Attitude Reference Update Method and Device

By dynamically adjusting the attitude benchmark, the problem of changing attitude angle benchmarks caused by different driving habits is solved, and the accuracy and detection accuracy of behavior detection are improved.

CN114399749BActive Publication Date: 2025-06-13JILUO TECH (SHANGHAI) CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202111444087.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-11-30
Publication Date
2025-06-13
Estimated Expiration
2041-11-30

AI Technical Summary

Technical Problem

In the prior art, the attitude angle reference changes due to different driving habits, resulting in poor behavior detection accuracy.

Method used

By obtaining the attitude reference in the preset time period based on the predicted attitude, calculating the deviation, and comparing the deviation with the maximum and minimum attitude reference in the preset time period, the attitude reference is dynamically adjusted.

Benefits of technology

It improves the accuracy of abnormal behavior detection, avoids mis-detection caused by driving habits or changes in drivers, and improves detection accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114399749B_ABST
    Figure CN114399749B_ABST
Patent Text Reader

Abstract

The present invention provides a method and device for updating an attitude reference. The method includes: obtaining an attitude reference within a preset time period according to a predicted attitude previously obtained based on a target image; obtaining a deviation between the attitude reference corresponding to the moment immediately preceding the predicted attitude within the preset time period and the predicted attitude; comparing the deviation with the maximum attitude reference and the minimum reference in the attitude reference within the preset time period respectively, and determining the attitude reference corresponding to the moment of the predicted attitude based on the predicted attitude and the attitude reference within the preset time period according to the comparison results. By calculating the deviation between the predicted attitude obtained by detecting the target image and the attitude reference at the moment immediately preceding the moment corresponding to the predicted attitude within the preset time period, and comparing the deviation with the maximum attitude reference and the minimum attitude reference within the preset time period, the present invention determines the attitude reference corresponding to the predicted attitude, thereby dynamically adjusting the attitude reference.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a method and device for updating an attitude reference. Background Art

[0002] With the rise of deep learning, the research on driver behavior attitude monitoring based on images and videos is constantly developing, and the research attitude area is constantly expanding. With the increasing development and popularity of vehicles, it has become increasingly important to detect the state changes (such as fatigue driving, playing with mobile phones, smoking, etc.) of drivers during driving in real time to avoid accidents.

[0003] Currently, a behavior attitude detection method is usually used to detect the behavior attitude of the acquired driver image, and the detected behavior attitude is compared with a preset attitude reference, so as to obtain a behavior attitude abnormal detection result.

[0004] Due to the camera positions of different vehicles, different drivers, and different driving habits, the corresponding attitude reference needs to be continuously re-acquired during the driving process. However, most of the attitude references in the above behavior attitude abnormal detection methods are preset, so it is difficult to accurately determine whether the behavior attitude of the current driver is abnormal according to the preset attitude reference. Summary of the Invention

[0005] The present invention provides a method and device for updating a face attitude reference, which are used to solve the defect in the prior art that the attitude angle reference changes due to different driving habits, resulting in poor accuracy of behavior detection, and to realize the automatic update of the attitude reference and improve the accuracy of the detection result.

[0006] The present invention provides a method for updating an attitude reference, including: obtaining an attitude reference within a preset time period according to a predicted attitude obtained in advance based on a target image; obtaining a deviation between the attitude reference within the preset time period and the attitude reference corresponding to the moment prior to the predicted attitude adjacent thereto; comparing the deviation with the maximum attitude reference and the minimum reference in the attitude reference within the preset time period respectively, and determining the attitude reference corresponding to the moment of the predicted attitude based on the comparison result, the predicted attitude, and the attitude reference within the preset time period.

[0007] A method for updating an attitude reference provided by the present invention, based on the comparison result, determining the attitude reference at the moment corresponding to the predicted attitude based on the predicted attitude and the attitude reference within the preset time period, includes: based on that the deviation is greater than the minimum attitude reference within the preset time period and less than the maximum attitude reference within the preset time period, obtaining the attitude reference at the moment corresponding to the predicted attitude according to the predicted attitude, the deviation, and the attitude reference within the preset time period.

[0008] A method for updating an attitude reference provided by the present invention, the attitude reference at the moment corresponding to the predicted attitude, represents:

[0009]

[0010] wherein, A(t) represents the attitude reference at the moment corresponding to the predicted attitude, x represents the predicted attitude, Δ represents the deviation, and M represents the time length.

[0011] A method for updating an attitude reference provided by the present invention, obtaining the deviation between the attitude references corresponding to the moments adjacent to and prior to the predicted attitude according to the attitude reference within the preset time period and the predicted attitude, includes: obtaining the attitude reference corresponding to the moment adjacent to and prior to the predicted attitude according to the attitude reference within the preset time period; obtaining the deviation based on the predicted attitude and the attitude reference at the adjacent prior moment.

[0012] A method for updating an attitude reference provided by the present invention, obtaining the attitude reference within the preset time period according to the predicted attitude obtained in advance based on the target image, includes: based on the predicted attitude obtained in advance based on the target image, obtaining the attitude reference within the preset time period including the moment corresponding to the predicted attitude by using a sliding window with a preset length.

[0013] A method for updating an attitude reference provided by the present invention, based on the comparison result, determining the attitude reference at the moment corresponding to the predicted attitude based on the predicted attitude and the attitude reference within the preset time period, further includes: based on that the deviation is less than or equal to the minimum attitude reference within the preset time period, or based on that the deviation is greater than or equal to the maximum attitude reference within the preset time period, obtaining the attitude reference corresponding to the moment adjacent to and prior to the predicted attitude according to the attitude reference within the preset time period, and using it as the attitude reference at the moment corresponding to the predicted attitude.

[0014] The present invention also provides an attitude reference updating device, including: a data acquisition module, which acquires an attitude reference within a preset time period according to a predicted attitude obtained in advance based on a target image; a deviation acquisition module, which obtains a deviation between the attitude reference corresponding to the moment prior to the predicted attitude based on the attitude reference within the preset time period and the predicted attitude; a reference acquisition module, which compares the deviation with the maximum attitude reference and the minimum reference in the attitude reference within the preset time period respectively, and determines the attitude reference corresponding to the moment of the predicted attitude based on the comparison result, the predicted attitude and the attitude reference within the preset time period.

[0015] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of the attitude reference updating method as described in any one of the above are implemented.

[0016] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the attitude reference updating method as described in any one of the above are implemented.

[0017] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, the steps of the attitude reference updating method as described in any one of the above are implemented.

[0018] The face attitude reference updating method and device provided by the present invention obtain an attitude reference within a preset time period corresponding to a predicted attitude detected based on a target image, obtain a deviation between the attitude reference corresponding to the moment prior to the predicted attitude based on the predicted attitude, and determine the attitude reference corresponding to the predicted attitude according to the comparison between the deviation and the maximum attitude reference and the minimum attitude reference within the preset time period, so as to dynamically adjust the attitude reference, improve the accuracy of subsequent abnormal behavior detection, avoid misdetection caused by changes in driving habits or drivers, and improve the detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 is a flowchart of the attitude reference updating method provided by the present invention;

[0021] Figure 2It is a schematic structural diagram of the attitude reference updating device provided by the present invention;

[0022] Figure 3 It is a schematic structural diagram of the electronic device provided by the present invention. Detailed implementation manners

[0023] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0024] Figure 1 It shows a schematic flowchart of a method for updating an attitude reference according to the present invention. The method includes:

[0025] S01, obtaining an attitude reference within a preset time period according to a predicted attitude obtained in advance based on a target image;

[0026] S02, obtaining a deviation between the attitude reference corresponding to the moment prior to and adjacent to the predicted attitude according to the attitude reference and the predicted attitude within the preset time period;

[0027] S03, comparing the deviation with the maximum attitude reference and the minimum reference in the attitude reference within the preset time period respectively, and determining the attitude reference corresponding to the moment of the predicted attitude based on the comparison result, the predicted attitude and the attitude reference within the preset time period.

[0028] It should be noted that S0N in this specification does not represent the sequence of the attitude reference updating method. The attitude reference updating method of the present invention will be specifically described below.

[0029] Step S01, obtaining an attitude reference within a preset time period according to a predicted attitude obtained in advance based on a target image.

[0030] In this embodiment, obtaining an attitude reference within a preset time period according to a predicted attitude detected in advance based on a target image includes: obtaining an attitude reference within a preset time period including the moment corresponding to the predicted attitude based on a sliding window with a preset length according to the predicted attitude obtained in advance based on the target image.

[0031] Furthermore, a pose reference within a preset time period is selected from historical pose references based on a sliding window of a preset length that includes the time period corresponding to the predicted pose as the pose reference within the preset time period. It should be noted that the preset length of the sliding window is the same as the time length of the preset time period. Additionally, assuming the time corresponding to the predicted pose is t, the preset time period can be [t - i, t + i], and the preset length of the corresponding sliding window is 2i; or the preset time period can be [t - 2i, t], which can be specifically set according to actual design requirements.

[0032] In an alternative embodiment, obtaining a pose reference within a preset time period based on a predicted pose obtained in advance from a target image includes: selecting a corresponding preset time period according to the time corresponding to the predicted pose obtained in advance from the target image, where the corresponding preset time period includes the time corresponding to the predicted pose; and selecting a corresponding pose reference from historical pose references according to the preset time period to obtain the pose reference within the preset time period.

[0033] In an alternative embodiment, before obtaining the pose reference within the preset time period, it further includes: obtaining a target image; performing pose detection on the target image to obtain a predicted pose. It should be noted that the target image can be a face image. By detecting the face pose, it is convenient to, after obtaining the corresponding pose reference, determine whether the predicted pose meets the standard based on the obtained pose reference, so as to determine whether there are abnormal behaviors of the corresponding driver, such as fatigue driving, distracted attention, smoking, playing with mobile phones, etc.

[0034] In a possible implementation manner, performing pose detection on the target image to obtain a predicted pose includes: inputting the obtained target image into a pose detection model to obtain the predicted pose output by the pose detection model, where the pose detection model is trained based on the target training image and its corresponding true pose angles.

[0035] In another possible implementation manner, performing pose detection on the target image to obtain a predicted pose includes: performing pose detection on the obtained target image; and labeling the detected pose angles to obtain the predicted pose.

[0036] In an alternative embodiment, when obtaining the target image, the target image to be subjected to pose detection can be obtained based on an electronic device or application platform applying a pose detection method; or the target image can be obtained based on a terminal device connected to the electronic device or application platform applying the pose detection method. It should be noted that the above terminal device or application platform can obtain the target image through a visual sensor connected thereto, and the visual sensor includes millimeter-wave radar, lidar, detector, camera, and other imaging devices, etc.

[0037] It should be noted that the obtained target image can be a single-frame picture or a picture frame sequence taken for different driver targets in different driving scenarios, or an image frame or an image frame sequence obtained by segmenting the shots of videos taken for different driver targets in different driving scenarios.

[0038] Step S02: Obtain the deviation between the attitude reference corresponding to the moment prior to the predicted attitude and adjacent to the predicted attitude, based on the attitude reference and the predicted attitude within a preset time period.

[0039] In this embodiment, obtaining the deviation between the attitude reference corresponding to the moment prior to the predicted attitude and adjacent to the predicted attitude, based on the attitude reference and the predicted attitude within a preset time period, includes: obtaining the attitude reference corresponding to the moment prior to the predicted attitude based on the attitude reference within the preset time period; and obtaining the deviation based on the predicted attitude and the attitude reference at the adjacent prior moment.

[0040] It should be noted that if the corresponding moment of the predicted attitude x is t, then first, according to the moment t corresponding to the predicted attitude x, obtain the moment prior to the predicted attitude within the preset time period, that is, the moment t - 1, and then obtain the attitude reference A(t - 1) corresponding to the moment t - 1; secondly, based on the predicted attitude x and the attitude reference A(t - 1) at the adjacent prior moment, obtain the deviation Δ = abs(x - A(t - 1)).

[0041] Step S03: Compare the deviation with the maximum attitude reference and the minimum reference in the attitude reference within the preset time period respectively, and based on the comparison results, determine the attitude reference corresponding to the moment of the predicted attitude based on the predicted attitude and the attitude reference within the preset time period.

[0042] In this embodiment, determining the attitude reference corresponding to the moment of the predicted attitude based on the comparison results, based on the predicted attitude and the attitude reference within the preset time period, includes: based on the deviation being greater than the minimum attitude reference within the preset time period and less than the maximum attitude reference within the preset time period, obtain the attitude reference corresponding to the moment of the predicted attitude according to the predicted attitude, the deviation, and the attitude reference within the preset time period.

[0043] It should be noted that before the deviation is greater than the minimum attitude reference within the preset time period and less than the maximum attitude reference within the preset time period, it also includes: obtaining the maximum attitude reference max and the minimum attitude reference min according to the attitude reference within the preset time period, so as to facilitate subsequent use of the maximum attitude reference max and the minimum attitude reference min to determine whether the deviation conforms to the value range formed by the maximum attitude reference max and the minimum attitude reference min.

[0044] Furthermore, according to the predicted pose, the deviation, and the pose reference within a preset time period, the pose reference at the moment corresponding to the predicted pose is obtained, including: obtaining the time length of the preset time period according to the preset time period; obtaining the pose reference at the moment corresponding to the predicted pose according to the time length, the predicted pose, the pose reference at the adjacent previous moment, and the deviation. Specifically:

[0045] First, according to the preset time period, the time length of the preset time period is obtained. It should be noted that when obtaining the pose reference within the preset time period by using a sliding window with a preset length, the time length of the preset time period can be determined according to the preset length of the sliding window.

[0046] Second, according to the time length, the predicted pose, the pose reference at the adjacent previous moment, and the deviation, the pose reference at the moment corresponding to the predicted pose is obtained. In this embodiment, the pose reference at the moment corresponding to the predicted pose means:

[0047]

[0048] wherein, A(t) represents the pose reference at the moment corresponding to the predicted pose, x represents the predicted pose, Δ represents the deviation, and M represents the time length.

[0049] In an alternative embodiment, according to the comparison result, based on the predicted pose and the pose reference within the preset time period, determining the pose reference at the moment corresponding to the predicted pose further includes: based on the deviation being less than or equal to the minimum pose reference within the preset time period, or based on the deviation being greater than or equal to the maximum pose reference within the preset time period, obtaining the pose reference corresponding to the adjacent previous moment of the predicted pose according to the pose reference within the preset time period, and using it as the pose reference at the moment corresponding to the predicted pose.

[0050] For example, when the deviation Δ ≤ the minimum pose reference min, or the deviation Δ ≥ the maximum pose reference max, select the pose reference corresponding to the adjacent previous moment of the predicted pose from the pose references within the preset time period, and use it as the pose reference at the moment corresponding to the predicted pose, that is, A(t) = A(t - 1), where A(t) represents the pose reference at the moment corresponding to the predicted pose, and A(t - 1) represents the pose reference corresponding to the adjacent previous moment of the predicted pose.

[0051] In summary, the embodiment of the present invention obtains the pose reference within the preset time period corresponding to the predicted pose obtained by detecting based on the target image, obtains the deviation between the predicted pose and the pose reference at the previous moment corresponding to the predicted pose, and determines the pose reference corresponding to the predicted pose according to the comparison between the deviation and the maximum and minimum pose references within the preset time period, thereby dynamically adjusting the pose reference to improve the accuracy of subsequent abnormal behavior detection, avoiding misdetection caused by changes in driving habits or drivers, and improving the detection accuracy.

[0052] The attitude reference update device provided by the present invention will be described below. The attitude reference update device described below can be correspondingly referred to the attitude reference update method described above.

[0053] Figure 2 A schematic structural diagram of an attitude reference update device is shown. The device includes:

[0054] A data acquisition module 21, which acquires an attitude reference within a preset time period according to a predicted attitude obtained in advance based on a target image.

[0055] A deviation acquisition module 22, which obtains a deviation between the attitude reference corresponding to the moment adjacent to the predicted attitude and the attitude reference within a preset time period according to the attitude reference and the predicted attitude within the preset time period.

[0056] A reference acquisition module 23, which compares the deviation with the maximum attitude reference and the minimum reference in the attitude reference within the preset time period respectively, and determines the attitude reference corresponding to the moment of the predicted attitude based on the comparison result, the predicted attitude and the attitude reference within the preset time period.

[0057] In this embodiment, the data acquisition module 21 includes: a data acquisition unit, which acquires an attitude reference within a preset time period containing the moment corresponding to the predicted attitude based on a predicted attitude obtained in advance based on a target image and based on a sliding window with a preset length.

[0058] Furthermore, when acquiring the attitude reference within the preset time period containing the moment corresponding to the predicted attitude, the data acquisition unit selects the attitude reference within the time period containing the moment corresponding to the predicted attitude from the historical attitude references based on a sliding window with a preset length as the attitude reference within the preset time period. It should be noted that the preset length of the sliding window is the same as the time length of the preset time period. In addition, assuming that the moment corresponding to the predicted attitude is t, the preset time period can be [t - i, t + i], and the preset length of the corresponding sliding window is 2i; or the preset time period can be [t - 2i, t], which can be specifically set according to actual design requirements.

[0059] In an alternative embodiment, the data acquisition unit includes: a time period selection sub-unit, which selects a corresponding preset time period according to the moment corresponding to the predicted attitude obtained in advance based on a target image, where the corresponding preset time period contains the moment corresponding to the predicted attitude; a data acquisition sub-unit, which selects the corresponding attitude reference from the historical attitude references according to the preset time period to obtain the attitude reference within the preset time period.

[0060] In an alternative embodiment, the data acquisition module 21 further includes: an image acquisition unit that acquires a target image; and a detection unit that performs pose detection on the target image to obtain a predicted pose. It should be noted that, in one possible implementation, the detection unit includes: a detection subunit that inputs the acquired target image into a pose detection model to obtain a predicted pose output by the pose detection model, where the pose detection model is trained based on target training images and their corresponding true pose angles.

[0061] In another possible implementation, the detection unit includes: a detection subunit that performs pose detection on the acquired target image; and an annotation subunit that annotates the detected pose angles to obtain a predicted pose.

[0062] In an alternative embodiment, when the image acquisition unit acquires a target image, the target image to be subjected to pose detection may be obtained based on an electronic device or an application platform that applies a pose detection method; or, the target image may be obtained based on a terminal device connected to the electronic device or the application platform that applies the pose detection method. It should be noted that the above terminal device or application platform may obtain the target image through a vision sensor connected thereto, and the vision sensor includes a millimeter-wave radar, a lidar, a detector, a camera, and other imaging devices.

[0063] The deviation acquisition module 22 includes: a prior pose reference update unit that obtains a pose reference at a prior moment adjacent to the predicted pose according to the pose reference within a preset time period; and a deviation acquisition unit that obtains a deviation based on the predicted pose and the pose reference at the adjacent prior moment.

[0064] It should be noted that if the corresponding moment of the predicted pose x is t, then first, according to the moment t corresponding to the predicted pose x, the prior moment adjacent to the predicted pose within the preset time period is obtained, that is, the moment t - 1, and then the pose reference A(t - 1) corresponding to the moment t - 1 is obtained; secondly, based on the predicted pose x and the pose reference A(t - 1) at the adjacent prior moment, the deviation Δ = abs(x - A(t - 1)) is obtained.

[0065] The reference acquisition module 23 includes: a comparison unit that respectively compares the deviation with the maximum pose reference and the minimum reference in the pose references within the preset time period; and a reference acquisition unit that determines the pose reference corresponding to the moment of the predicted pose based on the comparison result and the pose references within the preset time period.

[0066] Specifically, the reference acquisition unit includes: a reference acquisition subunit, which, based on the deviation being greater than the minimum attitude reference within a preset time period and less than the maximum attitude reference within the preset time period, obtains the attitude reference at the moment corresponding to the predicted attitude according to the predicted attitude, the deviation, and the attitude reference within the preset time period. Further, the reference acquisition subunit includes: a length acquisition grandchild unit, which obtains the time length of the preset time period according to the preset time period; a reference acquisition grandchild unit, which obtains the attitude reference at the moment corresponding to the predicted attitude according to the time length, the predicted attitude, the attitude reference at the adjacent previous moment, and the deviation.

[0067] In an alternative embodiment, the apparatus further includes: an extreme value acquisition module, which, before the deviation is greater than the minimum attitude reference within the preset time period and less than the maximum attitude reference within the preset time period, obtains the maximum attitude reference max and the minimum attitude reference min according to the attitude reference within the preset time period, so as to subsequently use the maximum attitude reference max and the minimum attitude reference min to determine whether the deviation conforms to the value range formed by the maximum attitude reference max and the minimum attitude reference min.

[0068] In an alternative embodiment, the reference acquisition subunit is further configured to, based on the deviation being less than or equal to the minimum attitude reference within the preset time period, or based on the deviation being greater than or equal to the maximum attitude reference within the preset time period, obtain the attitude reference corresponding to the adjacent previous moment of the predicted attitude according to the attitude reference within the preset time period, and use it as the attitude reference at the moment corresponding to the predicted attitude.

[0069] Figure 3 The schematic physical structure diagram of an electronic device is illustrated. As Figure 3 shown, the electronic device may include: a processor 31, a communication interface 32, a memory 33, and a communication bus 34. Among them, the processor 31, the communication interface 32, and the memory 33 complete mutual communication through the communication bus 34. The processor 31 may call the logical instructions in the memory 33 to execute the attitude reference update method, which includes: obtaining the attitude reference within a preset time period according to the predicted attitude obtained in advance based on the target image; obtaining the deviation between the attitude reference at the moment corresponding to the predicted attitude and the attitude reference at the adjacent previous moment according to the attitude reference within the preset time period and the predicted attitude; comparing the deviation with the maximum attitude reference and the minimum reference in the attitude reference within the preset time period respectively, and determining the attitude reference at the moment corresponding to the predicted attitude based on the comparison result, the predicted attitude, and the attitude reference within the preset time period.

[0070] In addition, when the logical instructions in the above-mentioned memory 33 can be implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0071] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the attitude reference update method provided by the above-mentioned various methods. The method includes: obtaining an attitude reference within a preset time period according to a predicted attitude obtained in advance based on a target image; obtaining a deviation between the attitude reference corresponding to the adjacent previous moment of the predicted attitude according to the attitude reference within the preset time period and the predicted attitude; comparing the deviation with the maximum attitude reference and the minimum reference in the attitude reference within the preset time period respectively, and determining the attitude reference corresponding to the moment of the predicted attitude based on the comparison result, the predicted attitude, and the attitude reference within the preset time period.

[0072] On another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it realizes the execution of the attitude reference update method provided by the above-mentioned various methods. The method includes: obtaining an attitude reference within a preset time period according to a predicted attitude obtained in advance based on a target image; obtaining a deviation between the attitude reference corresponding to the adjacent previous moment of the predicted attitude according to the attitude reference within the preset time period and the predicted attitude; comparing the deviation with the maximum attitude reference and the minimum reference in the attitude reference within the preset time period respectively, and determining the attitude reference corresponding to the moment of the predicted attitude based on the comparison result, the predicted attitude, and the attitude reference within the preset time period.

[0073] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0074] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for updating an attitude reference, characterized in that, it includes: Obtain an attitude reference within a preset time period according to a predicted attitude obtained in advance based on a target image; Obtain the deviation between the attitude reference corresponding to the moment prior to the predicted attitude adjacent to the predicted attitude according to the attitude reference within the preset time period and the predicted attitude; Compare the deviation with the maximum attitude reference and the minimum reference in the attitude reference within the preset time period respectively, and according to the comparison result, determine the attitude reference corresponding to the moment of the predicted attitude based on the predicted attitude and the attitude reference within the preset time period; The step of determining the attitude reference corresponding to the moment of the predicted attitude based on the predicted attitude and the attitude reference within the preset time period according to the comparison result includes: When the deviation is greater than the minimum attitude reference within the preset time period and less than the maximum attitude reference within the preset time period, obtain the attitude reference corresponding to the moment of the predicted attitude according to the predicted attitude, the deviation and the attitude reference within the preset time period; The attitude reference corresponding to the moment of the predicted attitude means: ; where \(A(t)\) represents the attitude reference at the moment corresponding to the predicted attitude, \(x\) represents the predicted attitude, represents the deviation, and \(M\) represents the time length; The step of determining the attitude reference corresponding to the moment of the predicted attitude based on the predicted attitude and the attitude reference within the preset time period according to the comparison result further includes: When the deviation is less than or equal to the minimum attitude reference within the preset time period, or when the deviation is greater than or equal to the maximum attitude reference within the preset time period, obtain the attitude reference corresponding to the moment prior to the predicted attitude adjacent to the predicted attitude according to the attitude reference within the preset time period, and use it as the attitude reference corresponding to the moment of the predicted attitude.

2. The method for updating an attitude reference according to claim 1, characterized in that, The step of obtaining the deviation between the attitude reference corresponding to the moment prior to the predicted attitude adjacent to the predicted attitude according to the attitude reference within the preset time period and the predicted attitude includes: Obtain the attitude reference corresponding to the moment prior to the predicted attitude adjacent to the predicted attitude according to the attitude reference within the preset time period; Obtain the deviation based on the predicted attitude and the attitude reference at the adjacent prior moment.

3. The method for updating an attitude reference according to claim 1, characterized in that, The step of obtaining an attitude reference within a preset time period according to a predicted attitude obtained in advance based on a target image includes: Based on the predicted attitude obtained in advance based on the target image, obtain the attitude reference within the preset time period including the moment corresponding to the predicted attitude by using a sliding window with a preset length.

4. An attitude reference updating device, characterized in that, it includes: A data acquisition module, which obtains an attitude reference within a preset time period according to a predicted attitude obtained in advance based on a target image; A deviation acquisition module, which obtains the deviation between the attitude reference corresponding to the moment prior to the predicted attitude adjacent to the predicted attitude according to the attitude reference within the preset time period and the predicted attitude; A reference acquisition module compares the deviation with the maximum attitude reference and the minimum reference in the attitude reference within the preset time period respectively, and determines the attitude reference at the moment corresponding to the predicted attitude based on the comparison result, the predicted attitude and the attitude reference within the preset time period; The reference acquisition module includes: A reference acquisition unit determines the attitude reference at the moment corresponding to the predicted attitude based on the comparison result, the predicted attitude and the attitude reference within the preset time period; The reference acquisition unit includes: A reference acquisition subunit, when the deviation is greater than the minimum attitude reference within the preset time period and less than the maximum attitude reference within the preset time period, obtains the attitude reference at the moment corresponding to the predicted attitude according to the predicted attitude, the deviation and the attitude reference within the preset time period; The attitude reference at the moment corresponding to the predicted attitude means: ; Among them, A(t) represents the attitude reference at the moment corresponding to the predicted attitude, x represents the predicted attitude, represents the deviation, and M represents the time length; The reference acquisition subunit is further configured to: When the deviation is less than or equal to the minimum attitude reference within the preset time period, or when the deviation is greater than or equal to the maximum attitude reference within the preset time period, obtain the attitude reference corresponding to the moment adjacent to the predicted attitude in the prior time according to the attitude reference within the preset time period, and use it as the attitude reference at the moment corresponding to the predicted attitude.

5. An electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, When the processor executes the program, the steps of the attitude reference update method according to any one of claims 1 to 3 are implemented.

6. A non-transitory computer-readable storage medium, on which a computer program is stored, wherein, When the computer program is executed by a processor, the steps of the attitude reference update method according to any one of claims 1 to 3 are implemented.

7. A computer program product, including a computer program, wherein, When the computer program is executed by a processor, the steps of the attitude reference update method according to any one of claims 1 to 3 are implemented.

Citation Information

Patent Citations

  • Posture prompting method and system

    CN105448053A

  • Image-based adjacent inter-frame attitude information calculation method and a related device

    CN109658511A