Exhibition layout-based conference participant flow prediction method and system

Abstract

The invention relates to the technical field of conference seat arrangement flow prediction, in particular to a conference participant flow prediction method and system based on exhibition layout, and the method comprises the following steps: obtaining passing track data, analyzing a seat arrangement interference path, screening conflict blocks, counting the number of times of behavior overlapping, and outputting a flow prediction result. According to the method, the dynamic change of the passing track in the range of the exhibition area is collected, structural parameters such as the boundary direction of the exhibition area and the seat distance are combined, the behavior pattern of path interruption and deviation is refined, and the path coincidence frequency and the passing time interval are counted, so that the identification of the high-frequency passing area and the tracking of the behavior conflict are realized; the method is advantaged in that lattice point density and direction matching is utilized to analyze people flow direction characteristics of a limited area in an exhibition area, staying and overlapping conditions of participants in different blocks are quantified, a local behavior aggregation effect is effectively identified, people flow distribution prediction precision and site use efficiency are improved, and passage organization and space utilization conditions under a complex seat arrangement structure are improved.

Description

Technical Field

[0001] This invention relates to the field of conference seating flow prediction technology, and in particular to a method and system for predicting attendee flow based on exhibition layout. Background Technology

[0002] The field of conference seating arrangement and traffic prediction technology involves utilizing data analysis and optimization methods, combined with computer science, artificial intelligence, and mathematical models, to study how to predict and optimize crowd flow and distribution based on the layout and seating arrangements of specific scenarios such as exhibitions and conferences. This includes predicting crowd flow based on factors such as seating plans in different scenarios, participant behavior characteristics, and flow patterns to ensure the smooth running of events and improve venue utilization and participant experience. Traditional exhibition layout and participant traffic prediction methods rely on analyzing participant flow patterns, regional distribution, and the attractiveness of exhibition areas, using simple rules or empirical models to predict participant distribution at the exhibition site. This typically depends on statistical analysis of historical data, establishing a basic crowd flow prediction model and combining it with the spatial layout of the exhibition venue to estimate participant flow. However, because these methods rely heavily on historical data and experience, they lack real-time dynamic adjustments and consideration of complex environmental changes, often failing to accurately predict traffic distribution under complex conditions.

[0003] In the current process of predicting the flow of attendees in exhibition layouts, the main reliance is on historical data and experience models. This makes it difficult to adapt to the dynamic changes in the flow of people at the exhibition site over time. The passage paths fail to reflect the real-time impact of seating arrangements and on-site spatial layout on attendees' behavior, resulting in model prediction lag and error accumulation. Areas with frequent path intersections cannot be identified in a timely manner, and it is difficult to accurately define attendees' hotspots and passage bottlenecks. The structural interference caused by seating arrangements on passage trajectories is not effectively captured, and the flow distribution lacks the ability to respond to changes in the scene, which may lead to problems such as local congestion, uneven resource allocation, and decreased on-site organization efficiency. Summary of the Invention

[0004] To address the technical problems existing in the prior art, this invention provides a method for predicting attendee traffic based on exhibition layout, comprising the following steps: To achieve the above objectives, the present invention adopts the following technical solution: a method for predicting the flow of attendees based on exhibition layout, comprising the following steps: S1: Obtain infrared counter and monitoring data at the exhibition site, detect the path continuity and turning frequency of attendees' passage trajectories within each exhibition area at different times, construct flow trajectory morphology labels, and generate a set of exhibition area passage path labels; S2: Based on the exhibition area access path label set, combined with the seat arrangement, seat aisle width and minimum distance data between seat groups and exhibition area boundaries, analyze the path deviation and turning interruption phenomena caused by the seating structure in each access path, and generate a set of seating structure interference paths. S3: Based on the set of interference paths of the seating structure, calculate the grid distribution density of each block in the current exhibition area, and combine the number of open boundaries of the exhibition area and the orientation parameters of the booth to filter out exhibition area areas with directional conflicts and generate a list of seating structure conflict blocks. S4: Based on the list of conflict blocks in the seating structure, the attendees whose paths overlap within the same passage segment are counted, frequently passing segments are marked, and the path numbers and time sequences corresponding to the frequently passing segments are organized to generate a seating structure interference behavior sequence dataset. S5: Based on the seating arrangement interference behavior sequence dataset, calculate the total number of times the participants' behavior overlaps in each block, record the total overlap statistics, and output the participant traffic prediction data.

[0005] As a further aspect of the present invention, during the process of marking frequently passing sections, the number of times the path overlaps within a unit time window and the time interval between two adjacent path overlaps are calculated. The channel segments corresponding to the number of times the path overlaps exceeds a preset overlap threshold and the time interval is less than a preset traffic rhythm threshold are marked as the frequently passing sections.

[0006] As a further aspect of the present invention, the exhibition area access path label set includes path continuity indicators, turning frequency levels, and path morphology types; the seating structure interference path set includes path offset segments, turning interruption segments, and seating restriction segments; the seating structure conflict block list includes directional conflict block numbers, conflict type labels, and layout grid coordinates; the seating structure interference behavior sequence dataset includes path number sequences, frequent passage segment sequences, and time sequence indexes; and the participant flow prediction data includes the total number of participant behavior overlaps within the block, path coverage density distribution data, and unit block flow intensity.

[0007] As a further aspect of the present invention, the step of obtaining the exhibition area access path label set is as follows: S111: Obtain infrared counter and monitoring data at the boundaries of each exhibition area, intersection passages and main entrance passages at the exhibition site, extract the passage trajectory data of participants at different time periods, synchronously verify the time series and spatial coordinate series of each trajectory, and classify the trajectory areas according to the exhibition area boundary parameters to generate a trajectory area distribution sequence set; S112: Based on the trajectory area distribution sequence set, the changes in labels of adjacent areas in the trajectory are traversed, the ratio of spatial spacing to time interval is calculated for each group of continuous coordinate points, and continuous trajectory segments with a ratio greater than the path continuity benchmark value are selected to generate a passage continuity marker sequence set. S113: Based on the continuous passage marker sequence set, extract the direction angle change value sequence of continuous trajectory segments, count the number of direction changes in each trajectory, aggregate the turning frequency distribution according to the exhibition area label, and generate an exhibition area passage path label set by combining the regional label order and direction change characteristics of the passage path.

[0008] As a further aspect of the present invention, the step of obtaining the interference path set of the seating structure is as follows: S211: Based on the exhibition area access path label set, construct a continuous coordinate sequence in the access path, and combine the seating arrangement of the corresponding area of ​​the exhibition area to extract the offset of the projection position of the path node in the row and column of the seat to obtain the access path offset matching matrix. S212: Based on the aforementioned path offset matching matrix, combined with the turning node sequence of the path in each exhibition area, extract the direction angle difference between path nodes, and combine the minimum spacing data between node coordinates and seat group edge coordinates to determine spatial interference at turning change points and obtain a turning interruption interference index set. S213: Based on the steering interruption interference index set, perform aggregation operation on the node sequence in the corresponding travel path and the interference mark in the offset matching matrix, split the path segments with steering interruption and position offset marks from the original label set into independent sequences, and establish a seating structure interference path set.

[0009] As a further aspect of the present invention, the step of obtaining the list of conflict blocks of the seating structure is as follows: S311: Based on the set of interference paths of the seating structure, extract the grid point coordinate sequence corresponding to each path in the exhibition area, count the number of grid points covered by the path in each block, calculate the distribution density value using the ratio of the number of grid points to the area, and generate the grid point density matrix of the exhibition area. S312: Based on the grid density matrix of the exhibition area, combined with the number of open boundaries of each exhibition area and the orientation parameter of the booth, the offset direction vector of the interference path is extracted, and the angle between the offset direction of the path and the grid direction vector of the exhibition area is calculated to obtain the path direction matching deviation sequence. S313: Based on the path direction matching deviation sequence, filter the grid point indexes with included angles greater than the set direction conflict threshold, and aggregate and organize the corresponding exhibition area numbers and grid point positions to establish a list of conflict blocks in the seating structure.

[0010] As a further aspect of the present invention, the step of obtaining the seating arrangement interference behavior sequence dataset is as follows: S411: Based on the list of conflict blocks in the seating structure, obtain the sequence of participant numbers and the sequence of path numbers in the corresponding channel segment. Perform a consistency comparison on the path numbers that fall within the same channel segment grid range in the same time window, mark the combination of participant numbers that have path space overlap, and establish an overlap event index for each combination to generate a channel path overlap event index set. S412: Based on the channel path overlap event index set, sort the overlap events by time according to the timestamp sequence, count the number of path overlap occurrences within a set unit time window, and calculate the time interval between two adjacent overlap events. When the number of path overlap occurrences exceeds the preset overlap number threshold and the time interval is less than the preset passage rhythm threshold, determine the corresponding channel segment as a frequently passing segment and obtain a frequently passing segment marker set. S413: Based on the frequently used passage segment marker set, reorganize the path number sequence and its time order within the marked passage segment, serialize the path number and time index, establish the temporal expression structure of path overlap behavior, and generate a seating structure interference behavior sequence dataset.

[0011] As a further aspect of the present invention, the step of obtaining the predicted attendance data is as follows: S511: Based on the seating arrangement interference behavior sequence dataset, extract the frequent passage segment number and dwell time label in the path of each participant, count the total number of dwell segments in the exhibition area grid of each path, and sum up the number of dwell segments of all path numbers in each block to generate a block path dwell frequency array. S512: Based on the block path dwell frequency array, aggregate all participant numbers in each block, determine the overlap of dwell time intervals of overlapping paths, and if there is a time overlap between the dwell time periods of two paths, accumulate the number of overlap occurrences, traverse according to the participant number dimension, and obtain the behavior overlap time coverage matrix. S513: Based on the behavior overlap time coverage number matrix, index and aggregate the number of attendees overlap in each block to establish the attendee traffic value sequence for each block, and output the data frame structure according to the block number to generate attendee traffic prediction data.

[0012] An exhibition layout-based system for predicting visitor flow includes: The path feature extraction module is used to perform S1: acquire infrared counter and monitoring data at the exhibition site, detect the path continuity and turning frequency of the participants' passage trajectory within each exhibition area at different time periods, construct flow trajectory morphology labels, and generate a set of exhibition area passage path labels; The seating arrangement interference label module is used to perform S2: Based on the exhibition area access path label set, combined with the seat arrangement method, the width of the aisle between seats, and the minimum distance data between the seat group and the exhibition area boundary, analyze the path deviation and turning interruption phenomenon caused by the seating arrangement structure in each access path, and generate a set of seating arrangement structure interference paths; The conflict block filtering module is used to perform S3: based on the set of interference paths of the seating structure, calculate the grid distribution density of each block in the current exhibition area, and combine the number of open boundaries of the exhibition area and the booth facing direction parameter to filter exhibition area areas with directional conflicts and generate a list of conflict blocks of the seating structure. The behavior sequence analysis module is used to execute S4: based on the list of conflict blocks in the seating structure, to count the attendees whose paths overlap in the same channel segment, to mark the frequently passing segments, and to organize the path numbers and time sequence corresponding to the frequently passing segments to generate a seating structure interference behavior sequence dataset. The traffic prediction module is used to execute S5: based on the seating arrangement interference behavior sequence dataset, calculate the total number of times the participants' behavior overlaps in each block, record the total number of overlap statistics, and output the participants' traffic prediction data.

[0013] Compared with the prior art, the advantages and positive effects of the present invention are as follows: By collecting dynamic changes in traffic trajectories within the exhibition area and combining structural parameters such as exhibition area boundary direction and seat spacing, the behavioral patterns of path interruption and deviation are refined. The frequency of path overlap and the time interval between passages are statistically analyzed to identify high-frequency passage areas and track behavioral conflicts. By using grid density and direction matching to analyze the flow characteristics of people in restricted areas within the exhibition area, the dwell time and overlap of participants in different blocks are quantified, effectively identifying local behavioral clustering effects, improving the prediction accuracy of people flow distribution and the efficiency of site use, and improving the traffic organization and space utilization under complex seating structures. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2 This is a flowchart of the process for obtaining the access path tag set in the exhibition area of ​​this invention; Figure 3 This is a flowchart of the process for obtaining the interference path set of the seating structure in this invention; Figure 4 This is a flowchart of the process for obtaining the list of conflict blocks in the seating arrangement structure of the present invention; Figure 5 This is a flowchart of the process for obtaining the dataset of interference behavior sequences of the seating arrangement structure in this invention; Figure 6 This is a flowchart of the process for obtaining data on the predicted flow of attendees in this invention. Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0016] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0017] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0018] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0019] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0020] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0021] Please see Figure 1 This invention provides a method for predicting visitor flow based on exhibition layout, comprising the following steps: S1: Obtain infrared counters and monitoring data at the boundaries of each exhibition area, intersections, and main entrances of the exhibition site; detect the path continuity and turning frequency of attendees' passage trajectories within each exhibition area at different times; construct flow trajectory morphology labels; and generate a set of exhibition area passage path labels. S2: Based on the exhibition area access path label set, combined with the seat arrangement method, the width of the aisle between seats, and the minimum distance between the seat group and the exhibition area boundary, analyze the path deviation and turning interruption phenomenon caused by the seating structure in each access path, and generate a set of seating structure interference paths. S3: Based on the set of interference paths of the seating structure, calculate the grid distribution density of each block in the current exhibition area. Combine the number of open boundaries of the exhibition area and the orientation parameters of the booth, perform structural matching between the path offset direction and the grid direction of the exhibition area layout, filter out exhibition area areas with directional conflicts, and generate a list of seating structure conflict blocks. S4: Based on the list of conflict blocks in the seating structure, the attendee numbers that overlapped in the same passage segment were counted, the number of times the paths overlapped within a unit time window and the time interval between two adjacent paths overlapped were calculated. When the number of paths overlapped exceeded the preset overlap threshold and the time interval was less than the preset passage rhythm threshold, the corresponding passage segment was determined to be a frequently passing segment. The path numbers and time order corresponding to the frequently passing segments were sorted to generate a seating structure interference behavior sequence dataset. S5: Based on the seating arrangement interference behavior sequence dataset, calculate the total number of times the behavior of attendees overlaps in each block according to the number of frequently passing sections in each attendee's path and the number of time periods the path stays in the exhibition area, record the total number of behavioral overlap statistics for each block, and output the attendee traffic prediction data.

[0022] The exhibition area access path label set includes path continuity indicators, turning frequency levels, and path morphology types. The seating structure interference path set includes path offset segments, turning interruption segments, and seating restriction segments. The seating structure conflict block list includes directional conflict block numbers, conflict type labels, and layout grid coordinates. The seating structure interference behavior sequence dataset includes path number sequences, frequent passage segment sequences, and time sequence indexes. The participant flow prediction data includes the total number of participant behavior overlaps within the block, path coverage density distribution data, and unit block flow intensity.

[0023] Please see Figure 2 The specific steps of S1 are as follows: S111: Obtain infrared counter and monitoring data at the boundaries of each exhibition area, intersection passages and main entrance passages at the exhibition site, extract the passage trajectory data of participants at different time periods, synchronously verify the time series and spatial coordinate series of each trajectory, and classify the trajectory areas according to the exhibition area boundary parameters to generate a trajectory area distribution sequence set; The system reads the parameters of the digital twin model of Zone S stored on the exhibition center server, retrieves the two-dimensional planar boundary coordinate sets of each exhibition area numbered Area_01 to Area_05, and simultaneously reads the number of people passing through the infrared counter (device ID: IR_G1) at the main entrance and the entry timestamp via the bus interface at a sampling frequency of 100ms. It also simultaneously accesses real-time video stream data from 50 high-definition surveillance cameras (device IDs: CAM_001 to CAM_050) covering the entire venue. Using computer vision algorithms, it extracts the top coordinates of all attendees from the video stream, constructs an initial coordinate sequence, selects attendee ID_2025 as the tracking target, and extracts their original trajectory data from the time period between 10:00:00 and 10:10:00. This original trajectory contains 6000 discrete coordinate points. The system executes a time and space synchronization verification program, setting the trigger time of the infrared counter... Coordinates and time of monitoring and identification Alignment is performed if the time difference between the two is... If the time difference is less than the set synchronization tolerance threshold of 0.05 seconds, the data is considered valid, and then the preset exhibition area boundary parameters are invoked. For example, the boundary range of Area_01 exhibition area is defined as Miqie Meters, the coordinate sequence of participant ID_2025 Each point is compared with the boundary range; when a coordinate point is detected... When the geometric constraints of Area_01 are satisfied, the coordinate point is marked as belonging to Area_01, and all the classified coordinate points are connected in chronological order to generate a trajectory area distribution sequence set containing spatial location and area label.

[0024] S112: Based on the trajectory area distribution sequence set, the changes in labels of adjacent areas in the trajectory are traversed, the ratio of spatial spacing to time interval is calculated for each group of continuous coordinate points, and continuous trajectory segments with a ratio greater than the path continuity benchmark value are selected to generate a passage continuity label sequence set. Retrieve data from the trajectory region distribution sequence set, specifically the trajectory sequence for participant ID_2025. Perform a traversal operation to identify the node indices where the area label changes, such as the moment when switching from Area_01 to Area_02. Extract the five adjacent coordinate points before and after this moment as the analysis window, and calculate the values ​​of adjacent points. and Euclidean distance between and time difference Set path continuity benchmark value The value is 0.2 meters per second. This value is set based on the minimum moving speed of a normal adult viewing an exhibition at a slow pace. The ratio is calculated accordingly. If the calculation yields a certain segment meters per second Furthermore, the segment is determined to be a valid continuous movement segment, and a pass continuity marker sequence set is generated.

[0025] S113: Based on the continuous track marker sequence set, extract the direction angle change value sequence of continuous track segments, count the number of direction changes in each track, aggregate the turning frequency distribution according to the exhibition area label, and generate the exhibition area track label set by combining the regional label order and direction change characteristics of the track. Call the continuity marker sequence set to extract the coordinate sequence of participant ID_2025 in the continuous trajectory segment, and for each coordinate point... Calculate its instantaneous movement vector Then, the direction angle is obtained using the arctangent function. Construct a sequence of direction angle changes along the entire path and calculate the direction deflection angle between adjacent vectors. Set the directional change threshold in the path continuity benchmark. (This value is based on the limit setting for smooth turning in a normal walking streamline), and the sequence is traversed to count the number of tracks that satisfy the condition. The system tracks the number of sudden changes. For example, during the time period in Area_01, it detected 5 sharp turns greater than 45 degrees by attendees, while in Area_02, only 1 such turn occurred. The system aggregates this frequency data by area label, forming a feature record of "Area_01: 5 times, Area_02: 1 time". Simultaneously, it extracts the temporal sequence of trajectories crossing each area (e.g., Sequence: Area_01). Area_02 Area_05) integrates this spatiotemporal sequence with the turning behavior features in each area to generate a set of exhibition area access path labels that includes trajectory flow structure and local behavior features.

[0026] Based on the exhibition area access path label set, the continuous coordinate sequence of attendee ID_2025 within Area_01 was parsed out. Then, the seating arrangement parameters for that area were retrieved from the digital twin model of Area S, resulting in the following arrangement: The matrix distribution is set with row and column spacing as follows: Rice and Meters define a local grid coordinate system, which represents each physical coordinate point in the path. Project onto the seat row and column grid lines, and calculate the vertical distance from the node to the nearest seat row. Vertical distance from the column line For example, the coordinates of a certain node are The nearest seat center is The offsets are respectively rice, Meters, using formula Extract the overall offset here. Perform this operation on all nodes along the path to generate a travel path offset matching matrix that reflects the degree of deviation of the trajectory from the fixed seating layout.

[0027] S212: Based on the path offset matching matrix, combined with the turning node sequence of the path in each exhibition area, the direction angle difference between path nodes is extracted, and combined with the minimum spacing data between the node coordinates and the seat group edge coordinates, spatial interference judgment is performed on the turning change points to obtain the turning interruption interference index set. Based on the path offset matching matrix and the turning node sequence extracted from S113, the indices of nodes in the path where abrupt changes in direction are identified. Read the direction angle difference at this node. (e.g., 90°), and simultaneously retrieve the coordinates of the node and the coordinates of the nearest seat group edge. Spatial relationships, calculate minimum spacing Set a threshold for spatial interference judgment Meters (based on the minimum comfortable width for an adult to pass sideways), if calculated Miqie This means that a sharp turn of more than 60 degrees occurred at a distance of less than 0.6 meters from the seat. The system determined that this point was affected by physical interference from the seating structure, resulting in an unnatural sharp turn. The system then recorded this turning point as... Mark them as interference points, traverse the entire path to obtain the indexes of all such points, and generate a steering interruption interference index set.

[0028] S213: Based on the steering interruption interference index set, perform aggregation operations on the node sequence in the corresponding travel path and the interference markers in the offset matching matrix, and split the path segments with steering interruption and position offset markers from the original label set into independent sequences to establish a seating structure interference path set. Based on the turning interruption interference index set, the interference point location is located in the original travel path label set of participant ID_2025. The sequence segmentation operation is performed to separate the path segment containing continuous interference indices (such as indices 100 to 105) from the original complete trajectory, retaining the smooth passage segments in the original path. The separated segments are marked as "seat structure interference segments", and a unique interference event ID (such as Int_Path_001) is assigned to this independent sequence. This sequence retains the precise coordinates, timestamps and offset information when the interference occurred. This segmentation and marking process is performed on the trajectory of all participants, and all the interfered path segments are collected to establish a set of seat structure interference paths.

[0029] Please see Figure 4The specific steps of S3 are as follows: S311: Based on the set of interference paths of the seating structure, extract the grid point coordinate sequence corresponding to each path in the exhibition area, and count the number of grid points covered by the path in each block. Use the ratio of the number of grid points to the area to calculate the distribution density value and generate the grid point density matrix of the exhibition area. The set of interference paths for the seating arrangement structure is invoked to divide the exhibition area Area_01 into... Using a standard cell grid, extract the grid point coordinate sequence of each interference path segment within the exhibition area, and statistically analyze each grid cell. Number of path points falling inside For example, grid Covered by 3 different paths, containing a total of 50 trajectory points, with a known grid area. Using the density formula Calculate the distribution density value, and obtain point / Traverse all grids to generate a grid density matrix that reflects the spatial aggregation of disturbance behavior.

[0030] S312: Based on the grid density matrix of the exhibition area, combined with the number of open boundaries of each exhibition area and the orientation parameter of the booth, the offset direction vector of the interference path is extracted, and the angle between the offset direction of the path and the grid direction vector of the exhibition area is calculated to obtain the path direction matching deviation sequence. Based on the grid density matrix of the exhibition area, high-density grid regions are identified, combined with the number of open boundaries of the exhibition area (e.g., ...). ) and the direction vector of the booth facing (Northward) Extract the average offset direction vector of the interference path falling into this area. (like (Eastward), the angle between the path offset direction and the direction of the exhibition area planning grid point (i.e., the booth orientation) is calculated using the vector angle formula. Substituting the numerical values, we can see that this angle quantifies the degree of conflict between the actual pedestrian flow and the designed flow line. This operation is performed on all high-density areas to obtain the path direction matching deviation sequence.

[0031] S313: Based on the path direction matching deviation sequence, filter the grid point indexes with included angles greater than the set direction conflict threshold, and aggregate and organize the exhibition area number and grid point position to establish a list of conflict blocks in the seating structure. Based on the path direction matching deviation sequence, set the direction conflict threshold. (Indicating that the streamlines are nearly perpendicular to or opposite to the layout), filter out the included angles. Grid indexes, such as those mentioned above grid The exhibition area number Area_01 to which the grid belongs is aggregated with the grid point location coordinates and marked as a "structural conflict block". This indicates that the seating arrangement in this area seriously hinders the natural passage of the participants, forcing them to make a large lateral shift. All the selected block information is sorted out to establish a list of seating structure conflict blocks.

[0032] Please see Figure 5 The specific steps of S4 are as follows: S411: Based on the list of conflict blocks in the seating structure, obtain the sequence of participant numbers and the sequence of path numbers in the corresponding channel segment. Perform consistency comparison on the path numbers that fall within the same channel segment grid range in the same time window, mark the combination of participant numbers that have path space overlap, and establish an overlap event index for each combination to generate a channel path overlap event index set. As shown in Table 1, based on the list of conflict blocks in the seating arrangement, the physical channel segment Channel_A is locked, and the sequence of participant numbers within the corresponding grid range of this channel is obtained. With the path number sequence, a time synchronization window of 1 second is set for points falling into the same grid within the same time window. Perform a consistency comparison on the path, and if a match is detected... and Euclidean distance of path coordinates Meters, determine if path space overlap occurs, and mark the combination. And establish an index of overlapping events. Record the precise time and duration of the overlap, and generate an index set of channel path overlap events.

[0033] Table 1 Example of Channel Path Overlap Detection Data

[0034] S412: Based on the channel path overlap event index set, sort the overlap events by time according to the timestamp sequence, count the number of path overlaps within a set unit time window, and calculate the time interval between two adjacent overlap events. When the number of path overlaps exceeds the preset overlap number threshold and the time interval is less than the preset passage rhythm threshold, determine the corresponding channel segment as a frequently passing segment and obtain a frequently passing segment marker set. Based on the channel path overlap event index set, the overlap events are sorted by timestamp within a set unit time window. The total number of times the paths overlap within a second. For example, if 12 coincidences are recorded between 10:30:00 and 10:31:00, calculate the time interval between two adjacent coincidence events. Set a threshold for the number of overlaps. The second and passing time threshold seconds, if the statistical results satisfy And average interval The number of seconds indicates that the pedestrian flow within the passage segment is extremely frequent and continuous, thus determining the corresponding passage segment as a frequently passing section and obtaining a frequently passing section marker set.

[0035] S413: Based on the frequently used segment marker set, reorganize the path number sequence and its time order within the marked channel segment, serialize the path number and time index, establish the temporal expression structure of path overlap behavior, and generate a seating arrangement interference behavior sequence dataset. Based on the frequently used segment marker set, extract the path number sequence within the marked channel segment Channel_A, and serialize and reassemble the path number with its time index for passing through the region to construct a sequence such as Sequence= The temporal structure records in detail the process of congestion, that is, how the paths of different participants gradually overlap in the time dimension and lead to the tension of spatial resources. This structure is used as a temporal expression of the interference of seating arrangement on the flow of people, and a dataset of seating arrangement interference behavior sequence is generated.

[0036] Please see Figure 6 The specific steps of S5 are as follows: S511: Based on the seating arrangement interference behavior sequence dataset, extract the frequent passage segment number and dwell time label in the path of each participant, count the total number of dwell segments in the exhibition area grid of each path, and sum up the number of dwell segments of all path numbers in each block to generate a block path dwell frequency array. Based on the seating arrangement interference behavior sequence dataset, for the path of participant ID_2025, their movement speed data in frequently passing sections was extracted, and speed was filtered. The time period is marked as "stay time period", and the path is counted at the grid points in the exhibition area. Total number of time fragments within For example, if a participant stops at this location twice, the number of stops on all paths within that block is accumulated. If a total of 50 people stay here, the cumulative value is 120 times, generating a block path dwell frequency array that reflects the dwell stickiness of each block.

[0037] S512: Based on the block path dwell frequency array, aggregate all participant IDs in each block, and determine the overlap of dwell time intervals for overlapping paths. If the dwell time periods of two paths overlap, accumulate the number of overlap occurrences, and iterate along the participant ID dimension to obtain a behavior overlap time coverage matrix. Based on the block path dwell frequency array, aggregate all participant IDs within grid G10,5, and extract the dwell time intervals [Tu,start,Tu,end] and [Tv,start,Tv,end] for any two participants u and v. Determine whether the time intervals overlap, i.e., (Tu,start,Tu,end) ∩ (Tv,start,Tv,end) ≠ ∅. If there is an overlap, it is determined as one overlap, and the number of overlap occurrences is accumulated. Construct an N×N behavior overlap time coverage matrix along the participant ID dimension. The element Muv in the matrix represents the frequency of participants u and v staying at the same location. This matrix reveals the degree of group congestion caused by unreasonable seating arrangements.

[0038] S513: Based on the behavior overlap time coverage matrix, index and aggregate the number of attendees overlap in each block to establish a numerical sequence of attendee traffic for each block, and output the data frame structure according to the block number to generate attendee traffic prediction data. Based on the behavior overlap time coverage matrix, index aggregation statistics are performed on the submatrix corresponding to each block, and the sum of each column is calculated. This value represents the congestion pressure faced by participant V in this area. By summing and normalizing the pressure values ​​of all participants within the block, a sequence of participant flow values ​​for that block is obtained. For example, if the calculated flow pressure index of Channel_A is 85.5, the data frame structure containing coordinates and pressure values ​​is output according to the block number. This data frame can be directly used to predict the flow change trend after the future seating arrangement is adjusted, and generate the flow prediction data of the attendees.

[0039] Please see Figure 7 A system for predicting visitor flow based on exhibition layout, comprising: The path feature extraction module is used to perform S1: acquire infrared counter and monitoring data at the exhibition site, detect the path continuity and turning frequency of the participants' passage trajectory within each exhibition area at different time periods, construct flow trajectory morphology labels, and generate a set of exhibition area passage path labels; The seating arrangement interference label module is used to perform S2: Based on the exhibition area access path label set, combined with the seat arrangement method, the width of the aisle between seats, and the minimum distance data between the seat group and the exhibition area boundary, it analyzes the path deviation and turning interruption phenomenon caused by the seating arrangement structure in each access path, and generates a set of seating arrangement structure interference paths. The conflict block filtering module is used to execute S3: based on the set of interference paths of the seating structure, calculate the grid distribution density of each block in the current exhibition area, and combine the number of open boundaries of the exhibition area and the booth facing direction parameter to filter exhibition area areas with directional conflicts and generate a list of conflict blocks of the seating structure. The behavior sequence analysis module is used to execute S4: based on the list of conflict blocks in the seating structure, to count the attendees whose paths overlap in the same channel segment, to mark frequently passing segments, and to organize the path numbers and time sequence corresponding to the frequently passing segments to generate a seating structure interference behavior sequence dataset. The traffic prediction module is used to execute S5: based on the seating arrangement interference behavior sequence dataset, calculate the total number of times the participants' behavior overlaps in each block, record the total overlap statistics, and output the participants' traffic prediction data.

[0040] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for predicting attendee flow based on exhibition layout, characterized in that, Includes the following steps: S1: Obtain infrared counter and monitoring data at the exhibition site, detect the path continuity and turning frequency of attendees' passage trajectories within each exhibition area at different times, construct flow trajectory morphology labels, and generate a set of exhibition area passage path labels; S2: Based on the exhibition area access path label set, combined with the seat arrangement, seat aisle width and minimum distance data between seat groups and exhibition area boundaries, analyze the path deviation and turning interruption phenomena caused by the seating structure in each access path, and generate a set of seating structure interference paths. S3: Based on the set of interference paths of the seating structure, calculate the grid distribution density of each block in the current exhibition area, and combine the number of open boundaries of the exhibition area and the orientation parameters of the booth to filter out exhibition area areas with directional conflicts and generate a list of seating structure conflict blocks. S4: Based on the list of conflict blocks in the seating structure, the attendees whose paths overlap within the same passage segment are counted, frequently passing segments are marked, and the path numbers and time sequences corresponding to the frequently passing segments are organized to generate a seating structure interference behavior sequence dataset. S5: Based on the seating arrangement interference behavior sequence dataset, calculate the total number of times the participants' behavior overlaps in each block, record the total overlap statistics, and output the participant traffic prediction data.

2. The method for predicting attendee flow based on exhibition layout according to claim 1, characterized in that: During the process of marking frequently passing sections, the number of times the path overlaps within a unit time window and the time interval between two adjacent path overlaps are calculated. The channel segments corresponding to the number of times the path overlaps exceeds a preset overlap threshold and the time interval is less than a preset traffic rhythm threshold are marked as the frequently passing sections.

3. The method for predicting attendee flow based on exhibition layout according to claim 1, characterized in that: The exhibition area access path label set includes path continuity indicators, turning frequency levels, and path morphology types. The seating structure interference path set includes path offset segments, turning interruption segments, and seating restriction segments. The seating structure conflict block list includes directional conflict block numbers, conflict type labels, and layout grid coordinates. The seating structure interference behavior sequence dataset includes path number sequences, frequent passage segment sequences, and time sequence indexes. The participant flow prediction data includes the total number of participant behavior overlaps within the block, path coverage density distribution data, and unit block flow intensity.

4. The method for predicting attendee flow based on exhibition layout according to claim 1, characterized in that, The steps for obtaining the exhibition area access path tag set are as follows: S111: Obtain infrared counter and monitoring data at the boundaries of each exhibition area, intersection passages and main entrance passages at the exhibition site, extract the passage trajectory data of participants at different time periods, synchronously verify the time series and spatial coordinate series of each trajectory, and classify the trajectory areas according to the exhibition area boundary parameters to generate a trajectory area distribution sequence set; S112: Based on the trajectory area distribution sequence set, the changes in labels of adjacent areas in the trajectory are traversed, the ratio of spatial spacing to time interval is calculated for each group of continuous coordinate points, and continuous trajectory segments with a ratio greater than the path continuity benchmark value are selected to generate a passage continuity marker sequence set. S113: Based on the continuous passage marker sequence set, extract the direction angle change value sequence of continuous trajectory segments, count the number of direction changes in each trajectory, aggregate the turning frequency distribution according to the exhibition area label, and generate an exhibition area passage path label set by combining the regional label order and direction change characteristics of the passage path.

5. The method for predicting attendee flow based on exhibition layout according to claim 1, characterized in that, The steps for obtaining the interference path set of the seating structure are as follows: S211: Based on the exhibition area access path label set, construct a continuous coordinate sequence in the access path, and combine the seating arrangement of the corresponding area of ​​the exhibition area to extract the offset of the projection position of the path node in the row and column of the seat to obtain the access path offset matching matrix. S212: Based on the aforementioned path offset matching matrix, combined with the turning node sequence of the path in each exhibition area, extract the direction angle difference between path nodes, and combine the minimum spacing data between node coordinates and seat group edge coordinates to determine spatial interference at turning change points and obtain a turning interruption interference index set. S213: Based on the steering interruption interference index set, perform aggregation operation on the node sequence in the corresponding travel path and the interference mark in the offset matching matrix, split the path segments with steering interruption and position offset marks from the original label set into independent sequences, and establish a seating structure interference path set.

6. The method for predicting attendee flow based on exhibition layout according to claim 1, characterized in that, The steps for obtaining the list of conflict blocks in the seating structure are as follows: S311: Based on the set of interference paths of the seating structure, extract the grid point coordinate sequence corresponding to each path in the exhibition area, count the number of grid points covered by the path in each block, calculate the distribution density value using the ratio of the number of grid points to the area, and generate the grid point density matrix of the exhibition area. S312: Based on the grid density matrix of the exhibition area, combined with the number of open boundaries of each exhibition area and the orientation parameter of the booth, the offset direction vector of the interference path is extracted, and the angle between the offset direction of the path and the grid direction vector of the exhibition area is calculated to obtain the path direction matching deviation sequence. S313: Based on the path direction matching deviation sequence, filter the grid point indexes with included angles greater than the set direction conflict threshold, and aggregate and organize the corresponding exhibition area numbers and grid point positions to establish a list of conflict blocks in the seating structure.

7. The method for predicting attendee flow based on exhibition layout according to claim 1, characterized in that, The steps for obtaining the dataset of interference behavior sequences of the seating arrangement structure are as follows: S411: Based on the list of conflict blocks in the seating structure, obtain the sequence of participant numbers and the sequence of path numbers in the corresponding channel segment. Perform a consistency comparison on the path numbers that fall within the same channel segment grid range in the same time window, mark the combination of participant numbers that have path space overlap, and establish an overlap event index for each combination to generate a channel path overlap event index set. S412: Based on the channel path overlap event index set, sort the overlap events by time according to the timestamp sequence, count the number of path overlap occurrences within a set unit time window, and calculate the time interval between two adjacent overlap events. When the number of path overlap occurrences exceeds the preset overlap number threshold and the time interval is less than the preset passage rhythm threshold, determine the corresponding channel segment as a frequently passing segment and obtain a frequently passing segment marker set. S413: Based on the frequently used passage segment marker set, reorganize the path number sequence and its time order within the marked passage segment, serialize the path number and time index, establish the temporal expression structure of path overlap behavior, and generate a seating structure interference behavior sequence dataset.

8. The method for predicting attendee flow based on exhibition layout according to claim 1, characterized in that, The steps for obtaining the predicted attendee flow data are as follows: S511: Based on the seating arrangement interference behavior sequence dataset, extract the frequent passage segment number and dwell time label in the path of each participant, count the total number of dwell segments in the exhibition area grid of each path, and sum up the number of dwell segments of all path numbers in each block to generate a block path dwell frequency array. S512: Based on the block path dwell frequency array, aggregate all participant numbers in each block, determine the overlap of dwell time intervals of overlapping paths, and if there is a time overlap between the dwell time periods of two paths, accumulate the number of overlap occurrences, traverse according to the participant number dimension, and obtain the behavior overlap time coverage matrix. S513: Based on the behavior overlap time coverage number matrix, index and aggregate the number of attendees overlap in each block to establish the attendee traffic value sequence for each block, and output the data frame structure according to the block number to generate attendee traffic prediction data.

9. A system for predicting visitor flow based on exhibition layout, characterized in that, The system is used to implement the attendee flow prediction method based on exhibition layout as described in any one of claims 1-8, and the system includes: The path feature extraction module is used to perform S1: acquire infrared counter and monitoring data at the exhibition site, detect the path continuity and turning frequency of the participants' passage trajectory within each exhibition area at different time periods, construct flow trajectory morphology labels, and generate a set of exhibition area passage path labels; The seating arrangement interference label module is used to perform S2: Based on the exhibition area access path label set, combined with the seat arrangement method, the width of the aisle between seats, and the minimum distance data between the seat group and the exhibition area boundary, analyze the path deviation and turning interruption phenomenon caused by the seating arrangement structure in each access path, and generate a set of seating arrangement structure interference paths; The conflict block filtering module is used to perform S3: based on the set of interference paths of the seating structure, calculate the grid distribution density of each block in the current exhibition area, and combine the number of open boundaries of the exhibition area and the booth facing direction parameter to filter exhibition area areas with directional conflicts and generate a list of conflict blocks of the seating structure. The behavior sequence analysis module is used to execute S4: based on the list of conflict blocks in the seating structure, to count the attendees whose paths overlap in the same channel segment, to mark the frequently passing segments, and to organize the path numbers and time sequence corresponding to the frequently passing segments to generate a seating structure interference behavior sequence dataset. The traffic prediction module is used to execute S5: based on the seating arrangement interference behavior sequence dataset, calculate the total number of times the participants' behavior overlaps in each block, record the total number of overlap statistics, and output the participants' traffic prediction data.

Images