JP2005346617A - Passer-by behavior analysis system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a system capable of accurately analyzing behaviors of passers-by. <P>SOLUTION: A passer-by behavior analysis system for analyzing behaviors of passers-by within a prescribed space is provided with; a location detection means for obtaining respective location coordinates of passers-by passing the space; a tracking processing means for tracking a location of one passer-by to obtain the temporal change of the location coordinate of the passer-by; a storage means for storing data of temporal changes of location coordinates within a prescribed time per passer-by; a means for obtaining a track of each passer-by within the space from temporal change data of location coordinates stored in the storage means; and a display means for displaying tracks of all passers-by who have passed the prescribed space within the prescribed time. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a passer-behavior analysis system that analyzes a passer-by behavior in a space where a passer-by moves and stays at a railway station or the like.

Passengers using the station can flow and move smoothly is an important requirement for the comfort of the station, but passengers and transfer passengers are complicated, stoppages and shops to check the signs and entrances and exits There are many factors that impede the smooth flow, such as shopping at the store and staying at the meeting. This causes problems such as a decrease in walking speed due to passenger collisions and congestion.
Conventional passenger flow surveys, which have been conducted as basic information collection for studying improvement works that improve the comfort of the station environment, are mainly to grasp the number of people passing through the section, and manual surveys are the mainstream. Therefore, it was difficult to comprehensively analyze the speed, direction, and density of passengers moving across the station.

In order to solve such a problem, a camera of a visible image is installed at a road intersection, and the camera captures an image in a monitoring area of a predetermined range including a pedestrian crossing. By examining the situation and the number of people, a technique is known that can more efficiently control the signal for the vehicle and the signal for the pedestrian, and as a result, improve the safety of the pedestrian (see, for example, Patent Document 1). ).
JP 11-275562 A

  However, since the device disclosed in Patent Document 1 uses a visible video image, tracking processing may be difficult when passers-by interlaces, and the accurate position tracking processing of passers-by is performed. There is a problem that it is difficult. In addition, in order to avoid this problem, the installation position and the installation angle of the camera are greatly restricted, the imaging range becomes narrow, and there are problems that the calculation processing load increases when there are many passersby.

  The present invention has been made in view of such circumstances, and an object of the present invention is to provide a passer-by behavior analysis system that can accurately analyze a passer-by behavior using a distance image.

  The invention according to claim 1 is a passerby behavior analysis system for analyzing a passerby's behavior in a predetermined space, wherein a plurality of distance image acquisition means for acquiring a distance image, and the plurality of distance image acquisition means Synchronization means for synchronizing the distance image acquisition timing, coordinate system integration means for integrating the coordinate systems of the plurality of distance images respectively acquired by the distance image acquisition means into one coordinate system, and the distance image obtained by integrating the coordinate systems Position detecting means for determining the position coordinates of each passerby who passes through the space, tracking processing means for tracking the position of the same passerby and determining temporal changes in the position coordinates of each passer, and a predetermined time A first storage means for storing data of temporal changes in position coordinates for each passer-by, and a time change data of position coordinates stored in the storage means in the space Density calculating means for obtaining the density of each area from the number of passersby present in each of the divided areas, and second storage means in which the density value and the number of distance image acquiring means are stored in association with each other in advance. And behavior analysis means for analyzing the behavior of a passer-by based on data of temporal changes in position coordinates within a predetermined time stored in the first storage means, and the coordinate system integration means The number of distance image acquisition means related to the density value obtained by the density calculation means is read from the second storage means, and the distance image obtained by using the read number of distance image acquisition means It is characterized by integrating the coordinate system.

  The invention according to claim 2 is a passer-by behavior analysis system that analyzes the behavior of passers-by in a predetermined space, and is the same as the position detecting means for obtaining the position coordinates of each passer-by passing through the space. Tracking processing means for tracking the position of each passerby to obtain a temporal change in the position coordinates of each passer, and storage means for storing the data of the temporal change in the position coordinates within a predetermined time for each passerby; , Means for obtaining the trajectory of each passerby in the space from the temporal change data of the position coordinates stored in the storage means, and displaying the trajectories of all passers-by who pass within the predetermined time in the predetermined space And display means.

  The invention described in claim 3 is a passer-by behavior analysis system for analyzing the behavior of passers-by in a predetermined space, and is the same as the position detection means for obtaining the position coordinates of each passer-by passing through the space. Tracking processing means for tracking the position of each passerby to obtain a temporal change in the position coordinates of each passer, and storage means for storing the data of the temporal change in the position coordinates within a predetermined time for each passerby; , Means for determining the passing speed of each passerby in the space from the temporal change data of the position coordinates stored in the storage means, and the pass speed distribution of all passersby passing within a predetermined time in the predetermined space. And display means for displaying.

  The invention described in claim 4 is a passer-by behavior analysis system that analyzes the behavior of passers-by in a predetermined space, and is the same as the position detection means that obtains the position coordinates of each passer-by passing through the space. Tracking processing means for tracking the position of each passerby to obtain a temporal change in the position coordinates of each passer, and storage means for storing the data of the temporal change in the position coordinates within a predetermined time for each passerby; , Means for obtaining a passage trajectory of each passerby in the space from temporal change data of the position coordinates stored in the storage means, an approach relative distance between passersby in the space from the pass trajectory, and Means for obtaining a crossing angle formed by a traveling direction vector, and a display for displaying a position where the approach distance and the crossing angle are within a predetermined threshold range within a predetermined time in a predetermined space. Characterized by comprising a stage.

  The invention according to claim 5 is a passer-by behavior analysis system for analyzing the behavior of passers-by in a predetermined space, and is the same as the position detecting means for obtaining the position coordinates of each passer-by passing through the space. Tracking processing means for tracking the position of each passerby to obtain a temporal change in the position coordinates of each passer, and storage means for storing the data of the temporal change in the position coordinates within a predetermined time for each passerby; , Means for determining the number of passersby present in each area obtained by dividing the space from the temporal change data of the position coordinates stored in the storage means, and existing in the area of the predetermined space within the predetermined time And a display means for displaying the density distribution of the passerby.

  According to the present invention, since the behavior of the passerby is accurately analyzed using the distance image, the problem that occurs in the analysis process using the visible image is solved, and the speed, direction, and density of the passerby are comprehensively determined. Can be analyzed. Further, since the number of means for obtaining the distance image is determined according to the traffic density, the effect that it is possible to perform the optimum measurement with the optimum number is obtained.

  Hereinafter, a passerby behavior analysis system according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the embodiment. In this figure, reference numeral 1 denotes a server that performs an analysis process on the behavior of a passerby 7 in a predetermined space 6. Reference numeral 2 denotes a laser sensor that does not require a reflector or a position mark, and scans the surroundings two-dimensionally to measure the distance to an object. The laser sensor 2 scans a horizontal surface of 16.3 cm from the floor, and displays a laser distance image in an orthogonal coordinate system including both a stationary object and a moving object in a horizontal section corresponding to the height of the laser projection. Can be obtained. Reference numeral 3 denotes a client PC (personal computer) provided for each laser sensor 2 in order to control each laser sensor 2. Reference numeral 4 denotes a network for performing information communication between each client PC 3 and the server 1 in order to transfer the measurement data stored in the client PC 3 to the server 1 and to send instruction information from the server 1 to the client PC 3. Used for. Reference numeral 5 denotes a video camera that captures a visible image in a predetermined space 6, and the obtained image is taken into the server 1 and used for verification of a behavior analysis result.

  Next, the configuration of the server 1 will be described with reference to FIG. FIG. 2 is a block diagram showing a configuration of the server 1 shown in FIG. In this figure, reference numeral 11 denotes a data receiving unit that receives distance image data from each client PC 3. Reference numeral 12 denotes a distance image data storage unit that stores the distance image data received by the data receiving unit 11. Reference numeral 13 denotes a synchronization processing unit that synchronizes a plurality of distance image data received from a plurality of client PCs 3 in time and integrates them into one coordinate system. Reference numeral 14 is synchronized with time, extracts the foot portion of the passerby 7 from the distance image data integrated into one coordinate system, executes a tracking process that the passerby 7 tracks, and passes through the predetermined space 6. It is a locus | trajectory calculation part which calculates the passage locus | trajectory of the passerby who passed within the predetermined time. Reference numeral 15 denotes a trajectory data storage unit that stores the pass trajectory data of each passer-by obtained by the trajectory calculation unit 14. Reference numeral 16 denotes a density calculation unit that calculates and outputs the passer-by density of a small area obtained by dividing the space 6 into a plurality from the trajectory data stored in the trajectory data storage unit 15. Reference numeral 17 denotes a speed calculation unit that calculates and outputs a speed distribution of a passerby from the travel path data stored in the path data storage unit 15. Reference numeral 18 denotes a congestion degree calculation unit that detects a state where passers-by meets predetermined conditions from the passage locus data stored in the locus data storage unit 15 and outputs this state as a congestion degree. Reference numeral 19 denotes a trajectory output unit that reads and outputs the traffic trajectory data stored in the trajectory data storage unit 15. Reference numeral 20 denotes a display unit configured by a display. Reference numeral 21 denotes an analysis result storage unit that stores behavior analysis data (traffic trajectory, traffic speed, traffic density, congestion degree). Reference numeral 22 denotes a number determination unit that determines the number of operating laser sensors 2 according to the passerby density. Reference numeral 23 is a number table in which the number of operating laser sensors 2 is defined in advance for each passer's density level. Reference numeral 24 denotes a data transmission unit that notifies the client PC 3 of the determined number.

  Next, an operation of extracting and tracking the passerby 7 will be described. First, each client PC 3 transfers distance image data obtained from the laser sensor 2 to the server 1. In response to this, the data receiving unit 11 stores the distance image data received from each client PC 3 in the distance image data storage unit 12.

  Next, the synchronization processing unit 13 reads a plurality of distance image data stored in the distance image data storage unit 12, removes fixed objects such as walls and pillars, and performs background difference processing for creating data only for the moving object. Do. Since the laser sensor 2 always scans the same range at the same position, the position of the wall or the column does not change beyond the system error range of the system. Using this principle, a fixed object whose position does not change with time is stored as a background. As a result, temporal image synchronization and distance image data integrated in one coordinate system can be obtained. The synchronization processing unit 13 stores the distance image data in the storage unit 12. Then, the synchronization processing unit 13 notifies the locus calculation unit 14 of the end of the synchronization processing.

  In response to this, the trajectory calculation unit 14 reads out the distance image data from the distance image data storage unit 12, collects a plurality of laser beams that have hit the passersby from the distance image data, and which point cloud is one. A clustering process is performed to estimate whether the foot hits the foot. This is a process of extracting moving body data based on the first background difference, and clustering foot cross-section data of data collected within a radius of 10 cm in these point clouds.

  Next, the trajectory calculation unit 14 performs a grouping process to estimate which of the two legs obtained by the clustering process corresponds to one passer-by. This is a process of grouping the clustered foot cross-section data with a distance within 30 cm as one passer-by.

  Next, the trajectory calculation unit 14 performs a tracking process that tracks where the grouped passerby has moved. When a human walks, he or she moves using either the left or right foot as an axis foot. If this operation is extracted from time-sequential frames, the grouped foot cross-sections will overlap to some extent. Here, if there is an overlap of 3 frames or more in time series, the seeds that are the basis of the traffic track can be extracted by considering these as traffic tracks. Subsequently, a discrete Kalman filter is used to combine the seeds created in the tracking process to extend the trajectory. As a result, the trajectory of each passerby is obtained, and the trajectory calculation unit 14 stores the obtained trajectory in the trajectory data storage unit 15. The trajectory data is composed of the position coordinate values at each time (elapsed time from the start of measurement) for each passerby. A processing flow conceptually summarizing the tracking operation is shown in steps S1 to S5 in FIG.

  Next, an operation for outputting the trajectory data stored in the trajectory data storage unit 15 will be described. In response to the input operation of the analyst, the trajectory output unit 19 reads the trajectory data from the trajectory data storage unit 15, connects each passerby with a line that interpolates between the existing position coordinate values, and color-codes each travel direction. Are displayed on the display unit 20. FIG. 4 shows an example in which locus data is displayed on the display unit 20. FIG. 4A is a diagram in which locus data is plotted with respect to a plan view of a space. FIG. 4B is an enlarged view of a part of FIG. FIG. 6 shows an example in which the plan view of the station concourse shown in FIG.

  Next, based on the trajectory data stored in the trajectory data storage unit 15, the density calculation unit 16 that calculates and outputs the density of passers-by passing through the concourse of the station shown in FIG. 5 will be described. To do. First, the density calculation unit 16 divides the measurement target space into small areas (1 m square area) in response to an input operation of density distribution display by the analyst. Then, the density calculation unit 16 refers to the trajectory data stored in the trajectory data storage unit 15 and counts the number of trajectories that have passed through the divided area within a unit time (for example, 30 seconds). The count value is displayed on the display unit 20 as a density intensity value. FIG. 7 shows an example in which the plan view of the station concourse shown in FIG. 5 and the density value in each region are displayed in an overlapping manner.

  Next, based on the trajectory data stored in the trajectory data storage unit 15, the operation of the speed calculation unit 17 that calculates and outputs the speed of the passers-by passing through the concourse of the station shown in FIG. To do. First, the speed calculation unit 16 divides the measurement target space into small areas (1 m square area) in response to the input operation of the speed distribution display of the analyst. Then, the speed calculation unit 17 refers to the trajectory data stored in the trajectory data storage unit 15, and calculates the speed of the trajectory that has passed through the divided area in a predetermined direction within a unit time (for example, 30 seconds). Ask. This speed value is displayed on the display unit 20 as a speed distribution. FIG. 8 shows an example in which the plan view of the station concourse shown in FIG. 5 and the speed value in each area (the speed of the passers-by from the platform to the ticket gate) are displayed in an overlapping manner.

  Next, based on the trajectory data stored in the trajectory data storage unit 15, the congestion degree calculation unit 18 that calculates and outputs the congestion degree of passers-by passing through the concourse of the station shown in FIG. 5. Will be explained. Here, the degree of congestion will be described with reference to FIG. Congestion level indicates how many times each unit time a state in which the approach distance between passers-by and the crossing angle formed by the traveling direction vector are within a predetermined threshold range has occurred. Degree. The congestion degree calculation unit 18 reads the trajectory data stored in the trajectory data storage unit 15 in response to an input operation of the congestion level display by the analyst, and interpolates between the respective position coordinates for each passerby. Are connected to each other and displayed on the display unit 20, and the intersection angle formed by the approach distance and the traveling direction vector is obtained from each trajectory data. Then, the congestion degree calculation unit 18 superimposes and displays the position where the state in which the approach distance is 60 cm or less and the crossing angle formed by the traveling direction vector is 180 ° ± 45 ° is superimposed on the trajectory display. In FIG. 10, the plan view of the station concourse shown in FIG. 5, the trajectory, the approaching distance between the passers-by and the intersection angle formed by the traveling direction vector are within predetermined threshold values. An example in which the position where the state has occurred is displayed in an overlapping manner is shown.

  Next, with reference to FIG. 11, the operation of determining the number of laser sensors 2 to be used will be described based on the obtained passer density value. First, the data receiving unit 11 receives the distance image data from each client PC 3, and the density calculating unit 16 calculates the passerby density by the operation described above (steps S11 and S12). The calculated density value is output from the density calculation unit 16 to the number determination unit 22. In response to this, the number determination unit 22 determines whether or not the density value obtained by the immediately preceding process and the newly obtained density level have changed (step S13). This determination is performed by referring to the number table 23 to determine which density level the obtained density value corresponds to and determining whether the density level has changed.

Here, the structure of the number table will be described with reference to FIG. The number table 23 stores the number of laser sensors 3 to be used (1 to 4), the density level (0 to 3), and the density value range in advance. For example, when the maximum density value output from the density calculation unit 16 is “0.3 person / m ² ”, the density level is “2”, and the maximum density value newly output is “0.6”. In the case of “person / m ² ”, since the density level is “3”, it is considered that the density level has changed in the determination in step S13.

  Next, when the density level changes as a result of the determination in step S13, the number determination unit 22 reads the number corresponding to the new density level from the number table 23, and determines this read number as the used number (step). S14). Then, the number determination unit 22 notifies each client PC 3 of the number of use determined here via the data transmission unit 24 (step S15). In response to this, the client PC 3 controls the laser sensor 2 based on the notified number of used devices. Each client PC 3 determines in advance whether or not to operate the laser sensor 2 connected to itself for each value of the number of units used, and operates the laser sensor 2 connected to itself based on this rule. (Step S16). As a result, the number of laser sensors 2 to be used is increased or decreased based on the passerby density. When the density is low, measurement can be performed with a small number, and when the density is large, measurement can be performed with a large number. .

  Thus, by visualizing the passerby's trajectory data obtained by integrating the distance image data by color-coded for each direction, it is possible to grasp which part of the passerby who is moving to a specific destination is moving. . In addition, by providing the trajectory data of passersby as a threshold with the relative distance between passers-by at each time and the crossing angle formed by the pedestrian travel direction vector, it is possible to detect a collision phenomenon between passers-by, By visualizing the location where the collision phenomenon is detected, it is possible to grasp where such a phenomenon occurs in the concourse of the station. Also, after dividing the passerby's trajectory data by direction, assuming a predetermined area on the data, average the distance traveled for a certain time by a vector for each direction that passed the area, and the average value is calculated by time By dividing, the average moving speed of passersby in the area is obtained. By displaying the size of the average moving speed in different colors, it is possible to grasp in which part of the concourse it takes time and the difficulty of moving. Also, by counting the number of trajectories that have passed through a predetermined area provided on the passerby's trajectory data per unit time and dividing the counted value by the area, the passerby's flow density in that area is obtained. be able to. By displaying the size of the flow density in different colors, it is possible to grasp which part of the concourse is crowded. In addition, since the number of laser sensors to be used is determined according to the traffic density, it is possible to perform optimum measurement with the optimum number.

  Also, by combining the traffic density and the traffic speed, it is easy to move in the crowd, difficult to move in the vacant space, crowded and difficult to move, vacant and easy to move in the concourse. The characteristics of each part can be grasped. Further, by combining the passage speed and the collision detection, it is possible to grasp the relationship between the ease and difficulty of movement for each location in the concourse and the appearance frequency of collision between passers-by. Furthermore, by combining the traffic density and collision detection, it is possible to grasp the relationship between the degree of congestion at each location in the concourse and the frequency of appearance of collision between passengers.

  Note that a program for realizing the function of the processing unit in FIG. 1 is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into a computer system and executed, thereby executing a passerby action. Analysis processing may be performed. Here, the “computer system” includes an OS and hardware such as peripheral devices. The “computer system” includes a WWW system having a homepage providing environment (or display environment). The “computer-readable recording medium” refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, and a CD-ROM, and a storage device such as a hard disk built in the computer system. Further, the “computer-readable recording medium” refers to a volatile memory (RAM) in a computer system that becomes a server or a client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. In addition, those holding programs for a certain period of time are also included.

  The program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.

It is a schematic diagram which shows the structure of one Embodiment of this invention. It is a block diagram which shows the structure of the server 1 shown in FIG. It is explanatory drawing which shows operation | movement of a tracking process. It is explanatory drawing which shows the example of a display of a trajectory. It is a top view of the space of measurement object. It is explanatory drawing which shows the example of a display of an action analysis result. It is explanatory drawing which shows the example of a display of an action analysis result. It is explanatory drawing which shows the example of a display of an action analysis result. It is explanatory drawing which shows the figure explaining a collision phenomenon. It is explanatory drawing which shows the example of a display of an action analysis result. It is a flowchart which shows the operation | movement of a used number determination process. It is explanatory drawing which shows the table structure of a number table.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Server, 2 ... Laser sensor, 3 ... Client PC, 4 ... Network, 5 ... Video camera, 6 ... Space, 7 ... Passerby

Claims (5)

A passer-behavior analysis system that analyzes the behavior of passers-by in a predetermined space,
A plurality of distance image acquisition means for acquiring a distance image;
Synchronization means for synchronizing distance image acquisition timings of the plurality of distance image acquisition means;
Coordinate system integration means for integrating the coordinate systems of a plurality of distance images respectively acquired by the distance image acquisition means into one coordinate system;
Position detecting means for determining the position coordinates of each passerby who passes through the space from the distance image obtained by integrating the coordinate system;
Tracking processing means for tracking the position of the same passerby to obtain temporal changes in the position coordinates of each passer;
First storage means for storing, for each passerby, data of temporal changes in position coordinates within a predetermined time;
Density calculating means for obtaining the density of each area from the number of passersby existing for each area obtained by dividing the space from the temporal change data of the position coordinates stored in the storage means;
Second storage means in which the density value and the number of distance image acquisition means are stored in association with each other;
Action analysis means for analyzing the behavior of a passer-by based on data of temporal changes in position coordinates within a predetermined time stored in the first storage means,
The coordinate system integration means reads the number of distance image acquisition means related to the density value obtained by the density calculation means from the second storage means, and uses the read number of distance image acquisition means. A passer-behavior analysis system that integrates the coordinate system of distance images obtained in this way. A passer-behavior analysis system that analyzes the behavior of passers-by in a predetermined space,
Position detecting means for determining the position coordinates of each passerby that passes through the space;
Tracking processing means for tracking the position of the same passerby to obtain temporal changes in the position coordinates of each passer;
Storage means for storing data of temporal changes in position coordinates within a predetermined time for each passer-by,
Means for obtaining the passage trajectory of each passerby in the space from the temporal change data of the position coordinates stored in the storage means;
A passer behavior analysis system comprising: a display unit that displays a trajectory of all passers-by who pass within a predetermined time in a predetermined space. A passer-behavior analysis system that analyzes the behavior of passers-by in a predetermined space,
Position detecting means for determining the position coordinates of each passerby that passes through the space;
Tracking processing means for tracking the position of the same passerby to obtain temporal changes in the position coordinates of each passer;
Storage means for storing data of temporal changes in position coordinates within a predetermined time for each passer-by,
Means for determining the passing speed of each passerby in the space from temporal change data of the position coordinates stored in the storage means;
A passer behavior analysis system comprising: display means for displaying a passing speed distribution of all passers-by who pass within a predetermined time in a predetermined space. A passer-behavior analysis system that analyzes the behavior of passers-by in a predetermined space,
Position detecting means for determining the position coordinates of each passerby that passes through the space;
Tracking processing means for tracking the position of the same passerby to obtain temporal changes in the position coordinates of each passer;
Storage means for storing data of temporal changes in position coordinates within a predetermined time for each passer-by,
Means for obtaining the passage trajectory of each passerby in the space from the temporal change data of the position coordinates stored in the storage means;
Means for obtaining the crossing angle formed by the approaching relative distance and the traveling direction vector of each passerby in the space from the trajectory;
A passer-behavior analysis system comprising: a display unit configured to display a position where the approach distance and the crossing angle within a predetermined time in a predetermined space are within a predetermined threshold range. A passer-behavior analysis system that analyzes the behavior of passers-by in a predetermined space,
Position detecting means for determining the position coordinates of each passerby that passes through the space;
Tracking processing means for tracking the position of the same passerby to obtain temporal changes in the position coordinates of each passer;
Storage means for storing data of temporal changes in position coordinates within a predetermined time for each passer-by,
Means for determining the number of passersby present in each area obtained by dividing the space into smaller parts from the temporal change data of the position coordinates stored in the storage means;
A passer behavior analysis system comprising: display means for displaying a density distribution of passersby who existed within a predetermined time in the region of a predetermined space.

Images