Positioning verification method and apparatus, and electronic device and storage medium

By extracting and matching the current frame and reference frame feature information in the preset scene, the problem of low positioning accuracy in scenarios where GNSS positioning cannot be used, the accurate positioning determination in large indoor environments is achieved, and the positioning results are improved. accuracy.

WO2024230076A1PCT designated stage expired Publication Date: 2024-11-14ECARX (HUBEI) TECHCO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2023/126978
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-06
Filing Date
2023-10-27
Publication Date
2024-11-14

AI Technical Summary

Technical Problem

In scenarios where external edge positioning information such as GNSS cannot be used, the positioning result accuracy or failure is low, especially in large indoor similar and environmentally changing scenarios, such as underground garages or different floors of buildings, resulting in the unaccurate determination of the equipment to be located. Information about your own position.

Method used

By extracting the feature information of the current frame and reference frame generated based on preset scene positioning, the positioning verification result is determined according to the matching results of the feature information, the false positive results in the initial positioning result are filtered out, and the accuracy of the positioning result is improved. The specific method includes obtaining the feature information of the current frame and the reference frame, using image feature extraction rules such as SIFT, SURF, ORB, etc. to extract feature points, and matching feature points through preset matching rules and random sampling consistency rules, and determining positioning verification. result.

Benefits of technology

In scenarios where external edge positioning information such as GNSS cannot be used, the accuracy of positioning results is improved, and false positive results are effectively filtered out, ensuring the reliability of positioning results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2023126978_14112024_PF_FP_ABST
    Figure CN2023126978_14112024_PF_FP_ABST
Patent Text Reader

Abstract

A positioning verification method and apparatus, and an electronic device and a storage medium. The method comprises: acquiring the current frame and a reference frame which are generated by means of the positioning of a preset scene (S110); extracting feature information respectively corresponding to the current frame and the reference frame (S120); and determining a positioning verification result according to a matching result of the feature information (S130).
Need to check novelty before this filing date? Find Prior Art

Description

Positioning verification method, device, electronic device and storage medium

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on May 6, 2023, with application number 202310498500.1, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of positioning technology, for example, to a positioning verification method, device, electronic device and storage medium. Background Art

[0003] Positioning refers to the process by which a device to be positioned determines its own position. Taking repositioning as an example, repositioning technology can help a device to be positioned re-determine its own position in the event of positioning failure, such as when the captured image is too blurry due to excessive motion, or when the surrounding brightness is too high or too low. Positioning methods typically rely heavily on peripheral positioning information such as the Global Navigation Satellite System (GNSS). In some large indoor scenarios with similar but changing environments, such as underground garages or different floors of a building, the inability to obtain peripheral positioning information such as GNSS can lead to positioning failures or low positioning accuracy.

[0004] Summary of the Invention

[0005] The present application provides a positioning verification method, device, electronic device and storage medium to achieve the purpose of extracting feature information of a current frame and a reference frame generated based on preset scene positioning, and determining the positioning verification result according to the matching result of the extracted feature information. The method is applicable to scenarios where positioning cannot be performed using external positioning information such as GNSS, and filters out false positive results in the initial positioning results based on the current frame and the reference frame, thereby improving the accuracy of the positioning results.

[0006] The present invention provides a method for verifying positioning, which includes:

[0007] Obtain the current frame and reference frame generated by positioning of the preset scene;

[0008] Extract feature information corresponding to the current frame and the reference frame respectively;

[0009] The positioning verification result is determined according to the matching result of the feature information.

[0010] The present invention also provides a positioning verification device, which includes:

[0011] A data acquisition module, configured to acquire a current frame and a reference frame generated by positioning of a preset scene;

[0012] A feature extraction module configured to extract feature information corresponding to the current frame and the reference frame respectively;

[0013] The result determination module is configured to determine the positioning verification result according to the matching result of the feature information.

[0014] An embodiment of the present application further provides an electronic device, comprising:

[0015] at least one processor; and

[0016] a memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the positioning verification method described in any embodiment of the present application.

[0018] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the positioning verification method described in any embodiment of the present application when executed. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] FIG1 is a flow chart of a positioning verification method provided according to Embodiment 1 of the present application;

[0020] FIG2 is a flow chart of a positioning verification method provided according to Embodiment 2 of the present application;

[0021] FIG3 is a flow chart of a positioning verification method provided according to Embodiment 3 of the present application;

[0022] FIG4 is a flow chart of a positioning verification method provided according to Embodiment 4 of the present application;

[0023] FIG5 is a flow chart of a positioning verification method provided according to Embodiment 5 of the present application;

[0024] FIG6 is a flowchart of another positioning verification method provided according to Embodiment 5 of the present application;

[0025] FIG7 is a schematic structural diagram of a positioning verification device provided according to Example 6 of the present application;

[0026] FIG8 is a schematic diagram of the structure of an electronic device that implements the positioning verification method according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application. The described embodiments are only part of the embodiments of the present application, rather than all the embodiments.

[0028] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. The numbers used in this way can be interchanged where appropriate, so that the embodiments of the application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units listed, but may include other steps or units that are not listed or that are inherent to these processes, methods, products or devices.

[0029] Example 1

[0030] Figure 1 is a flowchart of a positioning verification method provided in Example 1 of the present application. This embodiment is applicable to situations where positioning results need to be verified. The method can be performed by a positioning verification device, which can be implemented in the form of hardware and / or software. The positioning verification device can be configured in an electronic device. For example, the electronic device can include various smart wearable devices, unmanned vehicles, mobile robots, and drones. As shown in Figure 1, a positioning verification method provided in Example 1 includes the following steps.

[0031] S110: Acquire a current frame and a reference frame generated by positioning of a preset scene.

[0032] Among them, the preset scenes can be understood as some scenes that cannot be positioned using external positioning information such as GNSS. The preset scenes can generally include some large indoor scenes that are similar and have changing environments, such as underground garages or different floors of buildings. Positioning can refer to the technology for the device to be positioned to determine its own posture information. For example, positioning can include repositioning or other positioning technologies. The devices to be positioned here can include a variety of smart wearable devices, driverless cars, mobile robots, and drones. The current frame can refer to a frame of image captured by the device to be positioned. The current frame can be, for example, the frame of image with the latest corresponding timestamp in the image frame captured by the device to be positioned, or any frame of image in the image frame captured by the device to be positioned. The embodiment of the present application does not limit this. The reference frame can refer to the frame of image that is most similar to the current frame among all the historical frame images captured by the device to be positioned. The posture corresponding to the reference frame can be used to determine the posture corresponding to the current frame.

[0033] In an embodiment of the present application, the current frame and reference frame generated by the positioning of the device to be positioned in the preset scene can be obtained through some positioning means, that is, the initial positioning result is obtained, so that the initial positioning result can be verified later, wherein the methods for obtaining the current frame and reference frame generated by the positioning of the preset scene can include the following: the current frame and historical frame collected by the device to be positioned in the preset scene, the historical frame is used to build a global semantic map or a global feature map, the feature information of the current frame and all key frames (i.e., historical frames) in the map is extracted, the feature information corresponding to the current frame and the key frame is matched, the key frame with the highest similarity to the current frame is determined, and the key frame is used as the reference frame; the current frame and reference frame generated by the positioning of the preset scene can be obtained through some third-party channels, for example, the current frame and reference frame can be determined by other positioning algorithms. The focus of the embodiment of the present application is to verify the initial positioning results corresponding to the current frame and the reference frame, that is, to determine whether the initial positioning result is wrong. Therefore, the embodiment of the present application does not limit the method for obtaining the current frame and reference frame generated by the positioning of the preset scene.

[0034] S120: Extract feature information corresponding to the current frame and the reference frame respectively.

[0035] Among them, feature information can refer to global or local feature information of the image, and the feature information may include: Scale-Invariant Feature Transform (SIFT) features of the image, Speeded Up Robust Features (SURF) features, Adaptive Accelerated Segmentation Experiment features and Rotated Binary Robust Independent Basic Features (Oriented FAST and Rotated BRIEF, ORB) features, Histogram of Oriented Gradients (HOG) features, Features From Accelerated Segment Test (FAST) features, Binary Robust Invariant Scalable Keypoints (BRISK) features, Local Binary Pattern (LBP) features, and Enhanced Efficient Binary Local Image Descriptor (BEBLID) features, etc.

[0036] In an embodiment of the present application, corresponding feature information can be extracted from the acquired current frame and reference frame respectively by calling image extraction rules pre-configured on the electronic device, wherein the feature information may include the following: SIFT features, SURF features, ORB features, HOG features, FAST features, BRISK features, LBP features, and BEBLID features, etc. The feature information extraction methods may include the following: corresponding image geometric features can be extracted from the acquired current frame and reference frame respectively by image geometric feature extraction rules pre-configured on the electronic device; the current frame and reference frame can be respectively input into a trained feature extraction network model, and the data results of the model can be used as the feature information corresponding to the current frame and the reference frame respectively.

[0037] S130: Determine the positioning verification result according to the matching result of the feature information.

[0038] Among them, the matching result may refer to the image feature matching result between the feature information corresponding to the current frame and the reference frame respectively. The positioning verification result can be understood as the verification result determined for the current frame and the reference frame in the original positioning information. The positioning verification result may include two results: a correct verification and a verification error. The positioning verification result may also be represented by 1 or Y to indicate a correct verification, and by 0 or N to indicate a verification error. For example, when the verification is correct, the positioning verification result may also include the posture information and timestamp corresponding to the current frame, etc., which is not limited in the embodiments of the present application.

[0039] In an embodiment of the present application, feature point matching can be performed on the feature information corresponding to the current frame and the reference frame respectively by calling the image matching rules pre-configured on the electronic device, and the matching of the feature points can be used as the matching result of the feature information, and the positioning verification result can be determined based on the matching result. The method of determining the positioning verification result based on the matching result can be: if the number of correctly matched features in the matching result exceeds a preset threshold, the positioning verification result is determined as a correct verification, otherwise, the positioning verification result is determined as a verification error; it can also be: if the proportion of the number of correctly matched features in the matching result exceeds a preset threshold, the positioning verification result is determined as a correct verification, otherwise, the positioning verification result is determined as a verification error. The embodiment of the present application does not limit this. The positioning verification result can be used to filter out false positive results in the initial positioning results based on the current frame and the reference frame, thereby improving the accuracy of the positioning results.

[0040] The technical solution of the embodiment of the present application is to obtain the current frame and reference frame generated by positioning of a preset scene, extract the feature information corresponding to the current frame and reference frame respectively, and determine the positioning verification result according to the matching results of the feature information. The embodiment of the present application extracts the feature information of the current frame and reference frame generated based on positioning of the preset scene, and determines the positioning verification result according to the matching results of the extracted feature information. It is suitable for scenarios where positioning cannot be performed using external positioning information such as GNSS, filters out false positive results in the initial positioning results based on the current frame and reference frame, and improves the accuracy of the positioning results.

[0041] Example 2

[0042] Figure 2 is a flow chart of a positioning verification method provided in Example 2 of this application. This example, based on the above examples, provides an implementation of a positioning verification method that can determine a positioning verification result based on the image dimension reduction ORB features of the current frame and the reference frame. As shown in Figure 2, the positioning verification method provided in Example 2 of this application includes the following steps.

[0043] S210: Acquire a current frame and a reference frame generated by positioning of a preset scene.

[0044] S220. Extract corresponding image ORB features from the current frame and the reference frame respectively, perform dimensionality reduction processing on the binary robust independent elementary features (BRIEF) descriptors in the image ORB features, and use the image ORB features after dimensionality reduction processing as the image dimensionality reduction ORB features corresponding to the current frame and the reference frame respectively.

[0045] Here, ORB may refer to an image descriptor. The image ORB feature has good scale invariance and rotation invariance, and can provide better image feature information for feature matching between the current frame and the reference frame. The image dimensionality reduction ORB feature may refer to the image ORB feature after feature dimensionality reduction.

[0046] In an embodiment of the present application, some image feature extraction rules can be called to extract corresponding image ORB features from the acquired current frame and reference frame, perform feature dimensionality reduction on the image ORB features, remove redundant information in the image ORB features and retain valid information, wherein the feature dimensionality reduction method may not be limited to: principal component analysis (PCA) and linear discriminant analysis (LDA), etc., and the image ORB features after dimensionality reduction processing are used as the image dimensionality reduction ORB features corresponding to the current frame and the reference frame respectively. The image ORB features in the embodiment of the present application are only used as examples, and other feature information of the image may include the following: SIFT features, SURF features, ORB features, HOG features, FAST features, BRISK features, LBP features and BEBLID features, etc. The above image features are all applicable to the technical solutions of the embodiment of the present application, and the embodiment of the present application is not limited to this.

[0047] S230 , calling a preset image matching rule to perform feature point matching on the image dimensionality reduction ORB features in the feature information corresponding to the current frame and the reference frame, and using the successfully matched image dimensionality reduction ORB features corresponding to the current frame and the reference frame as matching point pairs.

[0048] The preset image matching rule may refer to a pre-configured image matching rule, and the preset image matching rule may include brute force matching and nearest neighbor matching.

[0049] In an embodiment of the present application, one or more preset image matching rules can be pre-configured on the electronic device, and preset image matching rules such as brute force matching and nearest neighbor matching can be called to perform feature point matching on the image dimensionality reduction ORB features in the feature information corresponding to the current frame and the reference frame respectively. For example, by comparing the Euclidean distance or Hamming distance between the image dimensionality reduction ORB features of the current frame and the reference frame, a group of feature points with high similarity can be determined, and the image dimensionality reduction ORB features corresponding to the successfully matched current frame and the reference frame respectively can be used as matching point pairs.

[0050] S240 , using a random sampling consistency rule to determine inliers within the matching point pair that meet a preset matching accuracy and outliers that do not meet the preset matching accuracy.

[0051] Among them, the Random Sample Consensus (RANSAC) rule can refer to a rule that can be used to determine the inliers and outliers in the matching point pairs. The RANSAC rule can be used to remove incorrect matching pairs in the matching point pairs and improve the image matching accuracy. The preset matching accuracy can be understood as a pre-configured parameter for measuring the accuracy of image feature matching. The preset matching accuracy can include a preset Hamming distance threshold or a preset Euclidean distance threshold. An inlier can refer to a valid and correctly matched matching point pair in the matching point pair. An outlier can refer to an invalid or incorrectly matched matching point pair in the matching point pair.

[0052] In an embodiment of the present application, the RANSAC rule can be called to model and optimize the matching point pairs, and the incorrect matching pairs in the matching point pairs can be eliminated, that is, the inner points in the matching point pairs that meet the preset matching accuracy and the outer points that do not meet the preset matching accuracy are determined.

[0053] S250: Determine the positioning verification result based on the interior points and exterior points.

[0054] In the embodiment of the present application, the method of determining the positioning verification result based on the interior points and the exterior points may include at least one of the following:

[0055] When the number of inliers is greater than a preset verification threshold, the positioning verification result is determined to be correct; when the number of inliers is less than or equal to the preset verification threshold, the positioning verification result is determined to be incorrect;

[0056] When the ratio of the number of inner points to the number of outer points is greater than the preset verification ratio threshold, the positioning verification result is determined to be a correct verification. When the ratio of the number of inner points to the number of outer points is less than or equal to the preset verification ratio threshold, the positioning verification result is determined to be a verification error.

[0057] The technical solution of the embodiment of the present application is to obtain the current frame and reference frame generated by the positioning of the preset scene, extract the corresponding image ORB features from the current frame and the reference frame respectively, perform dimensionality reduction processing on the BRIEF descriptor in the image ORB features, and use the image ORB features after dimensionality reduction processing as the image dimensionality reduction ORB features corresponding to the current frame and the reference frame respectively, call the preset image matching rules to perform feature point matching on the image dimensionality reduction ORB features in the feature information corresponding to the current frame and the reference frame respectively, and use the image dimensionality reduction ORB features corresponding to the successfully matched current frame and reference frame respectively as matching point pairs, use the random sampling consistency rule to determine the inner points in the matching point pairs that meet the preset matching accuracy and the outer points that do not meet the preset matching accuracy, and determine the positioning verification result based on the inner points and the outer points. The embodiment of the present application extracts the image dimensionality reduction ORB features of the current frame and the reference frame generated based on the preset scene positioning, uses the preset image matching rules and the random sampling consistency rules to match the feature points of the image dimensionality reduction ORB features of the current frame and the reference frame and determine the corresponding inner points and outer points, and determines the positioning verification result based on the inner points and outer points. It can be applied to scenarios where positioning cannot be performed using external edge positioning information such as GNSS, filters out false positive results in the initial positioning results based on the current frame and the reference frame, and improves the accuracy of the positioning results.

[0058] Example 3

[0059] Figure 3 is a flowchart of a positioning verification method provided in Example 3 of this application. This example, based on the above examples, provides an implementation of a positioning verification method that can determine a positioning verification result based on the image BEBLID features of the current frame and the reference frame. As shown in Figure 3, the positioning verification method provided in Example 3 of this application includes the following steps.

[0060] S310: Acquire a current frame and a reference frame generated by positioning of a preset scene.

[0061] S320 , extracting corresponding image BEBLID features from the current frame and the reference frame respectively.

[0062] Among them, BEBLID may refer to an image descriptor. The BEBLID descriptor has high performance and robustness and can provide better image feature information for feature matching between the current frame and the reference frame.

[0063] In the embodiment of the present application, some image feature extraction rules can be called to extract corresponding image BEBLID features from the acquired current frame and reference frame respectively, so as to be used for feature matching later.

[0064] S330 , calling a preset image matching rule to perform feature point matching on the image BEBLID features in the feature information corresponding to the current frame and the reference frame, and using the successfully matched image BEBLID features corresponding to the current frame and the reference frame as matching point pairs.

[0065] S340 , using a random sampling consistency rule to determine the inner points within the matching point pair that meet the preset matching accuracy and the outer points that do not meet the preset matching accuracy.

[0066] S350: Determine the positioning verification result based on the interior points and exterior points.

[0067] The technical solution of the embodiment of the present application is to obtain the current frame and reference frame generated by the positioning of the preset scene, extract the corresponding image BEBLID features from the current frame and the reference frame respectively, call the preset image matching rules to perform feature point matching on the image BEBLID features in the feature information corresponding to the current frame and the reference frame respectively, and use the image BEBLID features corresponding to the successfully matched current frame and reference frame respectively as matching point pairs, use the random sampling consistency rule to determine the inliers that meet the preset matching accuracy and the outliers that do not meet the preset matching accuracy within the matching point pairs, and determine the positioning verification result based on the inliers and outliers. The embodiment of the present application extracts the image BEBLID features of the current frame and the reference frame generated based on the positioning of the preset scene, uses the preset image matching rules and the random sampling consistency rule to perform feature point matching on the image BEBLID features of the current frame and the reference frame and determine the corresponding inliers and outliers, and determines the positioning verification result based on the inliers and outliers. It can be applied to scenarios where positioning cannot be performed using external positioning information such as GNSS, filters out false positive results in the initial positioning results based on the current frame and the reference frame, and improves the accuracy of the positioning results.

[0068] Example 4

[0069] Figure 4 is a flowchart of a positioning verification method provided in Example 4 of this application. This embodiment, building on the previous embodiments, provides an implementation of a positioning verification method that can determine a positioning verification result based on the image BEBLID features and image dimensionality reduction BEBLID features of the current frame and reference frame. As shown in Figure 4, the positioning verification method provided in Example 4 of this application includes the following steps.

[0070] S410: Acquire a current frame and a reference frame generated by positioning of a preset scene.

[0071] S420: Calling a preset image matching rule to perform feature point matching on the image BEBLID features in the feature information corresponding to the current frame and the reference frame, and taking the successfully matched image BEBLID features corresponding to the current frame and the reference frame as a first matching point pair.

[0072] The first matching point pair may be understood as a matching point pair corresponding to the BEBLID feature of the image.

[0073] S430: Determine, by using a random sampling consistency rule, a first inner point in the first matching point pair that satisfies a first preset matching accuracy and a first outer point that does not satisfy the first preset matching accuracy.

[0074] The first inlier point and the first outlier point may be understood as inlier points and outliers corresponding to the BEBLID feature of the image.

[0075] S440: Determine a first positioning verification result according to the first interior point and the first exterior point.

[0076] The first positioning verification result may be understood as a positioning verification result determined based on the first inner point and the first outer point, that is, a positioning verification result corresponding to the BEBLID feature of the image.

[0077] In the embodiment of the present application, the method for determining the first positioning verification result based on the first interior point and the first exterior point can be the same as that used in the above embodiment, that is, including at least one of the following:

[0078] When the number of the first inliers is greater than a preset verification threshold, the positioning verification result is determined to be correct; when the number of the first inliers is less than or equal to the preset verification threshold, the positioning verification result is determined to be incorrect;

[0079] When the ratio of the number of first inner points to the number of first outer points is greater than a preset verification ratio threshold, the positioning verification result is determined to be a correct verification; when the ratio of the number of first inner points to the number of first outer points is less than or equal to the preset verification ratio threshold, the positioning verification result is determined to be a verification error.

[0080] S450 , extracting corresponding image BEBLID features from the current frame and the reference frame respectively, performing dimensionality reduction processing on the BEBLID descriptors in the image BEBLID features, and using the image BEBLID features after dimensionality reduction processing as the image dimensionality reduction BEBLID features corresponding to the current frame and the reference frame respectively.

[0081] The image dimension reduction BEBLID feature may refer to the image BEBLID feature after feature dimension reduction.

[0082] In an embodiment of the present application, some image feature extraction rules can be called to extract corresponding image BEBLID features from the acquired current frame and reference frame, respectively, and feature dimensionality reduction is performed on the image BEBLID features to remove redundant information in the image BEBLID features and retain valid information. The feature dimensionality reduction method may not be limited to: PCA and LDA, etc., and the image BEBLID features after dimensionality reduction processing are used as the image dimensionality reduction BEBLID features corresponding to the current frame and the reference frame respectively.

[0083] S460: Calling a preset image matching rule to perform feature point matching on the image dimension reduction BEBLID features in the feature information corresponding to the current frame and the reference frame, and using the successfully matched image dimension reduction BEBLID features corresponding to the current frame and the reference frame as a second matching point pair.

[0084] The second matching point pair can be understood as a matching point pair corresponding to the image dimension reduction BEBLID feature.

[0085] S470: Determine, by using a random sampling consistency rule, a second inner point in the second matching point pair that satisfies a second preset matching accuracy and a second outer point that does not satisfy the second preset matching accuracy.

[0086] The second inliers and the second outliers may be understood as inliers and outliers corresponding to the BEBLID features of the image dimension reduction.

[0087] S480: Determine a second positioning verification result according to the second inner point and the second outer point.

[0088] The second positioning verification result can be understood as a positioning verification result determined according to the second inner point and the second outer point, that is, a positioning verification result corresponding to the image dimension reduction BEBLID feature.

[0089] In the embodiment of the present application, the method of determining the second positioning verification result based on the second inner point and the second outer point can be the same as that in the above embodiment, and will not be repeated here.

[0090] S490: Determine a positioning verification result according to the first positioning verification result and the second positioning verification result.

[0091] In an embodiment of the present application, the method of determining the positioning verification result according to the first positioning verification result and the second positioning verification result can be: if the first positioning verification result is a verification error, the final positioning verification result is determined to be a verification error; if the first positioning verification result is a verification correct and the second positioning verification result is also a verification correct, the final positioning verification result is determined to be a verification correct; if the first positioning verification result is a verification correct and the second positioning verification result is a verification error, the final positioning verification result is determined to be a verification error, that is, the second positioning verification result shall prevail.

[0092] The technical solution of the embodiment of the present application is to obtain the current frame and the reference frame generated by the positioning of the preset scene, call the preset image matching rules to perform feature point matching on the image BEBLID features in the feature information corresponding to the current frame and the reference frame respectively, and use the image BEBLID features corresponding to the successfully matched current frame and the reference frame respectively as the first matching point pair, use the random sampling consistency rule to determine the first inner point in the first matching point pair that meets the first preset matching accuracy and the first outer point that does not meet the first preset matching accuracy, determine the first positioning verification result based on the first inner point and the first outer point, extract the corresponding image BEBLID features from the current frame and the reference frame respectively, and match the BEBLID features in the image BEBLID features. The descriptor is subjected to dimensionality reduction processing, and the BEBLID features of the image after the dimensionality reduction processing are used as the image dimensionality reduction BEBLID features corresponding to the current frame and the reference frame respectively. The preset image matching rule is called to perform feature point matching on the image dimensionality reduction BEBLID features corresponding to the current frame and the reference frame respectively. The image dimensionality reduction BEBLID features corresponding to the current frame and the reference frame respectively that are successfully matched are used as the second matching point pair. The random sampling consistency rule is used to determine a second inner point in the second matching point pair that meets the second preset matching accuracy and a second outer point that does not meet the second preset matching accuracy. The second positioning verification result is determined according to the second inner point and the second outer point. The positioning verification result is determined according to the first positioning verification result and the second positioning verification result. The embodiments of the present application extract image BEBLID features and image dimension reduction BEBLID features of the current frame and reference frame generated based on preset scene positioning, and use preset image matching rules and random sampling consistency rules to determine a first positioning verification result based on the image BEBLID features and a second positioning verification result based on the image dimension reduction BEBLID features. The positioning verification result is determined based on the first positioning verification result and the second positioning verification result. This embodiment is applicable to scenarios where positioning cannot be performed using external positioning information such as GNSS, filters out false positive results in the initial positioning results based on the current frame and reference frame, and improves the accuracy of the positioning results.

[0093] Example 5

[0094] Figure 5 is a flowchart of a positioning verification method provided in Example 5 of the present application. This embodiment, based on the above embodiments, provides an implementation of a positioning verification method that can determine the corresponding third and fourth positioning verification results based on the image dimension reduction ORB features and image BEBLID features of the current frame and the reference frame, respectively, and fuse the third and fourth positioning verification results using a Kalman filter to generate a final positioning verification result. As shown in Figure 5, the positioning verification method provided in Example 5 of the present application includes the following steps.

[0095] S510: Acquire a current frame and a reference frame generated by positioning of a preset scene.

[0096] S520: Calling a preset image matching rule to perform feature point matching on the image dimensionality reduction ORB features in the feature information corresponding to the current frame and the reference frame respectively, and taking the successfully matched image dimensionality reduction ORB features corresponding to the current frame and the reference frame respectively as a third matching point pair.

[0097] The third matching point pair can be understood as a matching point pair corresponding to the image dimension reduction ORB feature.

[0098] S530: Determine, by using a random sampling consistency rule, a third inner point within the third matching point pair that satisfies a third preset matching accuracy and a third outer point that does not satisfy the third preset matching accuracy.

[0099] Among them, the third inliers and the third outliers can be understood as inliers and outliers corresponding to the ORB features of the image dimensionality reduction.

[0100] S540: Determine a third positioning verification result according to the third interior point and the third exterior point.

[0101] The third positioning verification result can be understood as a positioning verification result determined according to the third inner point and the third outer point, that is, a positioning verification result corresponding to the image dimensionality reduction ORB feature.

[0102] In the embodiment of the present application, the method of determining the third positioning verification result based on the third inner point and the third outer point can be the same as that in the above embodiment, and will not be repeated here.

[0103] S550 , calling a preset image matching rule to perform feature point matching on the image BEBLID features in the feature information corresponding to the current frame and the reference frame respectively, and taking the successfully matched image BEBLID features corresponding to the current frame and the reference frame respectively as a fourth matching point pair.

[0104] The fourth matching point pair may be understood as a matching point pair corresponding to the BEBLID feature of the image.

[0105] S560: Determine, by using a random sampling consistency rule, a fourth inner point within the fourth matching point pair that satisfies a fourth preset matching accuracy and a fourth outer point that does not satisfy the fourth preset matching accuracy.

[0106] The fourth inlier point and the fourth outlier point may be understood as inlier points and outliers corresponding to the BEBLID feature of the image.

[0107] S570: Determine a fourth positioning verification result according to the fourth interior point and the fourth exterior point.

[0108] The fourth positioning verification result may be understood as a positioning verification result determined according to the fourth interior point and the fourth exterior point, that is, a positioning verification result corresponding to the BEBLID feature of the image.

[0109] In the embodiment of the present application, the method of determining the fourth positioning verification result based on the fourth inner point and the fourth outer point can be the same as that in the above embodiment, and will not be repeated here.

[0110] S580: Use a Kalman filter to fuse the third positioning verification result and the fourth positioning verification result to obtain a positioning verification result.

[0111] Among them, the Kalman filter can make the best estimate of the system state through the system input observation data, and is a highly efficient recursive filter.

[0112] In an embodiment of the present application, after obtaining the third positioning verification result and the fourth positioning verification result, the third positioning verification result and the fourth positioning verification result can be fused using a Kalman filter to obtain a more accurate and robust positioning verification result. The fusion process can be as follows: the third positioning verification result based on the image dimensionality reduction ORB feature is used as the Kalman filter state for prediction, the third positioning verification result is propagated using the state transfer matrix, and the positioning verification result at the next moment is predicted. The Kalman filter gain coefficient is calculated based on the predicted positioning verification result, and the fourth positioning verification result based on the image BEBLID feature is used as the system observation. The predicted system state and its covariance matrix are updated to obtain the fused positioning verification result.

[0113] On the basis of the above-mentioned application embodiment, FIG6 is a flow chart of another positioning verification method provided in the fifth embodiment of the present application. As shown in FIG6 , the positioning verification method takes the repositioning scene as an example, and can realize the combination of four different positioning verification methods. The four positioning verification methods here have been introduced in the above-mentioned embodiment and will not be repeated here. In the construction of the map module, the map module may include multiple map key frames, each key frame may contain image feature information, timestamp, posture and other information, wherein a key frame refers to a frame of image selected as a key frame every time a distance (for example, 0.5 meters) is passed during the image acquisition process; the image feature information may refer to the ORB feature or BEBLID feature extracted from the image; the timestamp may refer to the time of image exposure; the posture may refer to the posture information corresponding to the key frame. Obtaining the semantic information of the current frame may refer to obtaining the image feature information of the current image frame, which may be an ORB feature or a BEBLID feature.

[0114] The technical solution of the embodiment of the present application is to obtain the current frame and reference frame generated by the positioning of the preset scene, call the preset image matching rules to perform feature point matching on the image dimensionality reduction ORB features in the feature information corresponding to the current frame and the reference frame respectively, and use the image dimensionality reduction ORB features corresponding to the successfully matched current frame and the reference frame respectively as the third matching point pair, use the random sampling consistency rule to determine the third inner point in the third matching point pair that meets the third preset matching accuracy and the third outer point that does not meet the third preset matching accuracy, determine the third positioning verification result based on the third inner point and the third outer point, call the preset image matching rules to perform feature point matching on the image BEBLID features in the feature information corresponding to the current frame and the reference frame respectively, and use the image BEBLID features corresponding to the successfully matched current frame and the reference frame respectively as the fourth matching point pair, use the random sampling consistency rule to determine the fourth inner point in the fourth matching point pair that meets the fourth preset matching accuracy and the fourth outer point that does not meet the fourth preset matching accuracy, determine the fourth positioning verification result based on the fourth inner point and the fourth outer point, and use the Kalman filter to fuse the third positioning verification result and the fourth positioning verification result to obtain the positioning verification result. The embodiment of the present application extracts the image dimensionality reduction ORB features and image BEBLID features of the current frame and the reference frame generated based on the preset scene positioning, and respectively uses the preset image matching rule and the random sampling consistency rule to determine the third positioning verification result based on the image dimensionality reduction ORB features and the fourth positioning verification result based on the image BEBLID features. The Kalman filter is used to fuse the third positioning verification result and the fourth positioning verification result to obtain the positioning verification result. The embodiment of the present application is applicable to scenarios where positioning cannot be performed using external positioning information such as GNSS, filters out false positive results in the initial positioning results based on the current frame and the reference frame, and improves the accuracy of the positioning results.

[0115] Example 6

[0116] FIG7 is a schematic diagram of the structure of a positioning verification device provided in Example 6 of the present application. As shown in FIG7 , the device includes the following modules.

[0117] A data acquisition module 61 is configured to acquire a current frame and a reference frame generated by positioning of a preset scene;

[0118] A feature extraction module 62 is configured to extract feature information corresponding to the current frame and the reference frame respectively;

[0119] The result determination module 63 is configured to determine the positioning verification result according to the matching result of the feature information.

[0120] The technical solution of the embodiment of the present application is to obtain the current frame and reference frame generated by positioning of a preset scene through a data acquisition module 61, extract feature information corresponding to the current frame and reference frame respectively through a feature extraction module 62, and determine the positioning verification result according to the matching result of the feature information through a result determination module 63. The embodiment of the present application extracts feature information of the current frame and reference frame generated based on positioning of the preset scene, and determines the positioning verification result according to the matching result of the extracted feature information. This embodiment is suitable for scenarios where positioning cannot be performed using external positioning information such as GNSS, filters out false positive results in the initial positioning results based on the current frame and reference frame, and improves the accuracy of the positioning results.

[0121] Based on the above application embodiment, the feature extraction module 62 includes at least one of the following:

[0122] The first feature extraction unit is configured to extract corresponding image ORB features from the current frame and the reference frame respectively, perform dimensionality reduction processing on the BRIEF descriptor in the image ORB features, and use the image ORB features after dimensionality reduction processing as the image dimensionality reduction ORB features corresponding to the current frame and the reference frame respectively.

[0123] The second feature extraction unit is configured to extract corresponding image BEBLID features from the current frame and the reference frame respectively.

[0124] The third feature extraction unit is configured to extract corresponding image BEBLID features from the current frame and the reference frame respectively, perform dimensionality reduction processing on the BEBLID descriptors in the image BEBLID features, and use the image BEBLID features after dimensionality reduction processing as the image dimensionality reduction BEBLID features corresponding to the current frame and the reference frame respectively.

[0125] Based on the above application embodiment, the result determination module 63 includes:

[0126] The dimensionality reduction ORB feature matching unit is configured to call a preset image matching rule to perform feature point matching on the image dimensionality reduction ORB features in the feature information corresponding to the current frame and the reference frame respectively, and use the image dimensionality reduction ORB features corresponding to the current frame and the reference frame respectively that have successfully matched as matching point pairs.

[0127] The inner and outer point determination unit is configured to determine inner points that meet a preset matching accuracy and outer points that do not meet the preset matching accuracy within a matching point pair using a random sampling consistency rule.

[0128] The verification result determination unit is configured to determine the positioning verification result based on the interior points and the exterior points.

[0129] Based on the above application embodiment, the result determination module 63 includes:

[0130] The BEBLID feature matching unit is configured to call a preset image matching rule to perform feature point matching on the image BEBLID features in the feature information corresponding to the current frame and the reference frame, and use the image BEBLID features corresponding to the current frame and the reference frame that are successfully matched as matching point pairs.

[0131] The inner and outer point determination unit is configured to determine inner points that meet a preset matching accuracy and outer points that do not meet the preset matching accuracy within a matching point pair using a random sampling consistency rule.

[0132] The verification result determination unit is configured to determine the positioning verification result based on the interior points and the exterior points.

[0133] Based on the above application embodiment, the result determination module 63 includes:

[0134] The first feature matching unit is configured to call a preset image matching rule to perform feature point matching on the image BEBLID features in the feature information corresponding to the current frame and the reference frame respectively, and use the image BEBLID features corresponding to the current frame and the reference frame respectively that have successfully matched as the first matching point pair.

[0135] The first inner and outer point determining unit is configured to determine a first inner point that meets a first preset matching accuracy and a first outer point that does not meet the first preset matching accuracy within the first matching point pair by using a random sampling consistency rule.

[0136] The first verification result determining unit is configured to determine a first positioning verification result according to the first interior point and the first exterior point.

[0137] The second feature matching unit is configured to call a preset image matching rule to perform feature point matching on the image dimensionality reduction BEBLID features in the feature information corresponding to the current frame and the reference frame respectively, and use the image dimensionality reduction BEBLID features corresponding to the current frame and the reference frame respectively that have successfully matched as the second matching point pair.

[0138] The second inner and outer point determining unit is configured to determine a second inner point that meets a second preset matching accuracy and a second outer point that does not meet the second preset matching accuracy within the second matching point pair by using a random sampling consistency rule.

[0139] The second verification result determining unit is configured to determine a second positioning verification result according to the second inner point and the second outer point.

[0140] The final verification result determining unit is configured to determine the positioning verification result according to the first positioning verification result and the second positioning verification result.

[0141] Based on the above application embodiment, the result determination module 63 includes:

[0142] The third feature matching unit is configured to call a preset image matching rule to perform feature point matching on the image dimensionality reduction ORB features in the feature information corresponding to the current frame and the reference frame respectively, and use the image dimensionality reduction ORB features corresponding to the current frame and the reference frame respectively that are successfully matched as the third matching point pair.

[0143] The third inner and outer point determining unit is configured to determine, by using a random sampling consistency rule, a third inner point in the third matching point pair that meets a third preset matching accuracy and a third outer point that does not meet the third preset matching accuracy.

[0144] The third verification result determining unit is configured to determine a third positioning verification result according to the third interior point and the third exterior point.

[0145] The fourth feature matching unit is configured to call a preset image matching rule to perform feature point matching on the image BEBLID features in the feature information corresponding to the current frame and the reference frame respectively, and use the image BEBLID features corresponding to the current frame and the reference frame respectively that have successfully matched as the fourth matching point pair.

[0146] The fourth inner and outer point determining unit is configured to determine, within the fourth matching point pair, a fourth inner point that meets a fourth preset matching accuracy and a fourth outer point that does not meet the fourth preset matching accuracy by using a random sampling consistency rule.

[0147] The fourth verification result determining unit is configured to determine a fourth positioning verification result according to the fourth interior point and the fourth exterior point.

[0148] The verification result fusion unit is configured to use a Kalman filter to fuse the third positioning verification result and the fourth positioning verification result to obtain a positioning verification result.

[0149] Based on the above application embodiment, the verification result determination unit is configured to be at least one of the following:

[0150] When the number of inliers is greater than a preset verification threshold, the positioning verification result is determined to be correct; when the number of inliers is less than or equal to the preset verification threshold, the positioning verification result is determined to be incorrect;

[0151] When the ratio of the number of inner points to the number of outer points is greater than the preset verification ratio threshold, the positioning verification result is determined to be a correct verification. When the ratio of the number of inner points to the number of outer points is less than or equal to the preset verification ratio threshold, the positioning verification result is determined to be a verification error.

[0152] The positioning verification device provided in the embodiments of the present application can execute the positioning verification method provided in any embodiment of the present application, and has the corresponding functional modules and effects of the execution method.

[0153] Example 7

[0154] FIG8 shows a block diagram of an electronic device 70 that can be used to implement an embodiment of the present application. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.

[0155] As shown in FIG8 , the electronic device 70 includes at least one processor 71 and a memory connected to the at least one processor 71, such as a read-only memory (ROM) 72 and a random access memory (RAM) 73. The memory stores a computer program that can be executed by the at least one processor 71. The processor 71 can perform various appropriate actions and processes according to the computer program stored in the ROM 72 or the computer program loaded from the storage unit 78 into the RAM 73. In the RAM 73, various programs and data required for the operation of the electronic device 70 can also be stored. The processor 71, ROM 72, and RAM 73 are connected to each other via a bus 74. An input / output (I / O) interface 75 is also connected to the bus 74.

[0156] Multiple components in the electronic device 70 are connected to the I / O interface 75, including an input unit 76, such as a keyboard, a mouse, etc.; an output unit 77, such as various types of displays, speakers, etc.; a storage unit 78, such as a magnetic disk, an optical disk, etc.; and a communication unit 79, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 79 allows the electronic device 70 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0157] Processor 71 can be a variety of general-purpose and / or specialized processing components with processing and computing capabilities. Some examples of processor 71 include a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 71 performs the various methods and processes described above, such as the positioning verification method.

[0158] In some embodiments, the positioning verification method can be implemented as a computer program that is tangibly contained in a computer-readable storage medium, such as storage unit 78. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 70 via ROM 72 and / or communication unit 79. When the computer program is loaded into RAM 73 and executed by processor 71, one or more steps of the positioning verification method described above can be performed. Alternatively, in other embodiments, processor 71 can be configured to perform the positioning verification method in any other suitable manner (e.g., via firmware).

[0159] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard parts (ASSPs), system on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0160] Computer programs for implementing the methods of the present application may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0161] In the context of the present application, computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage medium can include electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage medium can be a machine-readable signal medium. Machine-readable storage medium includes an electrical connection based on one or more lines, a portable computer disk, a hard disk, RAM, ROM, erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device or any suitable combination of the foregoing. Storage medium can be a non-transitory storage medium.

[0162] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0163] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0164] A computing system may include a client and a server. The client and server are generally remote from each other and typically interact via a communication network. The client-server relationship arises through computer programs running on the respective computers and establishing a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, a host product within a cloud computing service ecosystem that addresses the management difficulties and limited scalability of physical hosts and virtual private server (VPS) services.

[0165] The various forms of processes shown above can be used to reorder, add, or delete steps. For example, the multiple steps described in this application can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of this application can be achieved. This document is not limited here.

Claims

1. A positioning verification method, comprising: Obtain the current frame and reference frame generated by positioning of the preset scene; Extracting feature information corresponding to the current frame and the reference frame respectively; The positioning verification result is determined according to the matching result of the feature information.

2. The method according to claim 1, wherein: The extracting feature information corresponding to the current frame and the reference frame respectively includes at least one of the following: Extracting corresponding image adaptive accelerated segmentation experimental features and rotational binary robust independent basic features (ORB) features from the current frame and the reference frame, respectively, performing dimensionality reduction processing on the binary robust independent basic features (BRIEF) descriptors in the image ORB features, and using the image ORB features after dimensionality reduction processing as the image dimensionality reduction ORB features corresponding to the current frame and the reference frame, respectively; Extracting corresponding image enhanced efficient local image feature descriptor BEBLID features from the current frame and the reference frame respectively; Corresponding image BEBLID features are extracted from the current frame and the reference frame respectively, BEBLID descriptors in the image BEBLID features are subjected to dimensionality reduction processing, and the image BEBLID features after dimensionality reduction processing are used as image dimensionality reduction BEBLID features corresponding to the current frame and the reference frame respectively.

3. The method according to claim 1, wherein: The determining of the positioning verification result according to the matching result of the feature information includes: Calling a preset image matching rule to perform feature point matching on the image dimension reduction ORB features in the feature information corresponding to the current frame and the reference frame respectively, and using the image dimension reduction ORB features corresponding to the current frame and the reference frame respectively that have successfully matched as matching point pairs; Determine the inner points in the matching point pair that meet the preset matching accuracy and the outer points that do not meet the preset matching accuracy by using a random sampling consistency rule; The positioning verification result is determined according to the inner point and the outer point.

4. The method according to claim 1, wherein: The determining of the positioning verification result according to the matching result of the feature information includes: Calling a preset image matching rule to perform feature point matching on the image BEBLID features in the feature information corresponding to the current frame and the reference frame respectively, and taking the image BEBLID features corresponding to the current frame and the reference frame respectively that have been successfully matched as matching point pairs; Determine the inner points in the matching point pair that meet the preset matching accuracy and the outer points that do not meet the preset matching accuracy by using a random sampling consistency rule; The positioning verification result is determined according to the inner point and the outer point.

5. The method according to claim 1, wherein: The determining of the positioning verification result according to the matching result of the feature information includes: Calling a preset image matching rule to perform feature point matching on the image BEBLID features in the feature information corresponding to the current frame and the reference frame respectively, and taking the image BEBLID features corresponding to the current frame and the reference frame respectively that have been successfully matched as a first matching point pair; Determine, by using a random sampling consistency rule, a first inner point in the first matching point pair that satisfies a first preset matching accuracy and a first outer point that does not satisfy the first preset matching accuracy; Determine a first positioning verification result according to the first interior point and the first exterior point; Calling a preset image matching rule to perform feature point matching on the image dimension reduction BEBLID features in the feature information corresponding to the current frame and the reference frame respectively, and using the image dimension reduction BEBLID features corresponding to the current frame and the reference frame respectively that have been matched successfully as a second matching point pair; Determine, by using a random sampling consistency rule, a second inner point in the second matching point pair that satisfies a second preset matching accuracy and a second outer point that does not satisfy the second preset matching accuracy; Determine a second positioning verification result according to the second inner point and the second outer point; The positioning verification result is determined according to the first positioning verification result and the second positioning verification result.

6. The method according to claim 1, wherein: The determining of the positioning verification result according to the matching result of the feature information includes: Calling a preset image matching rule to perform feature point matching on the image dimension reduction ORB features in the feature information corresponding to the current frame and the reference frame respectively, and using the image dimension reduction ORB features corresponding to the current frame and the reference frame respectively that have been matched successfully as a third matching point pair; Determine, by using a random sampling consistency rule, a third inner point within the third matching point pair that satisfies a third preset matching accuracy and a third outer point that does not satisfy the third preset matching accuracy; Determine a third positioning verification result according to the third inner point and the third outer point; Calling a preset image matching rule to perform feature point matching on the image BEBLID features in the feature information corresponding to the current frame and the reference frame respectively, and taking the image BEBLID features corresponding to the current frame and the reference frame respectively that have been matched successfully as a fourth matching point pair; Determine, by using a random sampling consistency rule, a fourth inner point within the fourth matching point pair that satisfies a fourth preset matching accuracy and a fourth outer point that does not satisfy the fourth preset matching accuracy; Determine a fourth positioning verification result according to the fourth inner point and the fourth outer point; The third positioning verification result and the fourth positioning verification result are fused using a Kalman filter to obtain the positioning verification result.

7. The method according to claim 3 or 4, wherein: The determining the positioning verification result according to the interior point and the exterior point comprises at least one of the following: When the number of the inner points is greater than a preset verification threshold, the positioning verification result is determined as a verification correct; when the number of the inner points is less than or equal to the preset verification threshold, the positioning verification result is determined as a verification error; When the ratio of the number of inner points to the number of outer points is greater than a preset verification ratio threshold, the positioning verification result is determined to be a correct verification; when the ratio of the number of inner points to the number of outer points is less than or equal to the preset verification ratio threshold, the positioning verification result is determined to be a verification error.

8. A positioning verification device, comprising: A data acquisition module, configured to acquire a current frame and a reference frame generated by positioning of a preset scene; A feature extraction module, configured to extract feature information corresponding to the current frame and the reference frame respectively; A result determination module is configured to determine a positioning verification result according to a matching result of the feature information.

9. An electronic device, comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the positioning verification method described in any one of claims 1 to 7.

10. A computer-readable storage medium, wherein: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the positioning verification method according to any one of claims 1 to 7 when executed.

Citation Information

Patent Citations

  • Slam method in dynamic environment

    CN110648354A

  • Repositioning method and device based on visual map

    CN111750864A

  • Heavy equipment production scene video monitoring and image stabilizing method and system based on gyroscope

    CN114979489A

  • Relocation frame determination method and device, vehicle and storage medium

    CN115512124A

  • Positioning verification method and device, electronic equipment and storage medium

    CN116258769A