Pedestrian climbing behavior detection device and method and computer readable storage medium

Abstract

The invention discloses a pedestrian climbing behavior detection device and method and a computer readable storage medium, and relates to the technical field of intelligent traffic. The equipment comprises an acquisition module, a target detection module and a determination module which are connected with one another, the acquisition module is used for acquiring a video frame image acquired by the camera; the target detection module is used for inputting the video frame image into a target detection model to obtain a rectangular bounding box corresponding to each traffic target in the video frame image, and the traffic targets comprise pedestrians and two-wheeled vehicles; the determination module is used for calculating a first intersection-to-union ratio between the rectangular bounding box corresponding to the pedestrian and the rectangular bounding box corresponding to the two-wheeled vehicle, if the first intersection-to-union ratio is smaller than or equal to a preset intersection-to-union ratio threshold value, the posture of the pedestrian is recognized through a posture detection model, and when the posture of the pedestrian conforms to preset action characteristics, it is determined that the pedestrian has a climbing behavior, and the pedestrian climbs the two-wheeled vehicle. And the attitude detection model comprises a Yolov8-pose model. By adopting the equipment, real-time detection can be realized, the detection precision can be improved, and the flexibility is high.

Description

Technical Field

[0001] The present application relates to the field of intelligent transportation technology, and in particular to a pedestrian climbing behavior detection device, method and computer-readable storage medium. Background Art

[0002] Road traffic is closely related to people's daily lives, and safety issues cannot be ignored. At present, related technologies mainly detect pedestrian climbing behavior through target detection, but this method cannot capture action details, has low detection sensitivity and accuracy, and also requires a large number of training pictures of climbing behavior to be collected and marked according to different usage scenarios, which is time-consuming and labor-intensive. Summary of the invention

[0003] In view of the above-mentioned defects or deficiencies in the relevant technology, it is desired to provide a pedestrian climbing behavior detection device, method and computer-readable storage medium, which can not only reduce the dependence on a large number of labeled pictures, but also perform real-time detection while ensuring detection accuracy, and is more flexible.

[0004] In a first aspect, the present application provides a pedestrian climbing behavior detection device, the pedestrian climbing behavior detection device comprising an acquisition module, a target detection module and a determination module connected to each other;

[0005] The acquisition module is configured to acquire a video frame image captured by a camera; the target detection module is configured to input the video frame image into a target detection model to obtain a rectangular bounding box corresponding to each traffic target in the video frame image, wherein the traffic targets include pedestrians and two-wheeled vehicles;

[0006] The determination module is configured to calculate a first intersection-and-union ratio between a rectangular bounding box corresponding to the pedestrian and a rectangular bounding box corresponding to the two-wheeled vehicle. If the first intersection-and-union ratio is less than or equal to a preset intersection-and-union ratio threshold, the posture of the pedestrian is identified through a posture detection model, and when the posture of the pedestrian meets the preset action characteristics, it is determined that the pedestrian is in climbing behavior, and the posture detection model includes a Yolov8-pose model.

[0007] Optionally, in some embodiments of the present application, the determination module is specifically used to extract the key points of the pedestrian's body using the posture detection model, and the key points of the human body include the left hip, the right hip, the left knee, the right knee, the left foot and the right foot;

[0008] Constructing a first connection line from the left hip to the left knee, a second connection line from the left knee to the left foot, a third connection line from the right hip to the right knee, and a fourth connection line from the right knee to the right foot;

[0009] And, calculate the left knee angle between the first line and the second line and the right knee angle between the third line and the fourth line, and if at least one of the left knee angle and the right knee angle is greater than a preset angle threshold, it is determined that the pedestrian is climbing.

[0010] Optionally, in some embodiments of the present application, the preset angle threshold is 30 degrees to 90 degrees.

[0011] Optionally, the target detection model in some embodiments of the present application includes a YOLOv4-tiny model, a YOLOv5 model or a Yolov8 model.

[0012] Optionally, in some embodiments of the present application, the determination module is also specifically used to continuously acquire multiple video frame images within a preset time length range if the first intersection-and-union ratio is greater than the preset intersection-and-union ratio threshold, and sequentially calculate the second intersection-and-union ratios corresponding to the multiple video frame images; when the second intersection-and-union ratio is less than or equal to the preset intersection-and-union ratio threshold, identify the posture of the pedestrian through the posture detection model.

[0013] Optionally, in some embodiments of the present application, the preset intersection-over-union ratio threshold is 0.5 to 0.9.

[0014] Optionally, in some embodiments of the present application, the pedestrian climbing behavior detection device further includes an alarm module, and the alarm module is configured to issue an alarm prompt when it is determined that the pedestrian has climbing behavior.

[0015] In a second aspect, the present application provides a pedestrian climbing behavior detection method, the pedestrian climbing behavior detection method is used for the pedestrian climbing behavior detection device described in any one of the first aspects, the pedestrian climbing behavior detection method comprises:

[0016] Get the video frame image captured by the camera;

[0017] Inputting the video frame image into a target detection model to obtain a rectangular bounding box corresponding to each traffic target in the video frame image, wherein the traffic targets include pedestrians and two-wheeled vehicles;

[0018] A first intersection-and-union ratio between a rectangular bounding box corresponding to the pedestrian and a rectangular bounding box corresponding to the two-wheeled vehicle is calculated. If the first intersection-and-union ratio is less than or equal to a preset intersection-and-union ratio threshold, the posture of the pedestrian is identified through a posture detection model. When the posture of the pedestrian meets the preset action characteristics, it is determined that the pedestrian is climbing. The posture detection model includes a Yolov8-pose model.

[0019] Optionally, in some embodiments of the present application, the posture of the pedestrian is identified by a posture detection model, and when the posture of the pedestrian meets a preset action feature, it is determined that the pedestrian has a climbing behavior, including:

[0020] Extracting key points of the pedestrian's body using the posture detection model, wherein the key points of the pedestrian's body include a left hip, a right hip, a left knee, a right knee, a left foot, and a right foot;

[0021] Constructing a first connection line from the left hip to the left knee, a second connection line from the left knee to the left foot, a third connection line from the right hip to the right knee, and a fourth connection line from the right knee to the right foot;

[0022] The left knee angle between the first line and the second line and the right knee angle between the third line and the fourth line are calculated. If at least one of the left knee angle and the right knee angle is greater than a preset angle threshold, it is determined that the pedestrian is climbing.

[0023] In a third aspect, the present application provides a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the pedestrian climbing behavior detection method described in any one of the second aspects.

[0024] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0025] The embodiments of the present application provide a pedestrian climbing behavior detection device, method and computer-readable storage medium, which use a target detection model to obtain a rectangular bounding box corresponding to each traffic target in a video frame image, where the traffic targets include pedestrians and two-wheeled vehicles, and then calculate a first intersection-and-union ratio between the rectangular bounding box corresponding to the pedestrian and the rectangular bounding box corresponding to the two-wheeled vehicle. If the first intersection-and-union ratio is less than or equal to a preset intersection-and-union ratio threshold, the posture detection model is directly used to identify the posture of the pedestrian, which can not only capture the action details in real time and accurately and improve the detection sensitivity and accuracy, but also eliminate the need to collect and mark a large number of training pictures of climbing behavior, greatly improving processing efficiency and flexibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 A structural block diagram of a pedestrian climbing behavior detection device provided in an embodiment of the present application;

[0028] Figure 2 A schematic diagram of the distribution of key points of a human body provided in an embodiment of the present application;

[0029] Figure 3 A structural block diagram of another pedestrian climbing behavior detection device provided in an embodiment of the present application;

[0030] Figure 4 A flowchart of a method for detecting climbing behavior of a pedestrian provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0032] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0033] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. Figures 1 to 4 The pedestrian climbing behavior detection device, method and computer-readable storage medium provided in the embodiments of the present application are described in detail.

[0034] Please refer to Figure 1, which is a structural block diagram of a pedestrian climbing behavior detection device provided in an embodiment of the present application. The pedestrian climbing behavior detection device 100 includes an acquisition module 101, a target detection module 102 and a determination module 103 connected to each other, wherein the acquisition module 101 can acquire a video frame image collected by a camera, and the target detection module 102 can input the video frame image into a target detection model to obtain a rectangular bounding box corresponding to each traffic target in the video frame image, wherein the traffic target includes pedestrians and two-wheeled vehicles, and the two-wheeled vehicles can be bicycles, electric vehicles and motorcycles, etc. For example, the target detection model includes but is not limited to the YOLOv4-tiny model, the YOLOv5 model and the Yolov8 model, etc. The YOLOv4-tiny model and the YOLOv5 model are more lightweight, and the Yolov8 model is a series of pre-trained data models trained by the manufacturer based on the COCO data set, which can detect more than 80 common objects including pedestrians and various types of motor vehicles in real time and accurately, significantly reducing the development cost. In addition, the target detection model of the embodiment of the present application can also be a Faster R-CNN model, an SSD model and a CenterNet model, etc., which improves the adaptability of the system.

[0035] Furthermore, the determination module 103 is capable of calculating the first intersection over union (IOU) between the rectangular bounding box corresponding to the pedestrian and the rectangular bounding box corresponding to the two-wheeled vehicle. If the first IOU is less than or equal to a preset IOU threshold, the posture of the pedestrian is identified through the posture detection model, and when the posture of the pedestrian meets the preset action characteristics, it is determined that the pedestrian is climbing. For example, the preset IOU threshold can be 0.5 to 0.9. For another example, the posture detection model includes the Yolov8-pose model, which is a series of pre-trained data models trained by the manufacturer based on the COCO data set, which can efficiently and accurately extract Figure 2 All 17 key points of the human body are shown, and there is no need to collect a large number of climbing-related image materials for annotation, which saves labor and time costs. In addition, based on the single-stage structure of YOLO, it can integrate target detection and posture estimation together and perform them synchronously. It can detect climbing behavior in real time in complex scenes, improve processing efficiency, and the calculation time is about 20ms to 50ms. In addition, the posture detection model of the embodiment of the present application can also be HRNet model, DeepLabCut model and MMPose model, etc. It should be noted that Figure 2Among them, a represents nose, b1 represents left eye, b2 represents right eye, c1 represents left ear, c2 represents right ear, d1 represents left shoulder, d2 represents right shoulder, e1 represents left elbow, e2 represents right elbow, f1 represents left wrist, f2 represents right wrist, g1 represents left hip, g2 represents right hip, h1 represents left knee, h2 represents right knee, i1 represents left foot, and i2 represents right foot. In addition, according to the comparison result of the first intersection-over-union ratio with the preset intersection-over-union ratio threshold, pedestrians riding bicycles can be quickly excluded, unnecessary posture analysis calculations can be reduced, and the processing flow can be optimized.

[0036] Furthermore, Figure 2 As shown in the example, the determination module 103 is specifically used to first use the posture detection model to extract the pedestrian's human body key points, which include the left hip g1, the right hip g2, the left knee h1, the right knee h2, the left foot i1 and the right foot i2; then construct a first connection line from the left hip g1 to the left knee h1, a second connection line from the left knee h1 to the left foot i1, a third connection line from the right hip g2 to the right knee h2 and a fourth connection line from the right knee h2 to the right foot i2, and then calculate the left knee angle α between the first connection line and the second connection line and the right knee angle β between the third connection line and the fourth connection line. If at least one of the left knee angle α and the right knee angle β is greater than a preset angle threshold, it can be quickly determined that the pedestrian has climbing behavior. For example, the preset angle threshold can be 30 degrees to 90 degrees.

[0037] Further, in the embodiment of the present application, the determination module 103 is also specifically used to continuously obtain multiple video frame images within a preset time range if the first intersection-and-union ratio is greater than a preset intersection-and-union ratio threshold, and sequentially calculate the second intersection-and-union ratio corresponding to the multiple video frame images. When the second intersection-and-union ratio is less than or equal to the preset intersection-and-union ratio threshold, the posture detection model is used to identify the posture of the pedestrian. For example, the preset time range can be 2s to 5s. The advantage of this setting is that even if the pedestrian is occasionally very close to the two-wheeled vehicle, the pedestrian will not move, and the two-wheeled vehicle in motion is moving. This method can distinguish the two and reduce misjudgment. In addition, some embodiments of the present application can also determine whether the pedestrian is riding a bicycle based on the geometric position of the key points of the human body (such as feet or buttocks) relative to the two-wheeled vehicle, which helps to improve the judgment accuracy.

[0038] Furthermore, if Figure 3 As shown, in the embodiment of the present application, the pedestrian climbing behavior detection device 100 also includes an alarm module 104. The alarm module 104 can issue an alarm prompt when it is determined that the pedestrian has climbing behavior, thereby effectively regulating the pedestrian's bad behavior and ensuring road traffic safety.

[0039] The pedestrian climbing behavior detection device provided in the embodiment of the present application obtains the rectangular bounding box corresponding to each traffic target in the video frame image through the target detection model, and the traffic targets include pedestrians and two-wheeled vehicles, and then calculates the first intersection-and-union ratio between the rectangular bounding box corresponding to the pedestrian and the rectangular bounding box corresponding to the two-wheeled vehicle. If the first intersection-and-union ratio is less than or equal to the preset intersection-and-union ratio threshold, the posture detection model is directly used to identify the posture of the pedestrian, which can not only capture the action details in real time and accurately, improve the detection sensitivity and accuracy, but also eliminate the need to collect and mark a large number of training pictures of climbing behavior, greatly improving the processing efficiency and flexibility.

[0040] Based on the above embodiments, the present application provides a method for detecting climbing behavior of pedestrians. The method can be used to detect climbing behavior of pedestrians. Figures 1 to 3 The pedestrian climbing behavior detection device 100 of the corresponding embodiment. Please refer to Figure 4 , which is a flow chart of a pedestrian climbing behavior detection method provided in an embodiment of the present application, the detection method specifically comprises the following steps:

[0041] S101, obtaining a video frame image captured by a camera.

[0042] S102, inputting the video frame image into the target detection model to obtain a rectangular bounding box corresponding to each traffic target in the video frame image, where the traffic targets include pedestrians and two-wheeled vehicles.

[0043] Exemplarily, the target detection models in the embodiments of the present application include but are not limited to the YOLOv4-tiny model, the YOLOv5 model, the Yolov8 model, the Faster R-CNN model, the SSD model and the CenterNet model, and the two-wheeled vehicles include but are not limited to bicycles, electric vehicles and motorcycles.

[0044] S103, calculating a first intersection-and-union ratio between a rectangular bounding box corresponding to the pedestrian and a rectangular bounding box corresponding to the two-wheeled vehicle. If the first intersection-and-union ratio is less than or equal to a preset intersection-and-union ratio threshold, the posture of the pedestrian is identified through a posture detection model, and when the posture of the pedestrian meets the preset action characteristics, it is determined that the pedestrian is in climbing behavior. The posture detection model includes a Yolov8-pose model.

[0045] Exemplarily, for example, in the process of determining whether a pedestrian has climbing behavior, the embodiment of the present application first uses a posture detection model to extract the pedestrian's body key points, which include the left hip g1, the right hip g2, the left knee h1, the right knee h2, the left foot i1 and the right foot i2; then construct a first connecting line from the left hip g1 to the left knee h1, a second connecting line from the left knee h1 to the left foot i1, a third connecting line from the right hip g2 to the right knee h2 and a fourth connecting line from the right knee h2 to the right foot i2, and then calculate the left knee angle α between the first connecting line and the second connecting line and the right knee angle β between the third connecting line and the fourth connecting line. If at least one of the left knee angle α and the right knee angle β is greater than a preset angle threshold, it can be quickly determined that the pedestrian has climbing behavior. For example, the preset angle threshold can be 30 degrees to 90 degrees.

[0046] If the first IoU is greater than the preset IoU threshold, multiple video frame images within a preset time range are continuously acquired, and the second IoU corresponding to the multiple video frame images are calculated in turn. When the second IoU is less than or equal to the preset IoU threshold, the posture detection model is used to identify the posture of the pedestrian. For example, the preset time range can be 2s to 5s, and the preset IoU threshold can be 0.5 to 0.9. The advantage of this setting is that even if the pedestrian is occasionally very close to the two-wheeled vehicle, the pedestrian will not move, while the two-wheeled vehicle is moving. Processing in this way can distinguish the two and reduce misjudgment.

[0047] It should be noted that, for the description of the same steps and the same contents in this embodiment as those in other embodiments, reference can be made to the description in other embodiments and will not be repeated here.

[0048] The pedestrian climbing behavior detection method provided in the embodiment of the present application obtains the rectangular bounding box corresponding to each traffic target in the video frame image through the target detection model, and the traffic targets include pedestrians and two-wheeled vehicles. Then, the first intersection-and-union ratio between the rectangular bounding box corresponding to the pedestrian and the rectangular bounding box corresponding to the two-wheeled vehicle is calculated. If the first intersection-and-union ratio is less than or equal to the preset intersection-and-union ratio threshold, the posture detection model is directly used to identify the posture of the pedestrian. This can not only capture the action details in real time and accurately, improve the detection sensitivity and accuracy, but also eliminate the need to collect and mark a large number of training pictures of climbing behavior, greatly improving the processing efficiency and flexibility.

[0049] As another aspect, an embodiment of the present application provides a computer-readable storage medium for storing program code, the program code for executing Figure 4 Any implementation manner of the pedestrian climbing behavior detection method in the corresponding embodiment.

[0050] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0051] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms. The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0052] In addition, each functional module in each embodiment of the present application may be integrated into a processing unit, or each module may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of a software functional unit. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium.

[0053] Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the whole or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the pedestrian climbing behavior detection method in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0054] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0055] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A pedestrian climbing behavior detection device, characterized in that: The pedestrian climbing behavior detection device includes an acquisition module, a target detection module and a determination module which are connected to each other; The acquisition module is configured to acquire a video frame image captured by a camera; the target detection module is configured to input the video frame image into a target detection model to obtain a rectangular bounding box corresponding to each traffic target in the video frame image, wherein the traffic targets include pedestrians and two-wheeled vehicles; The determination module is configured to calculate a first intersection-and-union ratio between a rectangular bounding box corresponding to the pedestrian and a rectangular bounding box corresponding to the two-wheeled vehicle. If the first intersection-and-union ratio is less than or equal to a preset intersection-and-union ratio threshold, the posture of the pedestrian is identified through a posture detection model, and when the posture of the pedestrian meets the preset action characteristics, it is determined that the pedestrian is in climbing behavior, and the posture detection model includes a Yolov8-pose model.

2. The pedestrian climbing behavior detection device according to claim 1, characterized in that: The determination module is specifically used to extract the key points of the pedestrian's body using the posture detection model, and the key points of the pedestrian's body include the left hip, the right hip, the left knee, the right knee, the left foot and the right foot; Constructing a first connection line from the left hip to the left knee, a second connection line from the left knee to the left foot, a third connection line from the right hip to the right knee, and a fourth connection line from the right knee to the right foot; And, calculate the left knee angle between the first line and the second line and the right knee angle between the third line and the fourth line, and if at least one of the left knee angle and the right knee angle is greater than a preset angle threshold, it is determined that the pedestrian is climbing.

3. The pedestrian climbing behavior detection device according to claim 2, characterized in that: The preset angle threshold is 30 degrees to 90 degrees.

4. The pedestrian climbing behavior detection device according to any one of claims 1 to 3, characterized in that: The target detection model includes a YOLOv4-tiny model, a YOLOv5 model or a Yolov8 model.

5. The pedestrian climbing behavior detection device according to claim 4, characterized in that: The determination module is also specifically used to continuously acquire multiple video frame images within a preset time range if the first intersection-and-union ratio is greater than the preset intersection-and-union ratio threshold, and to sequentially calculate the second intersection-and-union ratios corresponding to the multiple video frame images; when the second intersection-and-union ratio is less than or equal to the preset intersection-and-union ratio threshold, identify the posture of the pedestrian through the posture detection model.

6. The pedestrian climbing behavior detection device according to claim 5, characterized in that: The preset intersection-over-union ratio threshold is 0.5 to 0.

9.

7. The pedestrian climbing behavior detection device according to claim 4, characterized in that: The pedestrian climbing behavior detection device also includes an alarm module, which is configured to issue an alarm prompt when it is determined that the pedestrian has climbing behavior.

8. A method for detecting pedestrian climbing behavior, characterized in that: The pedestrian climbing behavior detection method is used for the pedestrian climbing behavior detection device according to any one of claims 1 to 7, and the pedestrian climbing behavior detection method comprises: Get the video frame image captured by the camera; Inputting the video frame image into a target detection model to obtain a rectangular bounding box corresponding to each traffic target in the video frame image, wherein the traffic targets include pedestrians and two-wheeled vehicles; A first intersection-and-union ratio between a rectangular bounding box corresponding to the pedestrian and a rectangular bounding box corresponding to the two-wheeled vehicle is calculated. If the first intersection-and-union ratio is less than or equal to a preset intersection-and-union ratio threshold, the posture of the pedestrian is identified through a posture detection model. When the posture of the pedestrian meets the preset action characteristics, it is determined that the pedestrian is climbing. The posture detection model includes a Yolov8-pose model.

9. The method for detecting pedestrian climbing behavior according to claim 8, characterized in that: The step of identifying the posture of the pedestrian by using a posture detection model and determining that the pedestrian has a climbing behavior when the posture of the pedestrian meets a preset action feature includes: Extracting key points of the pedestrian's body using the posture detection model, wherein the key points of the pedestrian's body include a left hip, a right hip, a left knee, a right knee, a left foot, and a right foot; Constructing a first connection line from the left hip to the left knee, a second connection line from the left knee to the left foot, a third connection line from the right hip to the right knee, and a fourth connection line from the right knee to the right foot; The left knee angle between the first line and the second line and the right knee angle between the third line and the fourth line are calculated. If at least one of the left knee angle and the right knee angle is greater than a preset angle threshold, it is determined that the pedestrian is climbing.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the pedestrian climbing behavior detection method described in any one of claims 8 to 9.

Images