Panic pedestrian wandering behavior identification method and system based on motion trail

Abstract

The invention relates to a panic pedestrian wandering behavior recognition method and system based on a motion trail, and the method comprises the following steps: obtaining a motion image of a pedestrian, recognizing the motion mass center of the pedestrian based on the image, and constructing the motion trail of the pedestrian; according to the pedestrian wandering behavior criterion, performing wandering behavior identification on the motion trail of the pedestrian to obtain an identification result; wherein the establishment of the pedestrian wandering behavior criterion comprises the following steps: drawing possible wandering motion trails in different scenes by referring to common wandering behaviors in public places; and analyzing the wandering motion trails of the pedestrians to obtain a pedestrian wandering behavior criterion. Compared with the prior art, different wandering motion modes in different scenes are considered, and pertinence of pedestrian wandering behavior recognition is improved.

Description

Technical Field

[0001] The present invention relates to the field of crowd stability analysis, and in particular to a method and system for identifying wandering behavior of panicked pedestrians based on motion trajectories. Background Art

[0002] With the increase in the coverage of network surveillance cameras in public places, the richness of video and image data, and the full development and widespread application of computer vision human behavior recognition technology. Domestic and foreign scholars have begun to study crowd evacuation and emergencies in the early 20th century. In various public place emergencies, pedestrians are easily affected by the environment and fall into panic, which in turn triggers various panic behaviors.

[0003] There are still some deficiencies in the current research: 1) At present, the research on pedestrian wandering motion trajectory does not consider different wandering motion modes in different scenarios. For example, patent application CN 114662521 A discloses a pedestrian wandering behavior detection method and system, including: obtaining a video compression code stream of the detection area, extracting multiple frames of image data contained in the video compression code stream; obtaining a pre-trained neural network model, inputting the image data frame by frame into the neural network model to generate a coordinate set composed of multiple target detection frame coordinates for characterizing the target movement path; based on the coordinate set, performing a target back-off detection in a single direction, if the back-off number exceeds a threshold, outputting a detection result that the target has wandering behavior. This method does not consider different wandering motion modes in different scenarios, and lacks a more targeted wandering behavior criterion based on motion trajectory. 2) Most of the existing research on panic crowd disturbance propagation based on computer vision is based on crowd flow or density information, and few of them extract dynamic characteristics such as motion trajectory, speed or acceleration from video image data for disturbance propagation analysis. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a method and system for identifying the wandering behavior of panicked pedestrians based on motion trajectories, thereby achieving accurate identification of the wandering behavior of pedestrians.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A method for identifying panic pedestrian wandering behavior based on motion trajectory includes the following steps:

[0007] Obtain the motion image of the pedestrian, identify the motion center of the pedestrian based on the image, and construct the pedestrian motion trajectory;

[0008] According to the pedestrian wandering behavior criterion, the pedestrian's movement trajectory is identified to obtain the identification result;

[0009] The steps for establishing the pedestrian wandering behavior criterion are as follows:

[0010] Refer to the common wandering behaviors in public places and draw the possible wandering movement trajectories in different scenarios;

[0011] The wandering trajectory of pedestrians is analyzed to obtain the criteria for pedestrian wandering behavior.

[0012] Furthermore, the steps of constructing the pedestrian motion trajectory are as follows:

[0013] Based on the motion image of pedestrians, key nodes of the human body are obtained to form different body segments, and the center of mass position of each body segment is calculated;

[0014] The center of mass of each body segment of the pedestrian is used to calculate the total center of mass of the human body;

[0015] After fitting the coordinates of the pedestrian's center of mass according to the time sequence, the movement trajectory of the pedestrian's center of mass over a period of time can be obtained.

[0016] Furthermore, the key nodes include: top of the head, left ear, right ear, left eye, right eye, nose, left corner of the mouth, right corner of the mouth, neck, left shoulder, right shoulder, left elbow, right elbow, left wrist, right wrist, left hip, right hip, left knee, right knee, left ankle and right ankle.

[0017] Furthermore, the calculation formula for the centroid position of the body segment is as follows:

[0018] x k =x p l p +x d l d

[0019] y k =y p l p +y d l d

[0020] z k =z p l p +z d l d

[0021] In the formula, (x k ,y k 、z k ) is the coordinate of the segment centroid, (x p ,y p 、z p ) is the coordinate of the proximal end of the segment, (x d ,y d 、z d) is the coordinate of the distal end of the segment, l p is the percentage of segment length from proximal to distal, l d It is the percentage of segment length from distal end to tip.

[0022] Furthermore, the calculation formula of the total center of mass position of the human body is as follows:

[0023]

[0024] In the formula, (x c ,y c 、z c ) is the coordinate of the center of mass of the human body; (x k ,y k 、z k ) is the coordinate of the kth segment; m k is the mass of the kth segment; M is the total mass of the body segments.

[0025] Furthermore, the wandering motion trajectory includes multiple elliptical motion trajectories, spiral motion trajectories, serpentine motion trajectories and Brownian motion trajectories.

[0026] Furthermore, the pedestrian wandering behavior criterion is as follows:

[0027]

[0028]

[0029] Where Oval.p. is the criterion for elliptical motion trajectory, Spiral.p. is the criterion for spiral motion trajectory, Snake.p. is the criterion for snake motion trajectory, Brown.p. is the criterion for Brownian motion trajectory, and N count is the threshold number of turning points in the pedestrian motion trajectory.

[0030] Furthermore, when the pedestrian moves back and forth, an inflection point appears in the pedestrian's motion trajectory, and whether the pedestrian's behavior is wandering behavior is determined based on whether the number of occurrences of the inflection point exceeds a threshold.

[0031] Furthermore, the inflection point is the point where the direction angle of the trajectory point of consecutive frames in the pedestrian motion trajectory changes by more than 90 degrees, and the calculation formula is as follows:

[0032]

[0033] Where θ is the direction angle of the trajectory point, P t-1 is the pedestrian trajectory coordinate point at the t-1th moment, P t is the pedestrian trajectory coordinate point at the tth moment, P t+1 is the pedestrian trajectory coordinate point at the t+1th moment.

[0034] According to two aspects of the present invention, a panic pedestrian wandering behavior recognition system based on motion trajectory is provided, comprising:

[0035] The pedestrian motion trajectory construction module is used to obtain the motion image of the pedestrian, identify the motion center of the pedestrian based on the image, and construct the pedestrian motion trajectory;

[0036] A pedestrian wandering behavior recognition module is used to recognize the wandering behavior of pedestrians based on their movement trajectories according to the pedestrian wandering behavior criterion to obtain a recognition result;

[0037] The steps for establishing the pedestrian wandering behavior criterion are as follows:

[0038] Refer to the common wandering behaviors in public places and draw the possible wandering movement trajectories in different scenarios;

[0039] The wandering trajectory of pedestrians is analyzed to obtain the criteria for pedestrian wandering behavior.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] 1. The present invention analyzes the characteristics of wandering motion trajectories according to different wandering motion modes that pedestrians may have in different environments, and proposes targeted wandering behavior recognition criteria according to each motion mode, thereby improving the targeted identification of pedestrian wandering behavior.

[0042] 2. The present invention uses computer vision technology to extract the object's motion center of mass and motion trajectory information from video or image information, and uses these acquired data in wandering behavior recognition and crowd stability analysis research, thereby improving the accuracy of pedestrian wandering behavior recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A schematic diagram of a process of identifying the wandering behavior of panicked pedestrians based on motion trajectories proposed by the present invention;

[0044] Figure 2 Schematic diagrams of possible wandering trajectories in different scenarios, where (2a) is a multiple elliptical motion trajectory, (2b) is a spiral motion trajectory, (2c) is a serpentine motion trajectory, and (2d) is a Brownian motion trajectory;

[0045] Figure 3 It is a schematic diagram of the turning point formed in the pedestrian movement trajectory;

[0046] Figure 4 Schematic diagram of the experimental scene in the embodiment, wherein (4a) is a physical schematic diagram of the experimental scene, and (4a) is a virtual schematic diagram of the experimental scene;

[0047] Figure 5 Schematic diagram of the motion trajectory of the center of mass of a pedestrian with wandering behavior. DETAILED DESCRIPTION

[0048] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0049] Example 1

[0050] This embodiment provides a method for identifying panic pedestrian wandering behavior based on motion trajectory. Figure 1 As shown, the following steps are included:

[0051] S1. Obtain the motion image of the pedestrian, identify the motion center of the pedestrian based on the image, and construct the pedestrian's motion trajectory.

[0052] In this embodiment, a video of a pedestrian falling down is shot and observed in the lobby on the first floor of the office building of Tongji University Zhixin Hall as an experimental scene. Figure 4 The camera is located on the right side of the gate, with a height of 1.6 meters. The location and site information are as follows: Figure 4 as shown in (4b).

[0053] The steps for constructing pedestrian motion trajectories are as follows:

[0054] Based on the motion image of the pedestrian, key nodes of the human body are obtained to form different body segments, and the center of mass position of each body segment is calculated. In this embodiment, 21 key nodes are used to form 13 segments, and the nodes are: top of the head, left ear, right ear, left eye, right eye, nose, left corner of the mouth, right corner of the mouth, neck, left shoulder, right shoulder, left elbow, right elbow, left wrist, right wrist, left hip, right hip, left knee, right knee, left ankle, and right ankle.

[0055] The calculation formula for defining the center of mass position of a body segment is as follows:

[0056] x k =x p l p +x d l d

[0057] y k =y p l p +y d l d

[0058] z k =z p l p +z d ld

[0059] In the formula, (x k ,y k 、z k ) is the coordinate of the segment centroid, (x p ,y p 、z p ) is the coordinate of the proximal end of the segment, (x d ,y d 、z d ) is the coordinate of the distal end of the segment, l p is the percentage of segment length from proximal to distal, l d It is the percentage of segment length from distal end to tip.

[0060] The centroid position of each body segment of the pedestrian is used to calculate the total centroid position of the human body. The calculation formula is as follows:

[0061]

[0062] In the formula, (x c ,y c 、z c ) is the coordinate of the center of mass of the human body; (x k ,y k 、z k ) is the coordinate of the kth segment; m k is the mass of the kth segment; M is the total mass of the body segments.

[0063] After fitting the coordinates of the pedestrian's centroid according to the time sequence, the motion trajectory of the pedestrian's centroid within a period of time can be obtained. Figure 4 The schematic diagram of the video information is collected, the video acquisition frequency f = 29.97fps (Frames Per Second), the video single frame resolution is 1920*1080 pixels, and the video frame image is intercepted using matlab, and the interception frequency FrameRate = 30. The upper left corner of the video frame image is defined as the coordinate origin, the horizontal axis is the x-axis, the horizontal axis value represents 1920 pixels, and the vertical axis is the y-axis, and the vertical axis value represents 1080 pixels; then identify the key points of the pedestrian's body, and obtain the centroid trajectory of the pedestrian's wandering behavior as shown in Figure 5 As shown in FIG. 1 , the pedestrian falling process lasts for 10 seconds, and a total of 30 frames of images are extracted.

[0064] S2. Identify the wandering behavior of the pedestrian based on the pedestrian's movement trajectory according to the pedestrian wandering behavior criterion to obtain a recognition result.

[0065] The steps to establish the pedestrian wandering behavior criterion are as follows:

[0066] Refer to the common wandering behaviors in public places and draw the dynamic trajectory diagrams of wandering that may appear in different scenarios. Figure 2 As shown, it includes four wandering trajectories and multiple elliptical motion trajectories as shown in Figure 2 As shown in (2a), the spiral motion trajectory is as follows Figure 2 As shown in (2b), the serpentine motion trajectory is Figure 2 As shown in (2c), the Brownian motion trajectory is Figure 2 As shown in (2d)

[0067] The wandering process of pedestrians in public places is often affected by scene factors. There are often a large number of obstacles in the scene or the pedestrians have a small range of activities. Therefore, the range of pedestrian wandering is small and the trajectory will be reciprocating. The wandering trajectory of pedestrians is analyzed to determine whether the pedestrian behavior is wandering. When the target movement is reciprocating, an inflection point will be formed, such as Figure 3 If the direction angle of the trajectory point of the continuous frames of the pedestrian's motion trajectory during the wandering process exceeds 90 degrees, the point is determined to be an inflection point, and the calculation formula is as follows:

[0068]

[0069] Where θ is the direction angle of the trajectory point, P t-1 is the pedestrian trajectory coordinate point at the t-1th moment, P t is the pedestrian trajectory coordinate point at the tth moment, P t+1 is the pedestrian trajectory coordinate point at the t+1th moment.

[0070] By counting the number of turning points in the pedestrian's motion trajectory, if it exceeds the threshold N count , the motion trajectory can be judged as wandering behavior. The pedestrian wandering behavior judgment criteria of four wandering trajectories are shown in Case 1 to Case 4.

[0071] Case 1 (elliptical motion trajectory):

[0072]

[0073] Case 2 (Spiral motion trajectory):

[0074]

[0075] Case 3 (snake-like motion trajectory):

[0076]

[0077] Case 4 (Brownian motion trajectory):

[0078]

[0079] The centroid trajectory can be used to visually see that a change of direction has occurred. According to the wandering trajectory criterion, the trajectory change frame moment satisfies cosθ<0. And since there are two trajectory change moments, namely N count ≥2, it can be concluded that pedestrian wandering behavior occurs in the video case of this embodiment.

[0080] In order to facilitate the visualization of the motion trajectory recognition model, the character η is used to represent the pedestrian motion trajectory recognition result. When the pedestrian motion trajectory is judged to be a wandering trajectory, η = 1; otherwise, η = 0. In summary, the recognition result of this embodiment is η = 1, that is, there is wandering behavior.

[0081] Example 2

[0082] This embodiment provides a panic pedestrian wandering behavior recognition system based on motion trajectory, including:

[0083] The pedestrian motion trajectory construction module is used to obtain the motion image of the pedestrian, identify the motion center of the pedestrian based on the image, and construct the pedestrian motion trajectory;

[0084] A pedestrian wandering behavior recognition module is used to recognize the wandering behavior of pedestrians based on their movement trajectories according to the pedestrian wandering behavior criterion to obtain a recognition result;

[0085] The steps for establishing the pedestrian wandering behavior criterion are as follows:

[0086] Refer to the common wandering behaviors in public places and draw the possible wandering movement trajectories in different scenarios;

[0087] The wandering trajectory of pedestrians is analyzed to obtain the criteria for pedestrian wandering behavior.

[0088] The rest is the same as in Example 1.

[0089] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for identifying panic pedestrian wandering behavior based on motion trajectory, characterized in that: The following steps are involved: Obtain the motion image of the pedestrian, identify the motion center of the pedestrian based on the image, and construct the pedestrian motion trajectory; According to the pedestrian wandering behavior criterion, the pedestrian's movement trajectory is identified to obtain the identification result; The steps for establishing the pedestrian wandering behavior criterion are as follows: Refer to the common wandering behaviors in public places and draw the possible wandering movement trajectories in different scenarios; The wandering trajectory of pedestrians is analyzed to obtain the criteria for pedestrian wandering behavior.

2. The method for identifying panicked pedestrian wandering behavior based on motion trajectory according to claim 1 is characterized in that: The steps for constructing the pedestrian motion trajectory are as follows: Based on the motion image of pedestrians, key nodes of the human body are obtained to form different body segments, and the center of mass position of each body segment is calculated; The center of mass of each body segment of the pedestrian is used to calculate the total center of mass of the human body; After fitting the coordinates of the pedestrian's center of mass according to the time sequence, the movement trajectory of the pedestrian's center of mass over a period of time can be obtained.

3. The method for identifying panicked pedestrian wandering behavior based on motion trajectory according to claim 2 is characterized in that: The key nodes include: top of the head, left ear, right ear, left eye, right eye, nose, left corner of the mouth, right corner of the mouth, neck, left shoulder, right shoulder, left elbow, right elbow, left wrist, right wrist, left hip, right hip, left knee, right knee, left ankle and right ankle.

4. The method for identifying panicked pedestrian wandering behavior based on motion trajectory according to claim 2 is characterized in that: The calculation formula for the centroid position of the body segment is as follows: x k =x p l p +x d l d and k =and p l p +y d l d z k =z p l p +z d l d In the formula, (x k ,y k 、z k ) is the coordinate of the segment centroid, (x p ,y p 、z p ) is the coordinate of the proximal end of the segment, (x d ,y d 、z d ) is the coordinate of the distal end of the segment, l p is the percentage of segment length from proximal to distal, l d It is the percentage of segment length from distal end to tip.

5. The method for identifying panicked pedestrian wandering behavior based on motion trajectory according to claim 2 is characterized in that: The calculation formula of the total center of mass position of the human body is as follows: In the formula, (x c ,y c 、z c ) is the coordinate of the center of mass of the human body; (x k ,y k 、z k ) is the coordinate of the kth segment; m k is the mass of the kth segment; M is the total mass of the body segments.

6. The method for identifying panicked pedestrian wandering behavior based on motion trajectory according to claim 1 is characterized in that: The wandering motion trajectory includes multiple elliptical motion trajectories, spiral motion trajectories, serpentine motion trajectories and Brownian motion trajectories.

7. The method for identifying panicked pedestrian wandering behavior based on motion trajectory according to claim 6 is characterized in that: The pedestrian wandering behavior judgment criteria are as follows: Where Oval.p. is the criterion for elliptical motion trajectory, Spiral.p. is the criterion for spiral motion trajectory, Snake.p. is the criterion for snake motion trajectory, Brown.p. is the criterion for Brownian motion trajectory, and N count is the threshold number of turning points in the pedestrian motion trajectory.

8. The method for identifying panicked pedestrian wandering behavior based on motion trajectory according to claim 1 is characterized in that: When the pedestrian moves back and forth, an inflection point appears in the pedestrian's motion trajectory, and whether the pedestrian's behavior is wandering behavior is determined based on whether the number of occurrences of the inflection point exceeds a threshold.

9. The method for identifying panicked pedestrian wandering behavior based on motion trajectory according to claim 8 is characterized in that: The inflection point is when the direction angle of the trajectory point in consecutive frames of the pedestrian motion trajectory changes by more than 90 degrees, and the calculation formula is as follows: Where θ is the direction angle of the trajectory point, P t-1 is the pedestrian trajectory coordinate point at the t-1th moment, P t is the pedestrian trajectory coordinate point at the tth moment, P t+1 is the pedestrian trajectory coordinate point at the t+1th moment.

10. A panic pedestrian wandering behavior recognition system based on motion trajectory, characterized in that: include: The pedestrian motion trajectory construction module is used to obtain the motion image of the pedestrian, identify the motion center of the pedestrian based on the image, and construct the pedestrian motion trajectory; A pedestrian wandering behavior recognition module is used to recognize the wandering behavior of pedestrians based on their movement trajectories according to the pedestrian wandering behavior criterion to obtain a recognition result; The steps for establishing the pedestrian wandering behavior criterion are as follows: Refer to the common wandering behaviors in public places and draw the possible wandering movement trajectories in different scenarios; The wandering trajectory of pedestrians is analyzed to obtain the criteria for pedestrian wandering behavior.

Images