment the sky area and the non-sky area.

[0008] S2: According to the IMU initial attitude and the collected IMU observation data, the inertial navigation mechanical arrangement is performed, and the GNSS data and INS data are fused using the Kalman filtering method to obtain the integrated navigation positioning result, wherein the integrated navigation positioning result includes the heading angle of the current carrier;

[0009] S3: Calculate the distance from the satellite projection position to the center of the sky image based on the current GNSS / INS integrated navigation positioning result, satellite ephemeris, the fisheye lens stereoscopic projection model and focal length, and calculate the position of the satellite projection in the sky image based on the current carrier heading angle and the position of the sky image center;

[0010] S4: Determine whether it is an NLOS signal based on the segmentation result of the sky image of the current positioning point and the position of the satellite projection in the sky image.

[0011] In one implementation, preprocessing the sky image of the current positioning point includes:

[0012] First, the read RGB image is transformed into a grayscale image and Gaussian filtering is performed;

[0013] The image processed by Gaussian filtering is subjected to edge detection using the Canny operator to obtain the edge information of the image;

[0014] The region growing method is used to obtain the candidate sky region seed points for the image after Gaussian filtering, which specifically includes: firstly obtaining the sky region seed points, clustering them using the K-Means++ clustering algorithm, obtaining the clustering results represented by a binary image, performing a connected domain analysis on the binary image, and obtaining the connected domain centroid as the candidate sky region seed point;

[0015] A brightness coefficient is obtained according to the mean value of the deviation between the whole image and the set grayscale reference value of the sky, and the mean value of the deviation between the evaluation area and the set grayscale reference value of the sky. According to the relationship between the brightness coefficient and the brightness threshold of the sky area, it is judged whether the cluster center is located in the sky area, so as to select the sky seed point from the candidate of the sky area seed point, wherein the evaluation area is a circle with the cluster center as the center and the preset distance as the radius;

[0016] The selected sky points are used as seed points, and the edge mask detected in edge detection is used to flood fill the pixels in the sky area until the edge is reached, and the sky image segmentation result P is obtained. skymask , P skymask It is a sky area or a non-sky area.

[0017] In one implementation, connected domain analysis is performed on the binary image to obtain the connected domain centroid as a candidate for the sky region seed point, including:

[0018] Perform connected domain statistics on the binary image, mark the centroid of each connected domain, and calculate the area of ​​each connected domain;

[0019] If the area of ​​the connected domain is greater than the set connected domain area threshold, the centroid of the connected domain is used as a candidate for the sky region seed point.

[0020] In one implementation, the brightness coefficient is obtained according to the average value of the deviation between the entire image and the set grayscale reference value of the sky and the average value of the deviation between the evaluation area and the set grayscale reference value of the sky, including:

[0021] Calculate the mean deviation between the entire image and the grayscale reference value of the sky. The specific calculation method is:

[0022]

[0023] Where D represents the mean deviation between the entire image and the grayscale reference value of the sky, V k Represents the gray value of the kth point in the entire image, V sky Indicates the grayscale reference value of the sky, and S represents the entire image;

[0024] Calculate the mean deviation between the evaluation area and the grayscale reference value of the sky. The specific calculation method is:

[0025]

[0026] Where M represents the mean deviation between the entire image and the grayscale reference value of the sky, and h i is the number of pixels with gray value i;

[0027] The brightness coefficient is obtained according to the formula:

[0028]

[0029] Where L is the brightness coefficient.

[0030] In one implementation, judging whether the cluster center is located in the sky area according to the relationship between the brightness coefficient and the brightness threshold of the sky area, thereby selecting the sky seed point from the candidate sky area seed points, includes:

[0031] Set the brightness threshold L of the sky area threshold1 and L threshold2 , if L threshold1 <L<L threshold2 , it means that the cluster center is located in the sky area, so the sky seed point is selected from the candidate sky area seed points.

[0032] In one embodiment, step S3 includes:

[0033] According to the current GNSS / INS integrated navigation positioning results and satellite ephemeris, the satellite altitude and azimuth are calculated. According to the stereoscopic projection model and focal length of the fisheye lens, the distance from the satellite projection position to the center of the sky image is calculated:

[0034]

[0035] Among them, f c is the focal length of the fisheye lens, ε sat is the satellite altitude angle, d pix is the distance from the satellite projection position to the center of the sky image. The coordinates of the center of the sky image are (x c ,y c );

[0036] According to the solved heading angle yaw, calculate the position of the satellite projection on the sky map:

[0037] x sat =x c +d pix ·cos(yaw+α sat )

[0038] y sat =y c -d pix ·sin(yaw+α sat )

[0039] Among them, yaw is the heading angle of the current carrier, α sat is the satellite azimuth, (x sat ,y sat ) are the position coordinates of the satellite projection on the sky map.

[0040] In one embodiment, the sky image segmentation result is P skymask and the satellite projection position coordinates on the image are (x sat ,y sat), step S4 comprises:

[0041] If P skymask (x sat ,y sat )=0, image matrix P skymask If the value of the position of the satellite projection on the image is 0, the satellite signal is an NLOS signal; otherwise, the signal is a LOS signal.

[0042] Based on the same inventive concept, the second aspect of the present invention provides a 24-hour GNSS NLOS recognition device based on a fisheye camera, the device comprising:

[0043] The image segmentation module is used to obtain the sky image of the current positioning point using a fisheye lens pointing to the sky, and determine whether the current environment is a night environment. If not, the sky image of the current positioning point is preprocessed to segment the sky area and the non-sky area. If yes, cycleGAN is used to perform style conversion on the sky image of the current positioning point and convert it into a daytime style sky image. Then, the daytime style sky image is preprocessed to segment the sky area and the non-sky area.

[0044] The integrated navigation module is used to perform inertial navigation mechanical arrangement according to the IMU initial attitude and the collected IMU observation data, and use the Kalman filtering method to fuse the GNSS data and INS data to obtain the integrated navigation positioning result, wherein the integrated navigation positioning result includes the heading angle of the current carrier;

[0045] The satellite position calculation module is used to calculate the distance from the satellite projection position to the center of the sky image based on the current GNSS / INS integrated navigation positioning result, satellite ephemeris, the fisheye lens stereoscopic projection model and focal length, and calculate the position of the satellite projection in the sky image based on the current carrier heading angle and the position of the sky image center;

[0046] The NLOS signal judgment module is used to judge whether it is an NLOS signal according to the segmentation result of the sky image of the current positioning point and the position of the satellite projection in the sky image.

[0047] Based on the same inventive concept, the third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the all-day GNSS NLOS recognition method based on a fisheye camera as described in the first aspect.

[0048] Based on the same inventive concept, the fourth aspect of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the all-day GNSS NLOS identification method based on a fisheye camera as described in the first aspect is implemented.

[0049] Compared with the prior art, the advantages and beneficial technical effects of the present invention are as follows:

[0050] The present invention discloses a GNSS NLOS recognition method based on a fisheye camera at all times of the day. The method adopts a fisheye camera to shoot toward the sky to obtain a sky map of the surrounding environment, and realizes GNSS NLOS recognition by image segmentation and satellite projection, while meeting the daytime and nighttime environments, thereby improving the GNSS positioning accuracy according to certain data processing methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 The present invention provides a flow chart of a method for all-day GNSS NLOS recognition based on a fisheye camera according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to solve the problem of receiving non-line-of-sight (NLOS) signals broadcast by satellites blocked by tall buildings or trees during GNSS positioning, a method of obtaining a projection of a satellite sky map by shooting a fisheye camera toward the sky is used to realize GNSS NLOS recognition. When performing data processing position calculation, its weight is reduced, which can effectively improve the GNSS positioning accuracy. The present invention discloses a GNSS NLOS recognition method based on a fisheye camera all day long. The method of shooting a fisheye camera toward the sky is used to obtain a sky map of the surrounding environment, and the GNSS NLOS recognition is realized by means of image segmentation and satellite projection, which satisfies both daytime and night environments, thereby improving the GNSS positioning accuracy according to certain data processing methods.

[0054] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] Embodiment 1

[0056] The present invention discloses a GNSS NLOS recognition method based on a fisheye camera, which comprises:

[0057] S1: Use a fisheye lens pointing to the sky to obtain a sky image of the current positioning point, and determine whether the current environment is a night environment. If not, preprocess the sky image of the current positioning point to segment the sky area and the non-sky area. If yes, use cycleGAN to perform style conversion on the sky image of the current positioning point and convert it into a daytime style sky image. Then preprocess the daytime style sky image to segment the sky area and the non-sky area.

[0058] S2: According to the IMU initial attitude and the collected IMU observation data, the inertial navigation mechanical arrangement is performed, and the GNSS data and INS data are fused using the Kalman filtering method to obtain the integrated navigation positioning result, wherein the integrated navigation positioning result includes the heading angle of the current carrier;

[0059] S3: Calculate the distance from the satellite projection position to the center of the sky image based on the current GNSS / INS integrated navigation positioning result, satellite ephemeris, the fisheye lens stereoscopic projection model and focal length, and calculate the position of the satellite projection in the sky image based on the current carrier heading angle and the position of the sky image center;

[0060] S4: Determine whether it is an NLOS signal based on the segmentation result of the sky image of the current positioning point and the position of the satellite projection in the sky image.

[0061] Specifically, the present invention provides a vision-based NLOS discrimination algorithm as a whole, including steps S1 to S4. In step S1, the sky image of the current positioning point is preprocessed or the daytime style sky image is preprocessed, including Gaussian filtering, edge detection, and region growth on the image, so as to segment the sky area, obtain environmental occlusion information, and obtain the image matrix P of the sky image segmentation result skymaskS2 performs GNSS / INS combined navigation to obtain the heading angle of the current carrier. S3 further calculates the position of the satellite projection on the sky map, and S4 realizes NLOS signal discrimination based on the fisheye image based on the sky image segmentation result and the position of the satellite on the image.

[0062] The present invention utilizes the visual features of the surrounding environment and the ephemeris information of the GNSS satellites, and through multiple experiments, verifies the effectiveness of the method in identifying GNSS NLOS and improves the positioning accuracy.

[0063] The specific implementation of each step of the present invention is described in detail below.

[0064] Step 1: Use a fisheye lens pointing to the sky to obtain the sky image of the current positioning point, study the vision-based NLOS discrimination algorithm, perform Gaussian filtering, edge detection, and region growth on the image, so as to segment the sky area, obtain environmental occlusion information, and obtain the image matrix P of the sky image segmentation result. skymask .

[0065] The specific implementation plan is:

[0066] (1) After determining that the read image is not in a dark environment, the python or matlab library function can be directly called to first transform the read RGB image into a color space, convert it into a grayscale image, and perform Gaussian filtering to remove the noise that is large in the edge detection image.

[0067] (2) Then, the processed image is subjected to edge detection using the Canny operator to obtain the edge information of the image.

[0068] (3) Apply the region growing method to the image processed by (1), first obtain the seed points of the sky region, and use the K-Means++ clustering algorithm to cluster them to obtain a binary image of the sky region. Perform a connected domain analysis on the binary image and obtain the centroid of the connected domain as a candidate for the seed point of the sky region.

[0069] The specific steps of step (3) are shown in the following table:

[0070]

[0071]

[0072] (4) Screening sky seed points.

[0073] The specific implementation steps of step (4) are shown in the following table:

[0074]

[0075] (5) Finally, flood filling is performed. The sky points selected in step (4) are used as seed points. The edge mask detected in step (2) is used to flood fill the pixels in the sky area until the edge is reached. The sky image segmentation result P is obtained. skymask .

[0076] Step 2: According to the IMU initial attitude and the collected IMU observation data, the inertial navigation mechanical arrangement is performed, and the GNSS and INS data are fused using the Kalman filtering method, which is generally a loose combination method. At this time, the three-axis attitude of the carrier at each moment can be obtained, and thus the heading angle yaw of the current carrier can be obtained.

[0077] Step 3: Calculate the satellite altitude angle ε based on the current GNSS / INS integrated navigation positioning result and satellite ephemeris sat and azimuth α sat According to the stereoscopic projection model of the fisheye lens and the focal length f c , calculate the satellite projection position to the center of the sky image (x c ,y c ) pix :

[0078]

[0079] According to the heading angle yaw calculated in step 2, calculate the position of the satellite projection on the sky map (x sat ,y sat ):

[0080] x sat =x c +d pix ·cos(yaw+α sat )

[0081] y sat =y c -d pix ·sin(yaw+α sat )

[0082] Step 4: Based on the above sky image segmentation result P skymask and the position of the satellite projection on the image (x sat ,y sat ), the NLOS signal can be determined. The projected sky image is a binary image, where the sky area is the image matrix P skymask The value is 1, and the non-sky area is 0. skymask (x sat ,y sat )=0, the satellite signal is an NLOS signal; otherwise, the signal is considered to be a LOS signal.

[0083] Among them, if the current environment is a night environment, first use the fisheye camera to obtain the sky image of the current position, and then use cycleGAN to perform style conversion on it to achieve style migration of the captured image (from the night environment image to the day environment image), but the content remains unchanged (the image can still be divided into the sky part and the non-sky part), and the night image is converted from the night style to the day style, and steps 1, 2, 3 and 4 are repeated.

[0084] Embodiment 2

[0085] Based on the same inventive concept, this embodiment discloses a 24-hour GNSS NLOS recognition device based on a fisheye camera, the device comprising:

[0086] The image segmentation module is used to obtain the sky image of the current positioning point using a fisheye lens pointing to the sky, and determine whether the current environment is a night environment. If not, the sky image of the current positioning point is preprocessed to segment the sky area and the non-sky area. If yes, cycleGAN is used to perform style conversion on the sky image of the current positioning point and convert it into a daytime style sky image. Then, the daytime style sky image is preprocessed to segment the sky area and the non-sky area.

[0087] The integrated navigation module is used to perform inertial navigation mechanical arrangement according to the IMU initial attitude and the collected IMU observation data, and use the Kalman filtering method to fuse the GNSS data and INS data to obtain the integrated navigation positioning result, wherein the integrated navigation positioning result includes the heading angle of the current carrier;

[0088] The satellite position calculation module is used to calculate the distance from the satellite projection position to the center of the sky image based on the current GNSS / INS integrated navigation positioning result, satellite ephemeris, the fisheye lens stereoscopic projection model and focal length, and calculate the position of the satellite projection in the sky image based on the current carrier heading angle and the position of the sky image center;

[0089] The NLOS signal judgment module is used to judge whether it is an NLOS signal according to the segmentation result of the sky image of the current positioning point and the position of the satellite projection in the sky image.

[0090] Since the device introduced in the second embodiment of the present invention is a device used to implement the all-day GNSS NLOS identification method based on the fisheye camera in the first embodiment of the present invention, based on the method introduced in the first embodiment of the present invention, the person skilled in the art can understand the specific structure and deformation of the device, so it is not repeated here. All devices used in the method in the first embodiment of the present invention belong to the scope of protection of the present invention.

[0091] Embodiment 3

[0092] Based on the same inventive concept, the present invention further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method described in the first embodiment is implemented.

[0093] Since the computer-readable storage medium introduced in the third embodiment of the present invention is the computer-readable storage medium used to implement the all-day GNSS NLOS identification method based on the fisheye camera in the first embodiment of the present invention, based on the method introduced in the first embodiment of the present invention, the person skilled in the art can understand the specific structure and deformation of the computer-readable storage medium, so it is not repeated here. All computer-readable storage media used in the method of the first embodiment of the present invention belong to the scope of protection of the present invention.

[0094] Embodiment 4

[0095] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in Embodiment 1 when executing the program.

[0096] Since the computer device introduced in the fourth embodiment of the present invention is a computer device used to implement the all-day GNSS NLOS identification method based on the fisheye camera in the first embodiment of the present invention, based on the method introduced in the first embodiment of the present invention, the person skilled in the art can understand the specific structure and deformation of the computer device, so it is not repeated here. All computer devices used in the method of the first embodiment of the present invention belong to the scope of protection of the present invention.

[0097] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0098] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0099] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic creative concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the embodiments of the present invention without departing from the spirit and scope of the embodiments of the present invention. Thus, if these modifications and variations of the embodiments of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A 24-hour GNSS NLOS recognition method based on a fisheye camera, characterized in that: The method includes: S1: Use a fisheye lens pointing to the sky to obtain a sky image of the current positioning point, and determine whether the current environment is a night environment. If not, preprocess the sky image of the current positioning point to segment the sky area and the non-sky area. If yes, use cycleGAN to perform style conversion on the sky image of the current positioning point and convert it into a daytime style sky image. Then, preprocess the daytime style sky image to segment the sky area and the non-sky area. The preprocessing of the sky image of the current positioning point includes: performing color space conversion on the read RGB image, converting it into a grayscale image and performing Gaussian filtering. S2: According to the IMU initial attitude and the collected IMU observation data, the inertial navigation mechanical arrangement is performed, and the GNSS data and INS data are fused using the Kalman filtering method to obtain the integrated navigation positioning result, wherein the integrated navigation positioning result includes the heading angle of the current carrier; S3: Calculate the distance from the satellite projection position to the center of the sky image based on the current GNSS / INS integrated navigation positioning result, satellite ephemeris, the fisheye lens stereoscopic projection model and focal length, and calculate the position of the satellite projection in the sky image based on the current carrier heading angle and the position of the sky image center; S4: Determine whether it is an NLOS signal based on the segmentation result of the sky image of the current positioning point and the position of the satellite projection in the sky image.

2. The all-day GNSS NLOS recognition method based on a fisheye camera as claimed in claim 1, characterized in that: Preprocess the sky image of the current positioning point, including: The image processed by Gaussian filtering is subjected to edge detection using the Canny operator to obtain the edge information of the image; The region growing method is used to obtain the candidate sky region seed points for the image after Gaussian filtering, which specifically includes: firstly obtaining the sky region seed points, clustering them using the K-Means++ clustering algorithm, obtaining the clustering results represented by a binary image, performing a connected domain analysis on the binary image, and obtaining the connected domain centroid as the candidate sky region seed point; A brightness coefficient is obtained according to the mean value of the deviation between the whole image and the set grayscale reference value of the sky, and the mean value of the deviation between the evaluation area and the set grayscale reference value of the sky. According to the relationship between the brightness coefficient and the brightness threshold of the sky area, it is judged whether the cluster center is located in the sky area, so as to select the sky seed point from the candidate of the sky area seed point, wherein the evaluation area is a circle with the cluster center as the center and the preset distance as the radius; The selected sky points are used as seed points, and the edge mask detected in edge detection is used to flood fill the pixels in the sky area until the edge is reached, and the sky image segmentation result P is obtained. skymask , P skymask It is a sky area or a non-sky area.

3. The all-day GNSS NLOS recognition method based on a fisheye camera as claimed in claim 2, characterized in that: Perform connected domain analysis on the binary image and obtain the connected domain centroid as a candidate for the sky region seed point, including: Perform connected domain statistics on the binary image, mark the centroid of each connected domain, and calculate the area of ​​each connected domain; If the area of ​​the connected domain is greater than the set connected domain area threshold, the centroid of the connected domain is used as a candidate for the sky region seed point.

4. The all-day GNSS NLOS recognition method based on a fisheye camera as claimed in claim 2, characterized in that: The brightness coefficient is obtained according to the average value of the deviation between the entire image and the grayscale reference value of the set sky, and the average value of the deviation between the evaluation area and the grayscale reference value of the set sky, including: Calculate the mean deviation between the entire image and the grayscale reference value of the sky. The specific calculation method is: Where D represents the mean deviation between the entire image and the grayscale reference value of the sky, V k Represents the gray value of the kth point in the entire image, V sky Indicates the grayscale reference value of the sky, and S represents the entire image; Calculate the mean deviation between the evaluation area and the grayscale reference value of the sky. The specific calculation method is: Where M represents the mean deviation between the entire image and the grayscale reference value of the sky, and h i is the number of pixels with gray value i; The brightness coefficient is obtained according to the formula: Where L is the brightness coefficient.

5. The all-day GNSS NLOS recognition method based on a fisheye camera as claimed in claim 4, characterized in that: According to the relationship between the brightness coefficient and the brightness threshold of the sky area, it is judged whether the cluster center is located in the sky area, so as to select the sky seed point from the candidate sky area seed points, including: Set the brightness threshold L of the sky area threshold1 and L threshold2 , if L threshold1 <L<L threshold2 , it means that the cluster center is located in the sky area, so the sky seed point is selected from the candidate sky area seed points.

6. The all-day GNSS NLOS recognition method based on a fisheye camera as claimed in claim 1, characterized in that: Step S3 includes: According to the current GNSS / INS integrated navigation positioning results and satellite ephemeris, the satellite altitude and azimuth are calculated. According to the stereoscopic projection model and focal length of the fisheye lens, the distance from the satellite projection position to the center of the sky image is calculated: Among them, f c is the focal length of the fisheye lens, ε sat is the satellite altitude angle, d pix is the distance from the satellite projection position to the center of the sky image. The coordinates of the center of the sky image are (x c ,y c ); According to the solved heading angle yaw, calculate the position of the satellite projection on the sky map: x sat =x c +d pix ·cos(yaw+α sat ) and sat =and c -d pix ·sin(yaw+α sat ) Among them, yaw is the heading angle of the current carrier, α sat is the satellite azimuth, (x sat ,y sat ) are the position coordinates of the satellite projection on the sky map.

7. The all-day GNSS NLOS recognition method based on a fisheye camera as claimed in claim 1, characterized in that: The sky image segmentation result is P skymask and the satellite projection position coordinates on the image are (x sat ,y sat ), step S4 comprises: If P skymask (x sat ,y sat )=0, image matrix P skymask If the value of the position of the satellite projection on the image is 0, the satellite signal is an NLOS signal; otherwise, the signal is a LOS signal.

8. The all-day GNSS NLOS recognition device based on fisheye camera is characterized by: The device includes: An image segmentation module is used to obtain a sky image of the current positioning point using a fisheye lens pointing to the sky, and determine whether the current environment is a night environment. If not, the sky image of the current positioning point is preprocessed to segment the sky area and the non-sky area. If yes, the sky image of the current positioning point is style-converted using cycleGAN to convert it into a daytime-style sky image, and then the daytime-style sky image is preprocessed to segment the sky area and the non-sky area. The preprocessing of the sky image of the current positioning point includes: performing color space transformation on the read RGB image, converting it into a grayscale image, and performing Gaussian filtering. The integrated navigation module is used to perform inertial navigation mechanical arrangement according to the IMU initial attitude and the collected IMU observation data, and use the Kalman filtering method to fuse the GNSS data and INS data to obtain the integrated navigation positioning result, wherein the integrated navigation positioning result includes the heading angle of the current carrier; The satellite position calculation module is used to calculate the distance from the satellite projection position to the center of the sky image based on the current GNSS / INS integrated navigation positioning result, satellite ephemeris, the fisheye lens stereoscopic projection model and focal length, and calculate the position of the satellite projection in the sky image based on the current carrier heading angle and the position of the sky image center; The NLOS signal judgment module is used to judge whether it is an NLOS signal according to the segmentation result of the sky image of the current positioning point and the position of the satellite projection in the sky image.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the all-day GNSS NLOS recognition method based on a fisheye camera is implemented as claimed in any one of claims 1 to 7.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the all-day GNSS NLOS recognition method based on a fisheye camera is implemented as described in any one of claims 1 to 7.

Images