Pedestrian microscopic behavior calculation and analysis method, device and equipment, storage medium and product

Abstract

The invention discloses a pedestrian microscopic behavior calculation and analysis method and device, equipment, a storage medium and a product, and relates to the technical field of computer vision, and the method comprises the steps: obtaining a first key point coordinate corresponding to a target pedestrian, the first key point coordinate being a three-dimensional space coordinate, and the first key point coordinate being used for representing a joint position of the target pedestrian; performing data storage, data cleaning and inverse perspective transformation on the first key point coordinates to obtain second key point coordinates; constructing a human skeleton model corresponding to the target pedestrian based on the second key point coordinates; and performing gait analysis and calculation on the target pedestrian through the human skeleton model to obtain a microscopic behavior index corresponding to the target pedestrian. The microcosmic behavior of the target pedestrian is calculated and analyzed by performing gait analysis and calculation on the human skeleton model constructed based on the key point coordinates of the target pedestrian, and the automatic and non-intervention method has the advantage of going deep into the microcosmic field of behavior analysis, so that the behavior analysis precision is improved.

Description

Technical Field

[0001] The present application relates to the field of computer vision technology, and in particular to a method, device, equipment, storage medium and product for computing and analyzing pedestrian micro-behavior. Background Art

[0002] Traditional machine learning methods in the field of behavior analysis are generally only used to count, obtain the trajectory of people in the area, or identify people's movements, and have not penetrated into the field of micro-behavior analysis; and motion capture methods require invasive experiments (sticking to the motion capture area, wearing motion capture suits, or wearing inertial sensors), and cannot observe people's micro-behavior in a natural state. Therefore, the industry is in urgent need of a method that can observe people's micro-behavior in a natural state. Summary of the invention

[0003] The main purpose of this application is to provide a method, device, equipment, storage medium and product for computing and analyzing pedestrian micro-behavior, aiming to solve the technical problem that existing methods cannot observe human micro-behavior in a natural state.

[0004] To achieve the above objectives, the present application provides a method for calculating and analyzing pedestrian micro-behavior, the method comprising the following steps:

[0005] Acquire first key point coordinates corresponding to the target pedestrian, where the first key point coordinates are three-dimensional space coordinates and are used to represent the joint position of the target pedestrian;

[0006] The coordinates of the first key points are saved, cleaned and transformed inversely to obtain the coordinates of the second key points;

[0007] Constructing a human skeleton model corresponding to the target pedestrian based on the coordinates of the second key point;

[0008] The gait of the target pedestrian is analyzed and calculated through the human skeleton model to obtain micro-behavior indicators corresponding to the target pedestrian.

[0009] In one embodiment, the step of obtaining the coordinates of the first key point corresponding to the target pedestrian includes:

[0010] Acquire a target video containing a target pedestrian image, and set a monitoring area and a monitoring frame in the target video;

[0011] In the monitoring area, the target pedestrian in the target video is tracked frame by frame based on the monitoring frame to obtain the first key point coordinates corresponding to the target pedestrian.

[0012] In one embodiment, the step of performing data storage, data cleaning and inverse perspective transformation on the first key point coordinates to obtain the second key point coordinates includes:

[0013] The first key point coordinates and the pedestrian ID of the target pedestrian are saved as data to obtain a comma separated value file, wherein the pedestrian ID and the first key point coordinates are in a one-to-one correspondence;

[0014] Data cleaning is performed on the comma separated value file, and an inverse perspective transformation is performed on the comma separated value file after the data cleaning to obtain the coordinates of the second key point corresponding to the target pedestrian.

[0015] In one embodiment, the step of constructing a human skeleton model corresponding to the target pedestrian based on the coordinates of the second key point includes:

[0016] Input the coordinates of the second key point into the three-dimensional modeling software to obtain the model point position;

[0017] The model points are point-connected according to the joint positions of the target pedestrian to obtain a human skeleton model corresponding to the target pedestrian.

[0018] In one embodiment, the step of performing gait analysis and calculation on the target pedestrian through the human skeleton model to obtain micro-behavior indicators corresponding to the target pedestrian includes:

[0019] Performing gait analysis on the target pedestrian according to the human skeleton model to obtain stride data corresponding to the target pedestrian in the target video;

[0020] The micro-behavior indicators corresponding to the target pedestrian are calculated based on the stride data, and the micro-behavior indicators include stride length, stride width, stride speed and double support time.

[0021] In one embodiment, after the step of performing gait analysis and calculation on the target pedestrian through the human skeleton model to obtain the micro-behavior index corresponding to the target pedestrian, the method further includes:

[0022] The specific numerical values ​​corresponding to each indicator in the micro-behavior indicator are converted into hue parameters, and the stride length, stride width, stride speed and double support time of the target pedestrian in the human skeleton model are visualized based on the hue parameters.

[0023] In addition, to achieve the above purpose, the present application also proposes a pedestrian micro-behavior computing and analyzing device, the pedestrian micro-behavior computing and analyzing device comprising:

[0024] A coordinate acquisition module, used to acquire first key point coordinates corresponding to a target pedestrian, wherein the first key point coordinates are three-dimensional space coordinates, and the first key point coordinates are used to represent a joint position of the target pedestrian;

[0025] A data processing module, used for performing data storage, data cleaning and inverse perspective transformation on the first key point coordinates to obtain the second key point coordinates;

[0026] A model building module, used to build a human skeleton model corresponding to the target pedestrian based on the coordinates of the second key point;

[0027] The gait analysis module is used to perform gait analysis and calculation on the target pedestrian through the human skeleton model to obtain micro-behavior indicators corresponding to the target pedestrian.

[0028] In addition, to achieve the above-mentioned purpose, the present application also proposes a pedestrian micro-behavior computing and analysis device, which includes: a memory, a processor, and a pedestrian micro-behavior computing and analysis program stored in the memory and executable on the processor, wherein the pedestrian micro-behavior computing and analysis program is configured to implement the steps of the pedestrian micro-behavior computing and analysis method as described above.

[0029] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a pedestrian micro-behavior calculation and analysis program is stored on the storage medium. When the pedestrian micro-behavior calculation and analysis program is executed by a processor, the steps of the pedestrian micro-behavior calculation and analysis method described above are implemented.

[0030] In addition, to achieve the above-mentioned purpose, the present invention also provides a computer program product, which includes a pedestrian micro-behavior calculation and analysis program, and when the pedestrian micro-behavior calculation and analysis program is executed by a processor, it implements the steps of the pedestrian micro-behavior calculation and analysis method described above.

[0031] The present application obtains the first key point coordinates corresponding to the target pedestrian, the first key point coordinates are three-dimensional space coordinates, and the first key point coordinates are used to represent the joint position of the target pedestrian; the first key point coordinates are subjected to data preservation, data cleaning and inverse perspective transformation to obtain the second key point coordinates; a human skeleton model corresponding to the target pedestrian is constructed based on the second key point coordinates; and the gait analysis and calculation of the target pedestrian is performed through the human skeleton model to obtain the micro-behavior indicators corresponding to the target pedestrian. Since the present application realizes the calculation and analysis of the micro-behavior of the target pedestrian by performing gait analysis and calculation on the human skeleton model constructed based on the key point coordinates of the target pedestrian to obtain the micro-behavior indicators corresponding to the target pedestrian, compared with the traditional calculation and analysis method of pedestrian micro-behavior, the above method of the present application avoids relying on data analysis of large amounts of data, realizes automatic and non-intrusive in-depth into the micro-field of behavior analysis, improves the accuracy of behavior analysis, and thus can observe the micro-behavior of pedestrians in a natural state. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A schematic diagram of the structure of a pedestrian micro-behavior computing and analyzing device in the hardware operating environment involved in the embodiment of the present application;

[0033] Figure 2 This is a flow chart of the first embodiment of the pedestrian micro-behavior calculation and analysis method of the present application;

[0034] Figure 3 A schematic diagram of a human skeleton model in the pedestrian micro-behavior computational analysis method of this application;

[0035] Figure 4 This is a flow chart of the second embodiment of the pedestrian micro-behavior calculation and analysis method of the present application;

[0036] Figure 5 This is a flowchart of the third embodiment of the pedestrian micro-behavior calculation and analysis method of the present application;

[0037] Figure 6 This is a structural block diagram of the first embodiment of the pedestrian micro-behavior calculation and analysis device of the present application.

[0038] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0039] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0040] Reference Figure 1 , Figure 1This is a schematic diagram of the structure of the pedestrian micro-behavior computing and analysis equipment in the hardware operating environment involved in the embodiment of the present application.

[0041] like Figure 1 As shown, the pedestrian micro-behavior computing and analysis device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0042] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the pedestrian micro-behavior computing and analysis device, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0043] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a pedestrian micro-behavior calculation and analysis program.

[0044] exist Figure 1 In the pedestrian micro-behavior computing and analysis device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the pedestrian micro-behavior computing and analysis device of the present application can be set in the pedestrian micro-behavior computing and analysis device, and the pedestrian micro-behavior computing and analysis device calls the pedestrian micro-behavior computing and analysis program stored in the memory 1005 through the processor 1001, and executes the pedestrian micro-behavior computing and analysis method provided in the embodiment of the present application.

[0045] The present application embodiment provides a method for calculating and analyzing pedestrian microscopic behavior, referring to Figure 2 , Figure 2This is a flow chart of the first embodiment of the pedestrian micro-behavior calculation and analysis method of the present application.

[0046] In this embodiment, the pedestrian micro-behavior calculation and analysis method includes the following steps:

[0047] Step S10: Obtain the first key point coordinates corresponding to the target pedestrian, where the first key point coordinates are three-dimensional space coordinates and are used to represent the joint position of the target pedestrian.

[0048] It should be noted that the execution subject of the method of this embodiment can be a terminal device with model building, data processing and program running functions, such as a smart terminal, a computer, etc., or an electronic device with the same or similar functions, such as the above-mentioned pedestrian micro-behavior computing and analysis device. The following takes the pedestrian micro-behavior computing and analysis device (hereinafter referred to as the analysis device) as an example to illustrate this embodiment and the following embodiments.

[0049] It is understandable that, since this embodiment is applied to the behavior analysis scenario between a person and an environment space, or a person and a building space, the target pedestrian may be a pedestrian moving in a target environment or a target building.

[0050] It should be understood that the first key point coordinates may refer to the coordinates corresponding to the joint positions of the target pedestrian in a three-dimensional space coordinate system, wherein the three-dimensional space coordinate system may be constructed based on the space where the target environment or the target building is located.

[0051] Step S20: performing data preservation, data cleaning and inverse perspective transformation on the first key point coordinates to obtain the second key point coordinates.

[0052] It should be understood that since the first key point coordinates are unprocessed initial coordinates, there may be abnormal data (such as null values, duplicate values, etc.), invalid data (i.e., data irrelevant to the calculation and analysis of pedestrian micro-behavior), etc. Based on this, this embodiment can save and clean the data of the first key point coordinates to remove the abnormal data and invalid data and then perform inverse perspective transformation to obtain the second key point coordinates, so that the coordinate data for the subsequent construction of the human skeleton model is purer, and the accuracy of model construction is improved.

[0053] Step S30: constructing a human skeleton model corresponding to the target pedestrian based on the coordinates of the second key points.

[0054] It should be noted that the above human skeleton model is a tool or representation method for analyzing and explaining human posture and movement, which usually includes the main joints of the human body (such as head, shoulders, elbows, wrists, hips, knees, ankles, etc.) and the bones connecting these joints. Figure 3 , Figure 3 This is a schematic diagram of the human skeleton model in the pedestrian micro-behavior calculation and analysis method of this application. It can be seen that the human skeleton model may include but is not limited to 17 joints, including but not limited to hips, right hips, right knees, right ankles, left hips, left knees, left ankles, spine, chest, neck, head, left shoulder, left elbow, left wrist, right shoulder, right elbow, and right wrist.

[0055] In a specific implementation, all the above-mentioned second key point coordinates can be placed in a unified spatial coordinate system, and all the second key point coordinates can be connected in the spatial coordinate system to obtain the human skeleton model corresponding to the above-mentioned target pedestrian.

[0056] Step S40: performing gait analysis and calculation on the target pedestrian using the human skeleton model to obtain micro-behavior indicators corresponding to the target pedestrian.

[0057] In a specific implementation, the human skeleton model can be adjusted to be displayed in frames for easy viewing, or the human skeleton model of each frame can be translated a certain distance to make the human skeleton model easier to view, and then the human skeleton model is placed in the space where the target environment or target building is located to analyze and calculate the gait of the target pedestrian, thereby obtaining the above-mentioned micro-behavior indicators. It should be noted that the above-mentioned micro-behavior indicators can be used to represent the relationship between the target pedestrian and the target environment or target building.

[0058] This embodiment obtains the first key point coordinates corresponding to the target pedestrian, the first key point coordinates are three-dimensional space coordinates, and the first key point coordinates are used to represent the joint position of the target pedestrian; the first key point coordinates are subjected to data preservation, data cleaning and inverse perspective transformation to obtain the second key point coordinates; a human skeleton model corresponding to the target pedestrian is constructed based on the second key point coordinates; and the gait analysis and calculation of the target pedestrian are performed through the human skeleton model to obtain the micro-behavior indicators corresponding to the target pedestrian. Since the present application realizes the calculation and analysis of the micro-behavior of the target pedestrian by performing gait analysis and calculation on the human skeleton model constructed based on the key point coordinates of the target pedestrian to obtain the micro-behavior indicators corresponding to the target pedestrian, compared with the traditional method of calculating and analyzing the micro-behavior of pedestrians, the above method of the present application avoids relying on data analysis of large amounts of data, realizes automatic and non-invasive in-depth into the micro-field of behavior analysis, improves the accuracy of behavior analysis, and thus can observe the micro-behavior of pedestrians in a natural state.

[0059] refer to Figure 4 , Figure 4 This is a flow chart of the second embodiment of the pedestrian micro-behavior calculation and analysis method of the present application.

[0060] In a feasible implementation manner, the step S10 may include:

[0061] Step S101: Acquire a target video containing a target pedestrian image, and set a monitoring area and a monitoring frame in the target video.

[0062] It is understandable that the target video of the target pedestrian image can be obtained by shooting with a monocular camera.

[0063] Step S102: Tracking the target pedestrian in the target video frame by frame in the monitoring area based on the monitoring frame to obtain the first key point coordinates corresponding to the target pedestrian.

[0064] It is understandable that in the process of computational analysis of pedestrian micro-behavior, that is, studying the relationship between human behavior and the environment and space, this embodiment hopes to observe the observed target pedestrian without intervention, so that the target pedestrian is in a natural state, so as to obtain accurate research data, and at the same time hopes to automatically acquire data in batches. In this case, the data of the surveillance video (i.e., the above-mentioned target video) can serve as an ideal data source. In particular, the target detection and target tracking algorithms can be combined to annotate pedestrians in continuous video frames, and then the three-dimensional coordinates of the key points of the pedestrians detected in each frame (i.e., the coordinates of the first key point mentioned above) can be obtained through the human body posture estimation algorithm.

[0065] In the specific implementation, for the target detection process, the corresponding area can be demarcated in the target video to identify pedestrians within a specific range. For example, the aspect ratio of the monitoring frame can be set to achieve the effect of distinguishing adults from children.

[0066] In a feasible implementation manner, the step S20 may include:

[0067] Step S201: The first key point coordinates and the pedestrian ID of the target pedestrian are saved as data to obtain a comma separated value file, wherein the pedestrian ID and the first key point coordinates are in a one-to-one correspondence.

[0068] Step S202: performing data cleaning on the comma separated value file, and performing inverse perspective transformation on the comma separated value file after data cleaning to obtain the coordinates of the second key point corresponding to the target pedestrian.

[0069] It should be noted that the above-mentioned comma-separated value file, namely CSV (Comma-Separated Values) file, is a common text file format used to store and exchange simple data structures and large data sets. CSV files store tabular data in plain text form, where each row of data consists of fields separated by commas. These fields usually represent columns in a spreadsheet, and each row represents a record.

[0070] It should be understood that the relative positions of the 17 joints of the first key point coordinates are accurate for a single person, but in a multi-person scene, the relative positions between people are inaccurate due to the perspective of the shooting, so it is necessary to perform an inverse perspective transformation on the first key point coordinates after data cleaning in the comma-separated value file. More specifically, in the inverse perspective transformation process, 4 coordinates on the horizontal plane of the real world and their corresponding 4 pixel coordinates of the image can be pre-set as parameters to construct a transformation matrix, and then all the processed first key point coordinates in the comma-separated value file are converted into the second key point coordinates through this matrix.

[0071] In a feasible implementation manner, the step S30 may include:

[0072] Step S301: input the coordinates of the second key point into three-dimensional modeling software to obtain model points.

[0073] Step S302: Connect the model points according to the joint positions of the target pedestrian to obtain a human skeleton model corresponding to the target pedestrian.

[0074] This embodiment obtains a target video containing a target pedestrian image, and sets a monitoring area and a monitoring frame in the target video; in the monitoring area, the target pedestrian in the target video is tracked frame by frame based on the monitoring frame to obtain the first key point coordinates corresponding to the target pedestrian; the first key point coordinates and the pedestrian ID of the target pedestrian are saved to obtain a comma separated value file, and the pedestrian ID and the first key point coordinates are in a one-to-one correspondence; the comma separated value file is cleaned, and the comma separated value file after data cleaning is inversely transformed to obtain the second key point coordinates corresponding to the target pedestrian; the second key point coordinates are input into the three-dimensional modeling software to obtain the model points; the model points are connected point by point according to the joint positions of the target pedestrian to obtain the human skeleton model corresponding to the target pedestrian. Compared with the traditional pedestrian micro-behavior calculation and analysis method, the above method of this embodiment reconstructs the human skeleton model of the target pedestrian based on the key point coordinates corresponding to the joint positions of the target pedestrian, so that the human skeleton model of the target pedestrian can be coupled with the space, thereby providing a data basis for the subsequent micro-behavior calculation and analysis process of the pedestrian. In addition, the data collection process in this embodiment has a low degree of intervention on the target pedestrian, reducing the adverse effects of behavioral observation on the target pedestrian.

[0075] refer to Figure 5 , Figure 5 This is a flow chart of the third embodiment of the pedestrian micro-behavior calculation and analysis method of the present application.

[0076] In a feasible implementation manner, the step S40 may include:

[0077] Step S401: performing gait analysis on the target pedestrian according to the human skeleton model to obtain stride data corresponding to the target pedestrian in the target video.

[0078] It should be understood that the perimeter (i.e., a complete stride cycle) needs to be segmented according to the number of strides of the ankle joint point. For example, a certain height threshold is set, and the ankle joint point is judged to be on the ground when the z coordinate is lower than this height. Based on this, a person takes two steps as a stride, and then the number of frames to which belongs to a stride is obtained. In particular, the above-mentioned height threshold can be obtained based on the simulated annealing algorithm, with the goal of maximizing the number of segmented steps, and the height threshold is automatically optimized.

[0079] Step S402: Calculate micro-behavior indicators corresponding to the target pedestrian based on the stride data, where the micro-behavior indicators include stride length, stride width, stride speed, and double support time.

[0080] In a specific implementation, the micro-behavior indicators corresponding to the target pedestrian can be calculated based on the following steps. First, the body orientation is calculated (obtained using the vertical direction of the shoulder line), the distance between two consecutive footfalls is projected on the body orientation as the stride length, the distance between the left and right feet landing alternately is projected on the shoulder line direction as the stride width, and the time of each step is calculated according to the input video frame rate (for example, if the frame rate is 30fps and there are 60 frames in one step, the time of each step is 60 / 30=2 seconds), the stride speed is obtained by dividing the stride length by the time of each step, and the double support time is calculated according to the frame rate and the number of frames in which both feet land.

[0081] In a feasible implementation manner, after step S40, the following may also be included:

[0082] Step S50: Convert the specific numerical values ​​corresponding to each indicator in the micro-behavior indicator into hue parameters, and visualize the stride length, stride width, stride speed and double support time of the target pedestrian in the human skeleton model based on the hue parameters.

[0083] In the specific implementation, taking the visualization of stride speed as an example, first map the stride speed of all strides to the range of 0-1, then convert the numbers into corresponding hues and assign them to the corresponding model of the corresponding frame. At this point, we can analyze and judge where the stride speed of people is large and where it is small in the space, so as to further carry out large-scale and refined batch environmental behavior research.

[0084] This implementation performs gait analysis on the target pedestrian based on the human skeleton model to obtain the stride data corresponding to the target pedestrian in the target video; calculates the micro-behavior indicators corresponding to the target pedestrian based on the stride data, and the micro-behavior indicators include stride length, stride width, stride speed and double support time; converts the specific numerical values ​​corresponding to each indicator in the micro-behavior indicators into hue parameters, and visualizes the stride length, stride width, stride speed and double support time of the target pedestrian in the human skeleton model based on the hue parameters. Compared with the traditional pedestrian micro-behavior calculation and analysis method, the above method of this embodiment calculates the stride length, stride width, stride speed and double support time corresponding to the target pedestrian based on the stride data corresponding to the target pedestrian in the target video, thereby realizing targeted gait analysis and calculation of the target pedestrian; at the same time, by visually displaying each indicator of the target pedestrian in the human skeleton model, it is convenient to analyze and identify the micro-behavior indicators of the target pedestrian.

[0085] In addition, an embodiment of the present application also proposes a storage medium, on which a pedestrian micro-behavior calculation and analysis program is stored. When the pedestrian micro-behavior calculation and analysis program is executed by a processor, the steps of the pedestrian micro-behavior calculation and analysis method described above are implemented.

[0086] Reference Figure 6 , Figure 6 This is a structural block diagram of the first embodiment of the pedestrian micro-behavior calculation and analysis device of the present application.

[0087] like Figure 6 As shown, the pedestrian micro-behavior computing and analyzing device proposed in the embodiment of the present application includes:

[0088] A coordinate acquisition module 601 is used to acquire first key point coordinates corresponding to a target pedestrian, where the first key point coordinates are three-dimensional space coordinates and are used to represent joint positions of the target pedestrian;

[0089] A data processing module 602 is used to save, clean and perform inverse perspective transformation on the first key point coordinates to obtain the second key point coordinates;

[0090] A model building module 603 is used to build a human skeleton model corresponding to the target pedestrian based on the coordinates of the second key point;

[0091] The gait analysis module 604 is used to perform gait analysis and calculation on the target pedestrian through the human skeleton model to obtain micro-behavior indicators corresponding to the target pedestrian.

[0092] This embodiment obtains the first key point coordinates corresponding to the target pedestrian, the first key point coordinates are three-dimensional space coordinates, and the first key point coordinates are used to represent the joint position of the target pedestrian; the first key point coordinates are subjected to data preservation, data cleaning and inverse perspective transformation to obtain the second key point coordinates; a human skeleton model corresponding to the target pedestrian is constructed based on the second key point coordinates; and the gait analysis and calculation of the target pedestrian are performed through the human skeleton model to obtain the micro-behavior indicators corresponding to the target pedestrian. Since the present application realizes the calculation and analysis of the micro-behavior of the target pedestrian by performing gait analysis and calculation on the human skeleton model constructed based on the key point coordinates of the target pedestrian to obtain the micro-behavior indicators corresponding to the target pedestrian, compared with the traditional method of calculating and analyzing the micro-behavior of pedestrians, the above method of the present application avoids relying on data analysis of large amounts of data, realizes automatic and non-invasive in-depth into the micro-field of behavior analysis, improves the accuracy of behavior analysis, and thus can observe the micro-behavior of pedestrians in a natural state.

[0093] Based on the first embodiment of the pedestrian micro-behavior computing and analyzing device of the present application, a second embodiment of the pedestrian micro-behavior computing and analyzing device of the present application is proposed.

[0094] In this embodiment, the coordinate acquisition module 601 is also used to acquire a target video containing an image of a target pedestrian, and set a monitoring area and a monitoring frame in the target video; in the monitoring area, the target pedestrian in the target video is tracked frame by frame based on the monitoring frame to obtain the first key point coordinates corresponding to the target pedestrian.

[0095] Furthermore, the data processing module 602 is also used to save the first key point coordinates and the pedestrian ID of the target pedestrian to obtain a comma separated value file, and the pedestrian ID and the first key point coordinates are in a one-to-one correspondence; the comma separated value file is cleaned, and the comma separated value file after the data cleaning is performed an inverse perspective transformation to obtain the second key point coordinates corresponding to the target pedestrian.

[0096] Furthermore, the model building module 603 is also used to input the coordinates of the second key points into three-dimensional modeling software to obtain model points; connect the model points according to the joint positions of the target pedestrian to obtain a human skeleton model corresponding to the target pedestrian.

[0097] Furthermore, the gait analysis module 604 is also used to perform gait analysis on the target pedestrian according to the human skeleton model to obtain the stride data corresponding to the target pedestrian in the target video; based on the stride data, the micro-behavior indicators corresponding to the target pedestrian are calculated, and the micro-behavior indicators include stride length, stride width, stride speed and double support time.

[0098] Furthermore, the gait analysis module 604 is also used to convert the specific numerical value corresponding to each indicator in the micro-behavior indicator into a hue parameter, and based on the hue parameter, visualize the stride length, stride width, stride speed and double support time of the target pedestrian in the human skeleton model.

[0099] Other embodiments or specific implementation methods of the pedestrian micro-behavior computing and analyzing device of the present application can refer to the above-mentioned method embodiments and will not be repeated here.

[0100] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0101] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0102] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0103] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for calculating and analyzing pedestrian microscopic behavior, characterized in that: The method comprises the following steps: Acquire first key point coordinates corresponding to the target pedestrian, where the first key point coordinates are three-dimensional space coordinates and are used to represent the joint position of the target pedestrian; The coordinates of the first key points are saved, cleaned and transformed inversely to obtain the coordinates of the second key points; Constructing a human skeleton model corresponding to the target pedestrian based on the coordinates of the second key point; The gait of the target pedestrian is analyzed and calculated through the human skeleton model to obtain micro-behavior indicators corresponding to the target pedestrian.

2. The pedestrian micro-behavior calculation and analysis method according to claim 1, characterized in that: The step of obtaining the coordinates of the first key point corresponding to the target pedestrian includes: Acquire a target video containing a target pedestrian image, and set a monitoring area and a monitoring frame in the target video; In the monitoring area, the target pedestrian in the target video is tracked frame by frame based on the monitoring frame to obtain the first key point coordinates corresponding to the target pedestrian.

3. The pedestrian micro-behavior calculation and analysis method according to claim 1, characterized in that: The step of performing data storage, data cleaning and inverse perspective transformation on the first key point coordinates to obtain the second key point coordinates includes: The first key point coordinates and the pedestrian ID of the target pedestrian are saved as data to obtain a comma separated value file, wherein the pedestrian ID and the first key point coordinates are in a one-to-one correspondence; Data cleaning is performed on the comma separated value file, and an inverse perspective transformation is performed on the comma separated value file after the data cleaning to obtain the coordinates of the second key point corresponding to the target pedestrian.

4. The pedestrian micro-behavior calculation and analysis method according to claim 1, characterized in that: The step of constructing a human skeleton model corresponding to the target pedestrian based on the coordinates of the second key point includes: Input the coordinates of the second key point into the three-dimensional modeling software to obtain the model point position; The model points are point-connected according to the joint positions of the target pedestrian to obtain a human skeleton model corresponding to the target pedestrian.

5. The pedestrian micro-behavior calculation and analysis method according to claim 1, characterized in that: The step of performing gait analysis and calculation on the target pedestrian through the human skeleton model to obtain microscopic behavior indicators corresponding to the target pedestrian includes: Performing gait analysis on the target pedestrian according to the human skeleton model to obtain stride data corresponding to the target pedestrian in the target video; The micro-behavior indicators corresponding to the target pedestrian are calculated based on the stride data, and the micro-behavior indicators include stride length, stride width, stride speed and double support time.

6. The pedestrian micro-behavior calculation and analysis method according to claim 1, characterized in that: After the step of analyzing and calculating the gait of the target pedestrian using the human skeleton model to obtain the microscopic behavior index corresponding to the target pedestrian, the method further includes: The specific numerical values ​​corresponding to each indicator in the micro-behavior indicator are converted into hue parameters, and the stride length, stride width, stride speed and double support time of the target pedestrian in the human skeleton model are visualized based on the hue parameters.

7. A pedestrian micro-behavior computing and analysis device, characterized in that: The pedestrian micro-behavior computing and analyzing device comprises: A coordinate acquisition module, used to acquire first key point coordinates corresponding to a target pedestrian, wherein the first key point coordinates are three-dimensional space coordinates, and the first key point coordinates are used to represent a joint position of the target pedestrian; A data processing module, used for performing data storage, data cleaning and inverse perspective transformation on the first key point coordinates to obtain the second key point coordinates; A model building module, used to build a human skeleton model corresponding to the target pedestrian based on the coordinates of the second key point; The gait analysis module is used to perform gait analysis and calculation on the target pedestrian through the human skeleton model to obtain micro-behavior indicators corresponding to the target pedestrian.

8. A pedestrian micro-behavior computing and analysis device, characterized in that: The device includes: a memory, a processor, and a pedestrian micro-behavior computing and analysis program stored in the memory and executable on the processor, wherein the pedestrian micro-behavior computing and analysis program is configured to implement the steps of the pedestrian micro-behavior computing and analysis method as described in any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, on which a pedestrian micro-behavior calculation and analysis program is stored. When the pedestrian micro-behavior calculation and analysis program is executed by a processor, the steps of the pedestrian micro-behavior calculation and analysis method as described in any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product includes a pedestrian micro-behavior calculation and analysis program, which, when executed by a processor, implements the steps of the pedestrian micro-behavior calculation and analysis method as described in any one of claims 1 to 6.

Images