Environmental art immersive collaborative design system based on multi-source data fusion

Abstract

The invention discloses an environment art immersive collaborative design system based on multi-source data fusion, belongs to the technical field of computer data processing, and particularly relates to an immersive display system based on data processing and intelligent control. According to the system, real-time positions, flowing speeds and staying states of audiences are collected through the people flow interaction behavior recognition module, a display space is divided into a plurality of subareas, and space-time corresponding control over light and shadow and a sound field is achieved; a three-dimensional illumination matrix is established, and local shadow and brightness attenuation are compensated in combination with illumination attenuation dynamic calculation and an adaptive enhancement mechanism; the sound environment analysis module constructs a three-dimensional sound pressure distribution model, and realizes sound field equalization and local compensation according to sound absorption attenuation calculation and a sound pressure correction strategy. The system realizes synchronous response of light, sound and audience behaviors through multi-source data fusion and dynamic feedback adjustment, improves the space immersion, reduces the no-load energy consumption of equipment, and is suitable for various scenes such as night games in scenic spots, garden display, cultural performance and the like.

Description

Technical Field

[0001] This invention relates to the field of computer data processing technology, specifically to an immersive collaborative design system for environmental art that integrates multi-source data. Background Technology

[0002] In recent years, with the rapid integration of digital media art and smart scenic area construction, immersive displays based on multi-sensory interaction have been widely used in theme parks, cultural scenic areas, and garden night tours. Unlike traditional theaters or indoor fixed stage performance spaces, immersive stages in scenic areas often adopt a mobile display structure, allowing the audience to interact in real time with light, shadow, sound effects, installations, and landscape elements as they move, pass through, or stop, presenting a dynamic and immersive viewing experience.

[0003] However, existing immersive stage systems mostly rely on static preset models and fixed control strategies, setting up scenes solely based on lighting, sound, and visual rendering parameters, which cannot effectively cope with the dynamic changes in audience flow in space and time. Specifically: In open-air exhibitions, audience density, movement speed, direction of travel, and dwell time significantly affect local lighting distribution, sound wave propagation, and reflection characteristics. However, existing systems cannot collect this dynamic data in real time and lack mechanisms to adjust lighting or sound fields based on audience behavior. Because lighting and sound effects control are still based on fixed preset parameters, local areas are prone to brightness attenuation, shadow coverage, or uneven sound pressure, leading to visual focus shift and sound field imbalance, thus reducing the immersive experience. Existing systems fail to integrate and analyze multi-source information such as audience behavior, lighting data, sound field data, and scene device status in real time, and also lack dynamic adjustment strategies based on multi-physics interaction, limiting the immersive experience. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an immersive collaborative design system for environmental art that integrates multi-source data, in order to solve the problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a multi-source data fusion-based immersive collaborative design system for environmental art, comprising: The pedestrian interaction behavior recognition module is used to divide the scenic area or open exhibition space into several stage sub-areas according to the terrain, traffic flow, and viewing path. Each sub-area corresponds to an independent light and shadow control domain and sound field control domain. It also collects the real-time location set of the audience in the i-th sub-area at time t and sets the entrance boundary for each sub-area. and export border Construct the number of new entrants entering the entrance boundary within time t. and the speed of the flow of people This is used to identify whether a command to activate the lighting and sound field devices has been triggered, and to construct the number of people leaving the exit boundary within time t. and leaving ratio To identify whether a command to shut down lighting and sound field devices has been triggered; The dynamic lighting analysis module is used to collect lighting datasets from sub-regions and establish a three-dimensional lighting matrix. The three-dimensional lighting matrix consists of the lighting intensity at the three-dimensional spatial position (x, y, z) corresponding to each voxel. Composition; and construction of the illumination attenuation coefficient of each voxel in the i-th sub-region at time t. To assess and identify voxels with anomalous light attenuation, and to evaluate the light intensity of voxels with anomalous light attenuation. Make corrections; The acoustic environment analysis module is used to collect sound field datasets in sub-regions, establish a three-dimensional sound field distribution model based on the sound energy propagation model, and construct the sound pressure intensity at the three-dimensional spatial location corresponding to each voxel. And combined with the sound absorption coefficient of the r-th reflecting surface in the path from the sound source to each voxel. and crowd sound absorption coefficient The sound absorption attenuation coefficient of each voxel is jointly constructed. To assess and identify anomalous voxels affecting sound absorption, and to evaluate the sound pressure intensity of these anomalous voxels. Make corrections.

[0006] Preferably, the human interaction behavior recognition module includes a partitioning unit, a video frame acquisition unit, and a dynamic trigger sub-region control unit; The partition unit is used to divide the scenic area or open exhibition space into several stage sub-areas according to the terrain, circulation routes and viewing paths, i=1,2,...,Q; Q represents the total number of sub-areas; each sub-area i corresponds to an independent light and shadow control domain and sound field control domain; The video frame acquisition unit is used to deploy visual recognition devices in each sub-region to acquire video frame sequences of the audience and identify the individual positions of the audience using YOLO or OpenPose object detection algorithms, thereby obtaining the set of real-time positions of the audience in the i-th sub-region at time t.

[0007] in, Real-time number of users in the sub-region; Let be the two-dimensional coordinates of the k-th viewer in the i-th sub-region; These represent the positions of the k-th viewer in the i-th sub-region in the horizontal x-direction and the vertical y-direction, respectively.

[0008] Preferably, the dynamic trigger sub-zone control unit is used to set the entry boundary in each sub-zone. and export border This is used for triggering decisions in the light and shadow control domain and the sound field control domain. The specific steps are as follows: Let the boundary of the entrance to the i-th sub-region be... Identify the number of new entrants who enter the entrance boundary within time t. ; The sliding window method is used, by sliding the time window. The positions of adjacent video frames are temporally differencing to calculate the average displacement vector of the pedestrian flow in the i-th sub-region. ; Decomposed into average displacement components of the population and the average vertical displacement component of the crowd ; The vector synthesis of the two displacement components yields the magnitude of the pedestrian velocity in this sub-region. ; Set threshold for new hires and movement speed threshold ,when or At that time, the command to start the lighting and sound field devices is triggered.

[0009] Preferably, the pedestrian interaction behavior recognition module further includes a second trigger sub-area control unit and a flow path sequence generation unit; The second trigger sub-zone control unit is used to record the audience dwell time from the start time of the light and sound field equipment to the current time within the i-th sub-zone, calculate the occupancy time ratio of the current sub-zone, and obtain the occupancy time ratio. ; Let the exit boundary of the i-th sub-region be... Identify the number of people who leave the exit boundary within time t. ; Calculate the number of people leaving the exit boundary within time t. The departure ratio is obtained by comparing the number of people who enter with the total number of people who leave. ; Preset stay threshold With departure ratio threshold ,when or The second trigger sub-zone control unit outputs a command to shut down the light and sound field devices; Generate flow path sequence units, used to generate flow path sequence units at the center point of each sub-region. As path nodes, sub-regions with a "closed trigger" status are identified and marked as flow endpoint nodes, arranged according to the order of their closed trigger status to form a flow path sequence: If multiple sub-regions are closed within the same time window, they are arranged from highest to lowest according to their departure ratios → the main flow path takes priority.

[0010] Preferably, the dynamic light and shadow analysis module includes a light parameter acquisition unit and a light distribution modeling unit; The lighting parameter acquisition unit is used to acquire multi-source lighting parameters of the stage scene in the sub-area in real time through a scene sensor array, photometering equipment, and ambient light sensor, including the position of the main light source. Light intensity at the location Auxiliary light source in position intensity Angle between auxiliary light source and surface normal and material reflectivity The multi-source illumination parameters are discretized into a spatial raster format to form an illumination dataset; The illumination distribution modeling unit is used to establish and train the illumination energy balance model based on the illumination radiation transfer model. The stage scene in the i-th sub-region is divided into m1×n1×h1 voxel units according to the actual three-dimensional coordinates. The coordinates (x,y,z) of each voxel correspond to a set of illumination datasets. m1 represents the number of voxels in the horizontal direction; n1 represents the number of voxels in the vertical direction; h1 represents the number of voxels in the vertical direction, i.e., the z-axis direction; Lighting datasets from different spatial points are acquired using multi-angle lighting sensors, cameras, and brightness probes installed at the top, sides, and front of the stage. These non-uniformly distributed lighting datasets are then mapped to corresponding voxel coordinate systems. Bilinear or Gaussian interpolation algorithms are used to spatially smooth and complete the lighting values ​​in the missing areas, forming a three-dimensional lighting matrix. This three-dimensional lighting matrix consists of the lighting intensity at each voxel's corresponding three-dimensional spatial position (x, y, z). The composition is the result of the fusion of main light source, auxiliary light, and reflected brightness factors.

[0011] Preferably, the light and shadow dynamic analysis module also includes a light attenuation dynamic calculation unit; The dynamic illumination attenuation calculation unit is used to dynamically calculate the illumination attenuation coefficient of each voxel in the i-th sub-region at time t, based on the change in illumination intensity at the corresponding three-dimensional spatial position of each voxel and the occlusion caused by crowds. ; The specific steps include: To quantify the gradual weakening of light due to scattering, absorption, and shading during its propagation through the air, and the variable shading areas formed by the movement and gathering of people, resulting in a local decrease in light intensity, the area of ​​the shading portion within the voxel region of the sub-region is divided. Total area of ​​sub-regions The ratio is calculated to obtain the occlusion percentage. ; Based on the average displacement component of the population and the average vertical displacement component of the crowd Further calculations Macro-dominant direction of the population in i sub-regions ; Identify the macro-dominant direction angle of the population This represents the macroscopic flow direction of people in the sub-region on a two-dimensional plane at time t, including: if and This indicates that the crowd is moving horizontally; if and This indicates that the crowd is moving vertically. Extract the cosine value of the incident angle of the light source. The dominant direction angle of the macro-flow of the population in the i-th sub-region : Calculate and obtain the occlusion angle ; like This indicates that the direction of crowd movement is the same as the direction of illumination, and there is a risk of light source attenuation effect within the sub-area due to crowd occlusion; if This indicates that the direction of crowd movement is opposite to the direction of light illumination, and there is no risk of light source attenuation effect in the sub-area due to crowd occlusion. Based on the percentage of occlusion and the angle of obstruction The coupling is performed to calculate the illumination attenuation coefficient of each voxel in the i-th sub-region at time t. .

[0012] Preferably, the light and shadow dynamic analysis module further includes a first correction unit, which is used to set the light attenuation threshold. If the illumination attenuation coefficient of each voxel in the i-th sub-region at time t This indicates abnormal light attenuation, which may cause localized shadows on actors or stage objects, potentially affecting the audience's visual focus. In this case, the light enhancement ratio is calculated for the voxel with abnormal light attenuation. and increase the proportion of light. Illumination intensity applied to each voxel at its corresponding three-dimensional spatial location Make corrections to obtain the corrected light intensity. ; During the adjustment process, the main light source, auxiliary light source, and light direction are adjusted so that the light intensity of each voxel at its corresponding 3D spatial position is updated after the light is projected onto the shadow area, until the corrected light intensity is achieved. Until the value is reached.

[0013] Preferably, the acoustic environment analysis module includes a sound field parameter acquisition unit and a sound field modeling unit; The sound field parameter acquisition unit is used to acquire the total number of sound sources on the stage within the sub-area. The sound power of the a-th sound source The propagation path length from the a-th sound source to the corresponding three-dimensional spatial position (x, y, z) of each voxel. Establish a sound field dataset and convert the collected sound field dataset into a spatial raster data format so that it can be fused and analyzed with the illumination distribution matrix in the same coordinate system; The sound field modeling unit is used to establish a three-dimensional sound field distribution model based on the sound energy propagation model. The stage scene in the i-th sub-region is divided into m1×n1×h1 voxel units according to the actual three-dimensional coordinates, which correspond to the three-dimensional coordinates of the illumination distribution modeling unit. The coordinates (x,y,z) of each voxel correspond to a set of acoustic parameters. The sound pressure distribution matrix was obtained using the inverse ray tracing algorithm. The sound pressure distribution matrix consists of the sound pressure intensity at the three-dimensional spatial position corresponding to each voxel. .

[0014] Preferably, the acoustic environment analysis module further includes a sound energy attenuation calculation unit and a second correction unit; The sound energy attenuation calculation unit is used to identify the r-th reflecting surface in the path from each voxel to each sound source. The reflecting surfaces include different building materials. After identifying the movement state of the crowd, it calculates the sound absorption attenuation coefficient of each voxel. ; Constructing the sound absorption coefficient of the crowd The identification process involves recognizing the population state settings within each voxel, and the steps are as follows: Calculate the ratio of the number of people within a voxel to the voxel's area to obtain the population density; When the crowd density is in the range of 0 people / m² to 0.5 people / m², it is identified as a dispersed state of single people with no mutual obstruction, and the crowd sound absorption coefficient is set to 0.1 to 0.15; When the crowd density is in the range of 0.5 people / m² to 2.0 people / m², it is identified as a dense standing state with mutual occlusion, and the crowd sound absorption coefficient is set to 0.2 to 0.3. When the crowd density is higher than 2.0 people / m², it is identified as a densely packed group of people sitting in their seats, with mutual obstruction. The crowd sound absorption coefficient is set to 0.4~0.7.

[0015] Preferably, the second correction unit is used to preset the sound absorption attenuation threshold. If the sound absorption attenuation coefficient of each voxel Exceeding the sound absorption attenuation threshold This indicates that the sound absorption effect on the voxel is abnormal, which may lead to a significant decrease in sound pressure intensity, insufficient local sound energy or imbalance of the sound field, and requires correction of the sound pressure intensity. For voxels with abnormal sound absorption effects, the sound pressure correction gain ratio is calculated based on the difference between the sound absorption attenuation coefficient of each voxel and the sound absorption attenuation threshold. ; and correct the gain ratio for sound pressure levels. Sound pressure intensity at each voxel corresponding to a three-dimensional spatial location Make corrections to obtain the corrected sound pressure level. During the adjustment process, the sound energy is effectively replenished to the abnormal voxel by adjusting the corresponding sound source power or sound source direction until the sound pressure intensity reaches the corrected sound pressure intensity. until.

[0016] This invention provides an immersive collaborative design system for environmental art that integrates multi-source data. It offers the following advantages: By fusing and analyzing audience behavior information, lighting distribution data, sound field propagation characteristics, and scene device status in real time, the system can automatically adjust lighting intensity and sound pressure distribution based on audience flow density, movement speed, and distribution changes, ensuring a stable visual and auditory experience during dynamic performances. In the dynamic lighting analysis module, identifying anomalous voxels and correcting lighting intensity effectively compensates for localized shadows or brightness attenuation, restoring the design illuminance level. In the sound environment analysis module, correcting sound pressure intensity based on sound absorption attenuation coefficients significantly reduces sound field unevenness caused by changes in audience density and differences in material reflectivity, maintaining sound propagation balance and clarity. The audience interaction behavior recognition module automatically triggers start / stop commands for lighting and sound equipment based on the dynamic changes of audience entering and exiting sub-areas, avoiding ineffective energy consumption and delayed response issues, achieving synchronized response and energy-saving control of lighting and sound equipment and audience behavior. By establishing a three-dimensional voxel-level distribution model of lighting and sound field, the system can maintain a continuous transition of light and shadow and sound pressure in complex terrain and multi-layered display spaces, reducing visual focus shift or sound field discontinuity caused by local deviations, thereby enhancing the overall spatial immersion and artistic expression. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the system flow of the present invention; Figure 2 This is a schematic diagram of the flow path in the application scenario of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1 This invention is applicable to scenic spots, garden night tours, and cultural performance venues of varying sizes. Please refer to [link / reference]. Figures 1-2 This invention provides an immersive collaborative design system for environmental art that integrates multi-source data, comprising: The pedestrian interaction behavior recognition module is used to divide the scenic area or open exhibition space into several stage sub-areas according to the terrain, traffic flow, and viewing path. Each sub-area corresponds to an independent light and shadow control domain and sound field control domain. It is used to collect the real-time location set of the audience or performers in the i-th sub-area at time t through visual recognition and infrared sensing devices, and to set the entrance boundary for each sub-area. and export border Construct the number of new entrants entering the entrance boundary within time t. and the speed of the flow of people This is used to identify whether a command to activate the lighting and sound field devices has been triggered, and to construct the number of people leaving the exit boundary within time t. and leaving ratio To identify whether a command to shut down lighting and sound field devices has been triggered; The dynamic lighting analysis module is used to collect lighting datasets from sub-regions and establish a three-dimensional lighting matrix. The three-dimensional lighting matrix consists of the lighting intensity at the three-dimensional spatial position (x, y, z) corresponding to each voxel. ; And construct the illumination attenuation coefficient of each voxel in the i-th sub-region at time t. And evaluate to identify voxels with anomalous light attenuation, and assess the light intensity of voxels with anomalous light attenuation. Make corrections; The acoustic environment analysis module is used to collect lighting datasets by deploying sound fields in the stage area and audience area. After establishing a three-dimensional sound field distribution model based on the Sound-Energy-Propagation Model, it constructs the sound pressure intensity at the corresponding three-dimensional spatial location for each voxel. And combined with the sound absorption coefficient of the r-th reflecting surface in the path from the sound source to each voxel. and crowd sound absorption coefficient The sound absorption attenuation coefficient of each voxel is jointly constructed. And an assessment was conducted to identify anomalous voxels affecting sound absorption, and the sound pressure intensity of these anomalous voxels was evaluated. Make corrections.

[0020] In this embodiment, by fusing and analyzing audience behavior information, lighting distribution data, sound field propagation characteristics, and scene device status in real time, the system can automatically adjust the lighting intensity and sound pressure distribution based on audience flow density, movement speed, and distribution changes, ensuring a stable visual and auditory experience during dynamic performances. In the dynamic lighting analysis module, identifying anomalous voxels and correcting lighting intensity effectively compensates for local shadows or brightness attenuation, restoring the design illuminance level. In the sound environment analysis module, correcting sound pressure intensity based on sound absorption attenuation coefficients significantly reduces sound field unevenness caused by changes in audience density and differences in material reflectivity, maintaining sound propagation balance and clarity. The audience interaction behavior recognition module automatically triggers start / stop commands for lighting and sound equipment based on the dynamic changes of audience entering and exiting sub-areas, avoiding ineffective energy consumption and delayed response issues, achieving synchronized response and energy-saving control of lighting and sound equipment and audience behavior. By establishing a three-dimensional voxel-level distribution model of lighting and sound field, the system can maintain a continuous transition of light and shadow and sound pressure in complex terrain and multi-layered display spaces, reducing visual focus shift or sound field discontinuity caused by local deviations, thereby enhancing the overall spatial immersion and artistic expression.

[0021] Example 2 This embodiment is an explanation based on Embodiment 1. Please refer to it. Figure 1 Specifically, the human interaction behavior recognition module includes a partitioning unit, a video frame acquisition unit, and a dynamic trigger sub-area control unit; The partition unit is used to divide the scenic area or open exhibition space into several stage sub-areas according to the terrain, circulation routes and viewing paths, i=1,2,...,n; n represents the total number of sub-areas; each sub-area i corresponds to an independent light and shadow control domain and sound field control domain; The video frame acquisition unit is used to deploy visual recognition devices in each sub-region to acquire video frame sequences of the audience and identify the individual positions of the audience using YOLO or OpenPose object detection algorithms, thereby obtaining the set of real-time positions of the audience in the i-th sub-region at time t.

[0022] in, Real-time number of users in the sub-region; Let be the two-dimensional coordinates of the k-th viewer in the i-th sub-region; These represent the positions of the k-th viewer in the i-th sub-region in the horizontal x-direction and the vertical y-direction, respectively.

[0023] Dynamically triggered sub-zone control unit, used to set the entry boundary in each sub-zone. and export border This is used for triggering decisions in the light and shadow control domain and the sound field control domain. The specific steps are as follows: Let the boundary of the entrance to the i-th sub-region be... Identify the number of new entrants who enter the entrance boundary within time t. for:

[0024] Among them, 1 This is an indicator function; it is 1 if the audience is inside the entrance boundary, and 0 otherwise. The sliding window method is used, by sliding the time window. The positions of adjacent video frames are temporally differencing to calculate the average displacement vector of the pedestrian flow in the i-th sub-region. :

[0025] in, Indicates the time interval between two frames; Decomposed into average displacement components of the population and the average vertical displacement component of the crowd ;

[0026]

[0027] The vector synthesis of the two displacement components yields the magnitude of the pedestrian velocity in this sub-region. Combined with the average displacement component of the population and the average vertical displacement component of the crowd The conversion yields the magnitude of the pedestrian flow velocity in the i-th sub-region. :

[0028] Average displacement vector of pedestrian flow in sub-region i , is a macroscopic population flow vector, no longer tracking each individual, but representing the dominant movement trend of people in the entire sub-region; the average displacement vector of people in the i-th sub-region. It is a two-dimensional vector that contains both direction and velocity information; =0 → This indicates that the group is moving horizontally to the right at a speed of 1m / s; =1 → This indicates that the group is moving vertically upwards at a speed of 1 m / s; the velocity modulus of the pedestrian flow is [missing value]. It represents the overall speed of the group. Regardless of which direction the group moves, as long as the scale is large, it indicates that the overall flow is fast; this makes it easier to trigger control logic. Set threshold for new hires and movement speed threshold ,when or At that time, the command to activate the lighting and sound field devices is triggered; The specific determination of the sample sub-region in Example 2 is shown in Table 1 below: Table 1. Example table of sample sub-regions in Example 2:

[0029] Principle: Real-time number of users in the sub-region and number of new employees The difference lies in the number of viewers already present within the sub-area, but the duration and speed of the crowd increase. The calculation (m / s) indicates that the moving speed threshold has not been reached. The pedestrian flow velocity model is used to distinguish between the two states of "static gathering" and "dynamic flow". When there is static gathering, the real-time number of people in the sub-area is... The large number of people, but little change in position, and the speed of the flow of people is approximately 0 m / s, indicating that the audience is resting, staying or waiting, and no dynamic light and shadow performance is needed; In a dynamically flowing state, the number of people in the sub-region in real time The number of people was large and there was obvious movement, when When a group begins to move, enter, leave, or focus on a particular performance point, dynamic light and shadow performances are required.

[0030] Or it could be a transitional state, involving minor movements. Between 0.1 m / s and 0.5 m / s, the group may move slightly and change positions, thus it is considered a non-moving signal.

[0031] In this embodiment, by introducing a multi-parameter joint judgment mechanism that includes real-time number of people in sub-areas, number of new entrants, and crowd flow speed modulus, the system can effectively distinguish between two crowd states: "static gathering" and "dynamic flow." When the audience is in a static state such as staying or resting, the lighting and sound equipment remains on standby; when an increased crowd flow trend is detected or the number of new entrants exceeds a threshold, the corresponding lighting and sound performance modules are automatically triggered, realizing an immersive scene adaptive response driven by crowd flow.

[0032] By setting entrance and exit boundaries and combining this with a sliding time window to calculate the average displacement vector and velocity modulus of the crowd flow, the system can spatially locate changes in audience distribution and achieve continuous temporal monitoring. This method ensures precise spatiotemporal alignment between the start and stop of light and sound effects and audience movement, avoiding delays and mismatches caused by traditional timed or manual triggering. When the audience moves between different sub-areas, this module can automatically achieve a gradual transition of light and sound effects based on the dynamic changes in the crowd flow velocity modulus and the number of new entrants. Thus, along the audience's path, visual and auditory effects seamlessly connect with their movement, creating a continuous and immersive viewing experience.

[0033] By using threshold-triggered control logic, relevant equipment can be activated only when significant changes in pedestrian flow or gathering trends are detected. This effectively avoids prolonged idle operation of equipment, reduces system energy consumption and maintenance costs, and extends the lifespan of lighting and sound devices.

[0034] Example 3 This embodiment is an explanation based on Embodiment 1. Please refer to it. Figure 1 Specifically, the pedestrian interaction behavior recognition module also includes a second trigger sub-area control unit and a flow path sequence generation unit; The second trigger sub-zone control unit is used to dynamically determine the audience's dwell time in each sub-zone, and triggers a command to shut down the lighting and sound field equipment when it detects that the audience is gradually leaving or the dwell time reaches a preset threshold, including the following steps: Within the i-th sub-region, record the audience dwell time from the start of the lighting and sound field equipment to the current time, calculate the occupancy ratio of the current sub-region, and obtain the occupancy ratio. ; Let the exit boundary of the i-th sub-region be... The number of people leaving the exit boundary within time t. for:

[0035] Among them, 1 This is an indicator function; it is 1 if the audience is inside the exit boundary, and 0 otherwise. Calculate the number of people leaving the exit boundary within time t. The departure ratio is obtained by comparing the number of people who enter with the total number of people who leave. ; Preset stay threshold With departure ratio threshold ,when or The second trigger sub-zone control unit outputs a command to shut down the light and sound field devices; Generate flow path sequence units, used to generate flow path sequence units at the center point of each sub-region. As path nodes, sub-regions with a "closed trigger" status are identified and marked as flow endpoint nodes, arranged according to the order of their closed trigger status to form a flow path sequence: If multiple sub-regions are closed within the same time window, they are arranged from highest to lowest according to their departure ratios → the main flow path takes priority.

[0036] The specific sample sub-regions determined in Example 3 are shown in Table 2 below: Table 2. Example table of sample sub-regions in Example 2:

[0037] In this embodiment, by recording the audience's dwell time in real time and calculating the ratio of the total time spent, this module can accurately reflect the audience's engagement continuity in each sub-area. Combined with the dynamic monitoring of the departure ratio, the system can identify the gradual dispersal or diminishing interest of the audience, and automatically shut down the lighting and sound equipment when the preset dwell time threshold or departure ratio threshold is reached, achieving adaptive end control of the performance content and avoiding resource waste. The flow path sequence generation unit uses the center point of each sub-area as the path node and constructs the main flow path according to the sub-area closing order and departure ratio, realizing intelligent tracking of the audience's movement direction. When some sub-areas are closed, the system can determine the main flow direction of the crowd, providing a predictive basis for the early activation of lighting and sound in adjacent sub-areas, ensuring the spatial continuity and rhythmic connection of the overall immersive experience. Compared with the traditional timed shutdown mode, this module judges the performance status based on real-time crowd flow data, which can significantly reduce the idle operation and ineffective energy consumption of the equipment. When the audience gradually leaves a certain area or stays for too long, the system shuts down the equipment in that area in a timely manner, realizing an energy-saving operation mechanism of "lights off when people leave, sounds cease and scenery becomes quiet".

[0038] The system can not only automatically wrap up and connect performance scenes based on real-time data, but also provide maintenance personnel with clear information on the dynamic status and flow trends of sub-areas, reducing manual intervention and improving the organizational efficiency of on-site performances and the audience's immersive participation.

[0039] Example 4 This embodiment is an explanation based on Embodiment 1. Please refer to it. Figure 1 Specifically, the dynamic light and shadow analysis module includes a light parameter acquisition unit and a light distribution modeling unit; The lighting parameter acquisition unit is used to acquire multi-source lighting parameters of the stage scene in the sub-area in real time through a scene sensor array, photometering equipment, and ambient light sensor, including the position of the main light source. Light intensity at the location Auxiliary light source in position intensity Angle between auxiliary light source and surface normal and material reflectivity The multi-source illumination parameters are discretized into a spatial raster format to form an illumination dataset; The illumination distribution modeling unit is used to establish and train an illumination energy balance model based on the radiative-transfer-model. The stage scene in the i-th sub-region is divided into m1×n1×h1 voxel units according to the actual three-dimensional coordinates. The coordinates (x,y,z) of each voxel correspond to a set of illumination datasets. m1 represents the number of voxels in the horizontal direction; n1 represents the number of voxels in the vertical direction; h1 represents the number of voxels in the vertical direction, i.e., the z-axis direction; Lighting datasets from different spatial points are acquired using multi-angle lighting sensors, cameras, and brightness probes installed at the top, sides, and front of the stage. These non-uniformly distributed lighting datasets are then mapped to corresponding voxel coordinate systems. Bilinear or Gaussian interpolation algorithms are used to spatially smooth and complete the lighting values ​​in the missing areas, forming a three-dimensional lighting matrix. This three-dimensional lighting matrix consists of the lighting intensity at each voxel's corresponding three-dimensional spatial position (x, y, z). The composition is the result of the fusion of main light source, auxiliary light, and reflected brightness factors; Light intensity at each voxel's corresponding three-dimensional spatial location The calculation formula is:

[0040] in, Indicates the position of the main light source The light intensity at that location, Indicates the location Material reflectivity ; The cosine of the incident angle of the main light source is the cosine of the angle between the direction of the main light source and the surface normal. This is used to ensure that the light contributes only when the surface is oriented towards the light source, thus avoiding negative values; Indicates the s-th auxiliary light source in The light intensity at that location, Indicates in The total number of auxiliary light sources converged at the location. Let represent the angle between the s-th auxiliary light source and the surface normal; where, in the physical modeling of illumination, the local surface illumination intensity distribution follows the law of conservation of light energy and the law of reflection, where, The intensity of light received at a point = incident light intensity × material reflectivity × incident angle influence factor. This represents "main light source illumination energy × material reflectivity × light angle." When light shines perpendicularly (angle 0°) onto the surface, the light intensity received per unit area is maximum. When the light is tilted (angle increases), the effective illumination area decreases, and therefore the light intensity received per unit area also decreases accordingly. This determines the reflected brightness. This means that the larger the angle of incidence of light (the more oblique), the smaller the effective illuminance, therefore multiplied by... According to Lambert's cosine law, the reaction is related to the auxiliary light source.

[0041] By collaborating with the lighting parameter acquisition unit and the lighting distribution modeling unit, real-time modeling and spatial compensation of the stage area's lighting distribution are achieved. The principle involves using multi-angle sensors and metering equipment to collect multi-source lighting parameters, including the main light source, auxiliary light source, and reflected brightness. A light energy balance relationship is established based on a light radiation transfer model, and the unevenly sampled lighting data is mapped into a three-dimensional voxel space. An interpolation algorithm is used to smoothly fill in missing areas, constructing a three-dimensional lighting matrix that reflects the spatial gradient changes of stage lighting. Following the conservation of light energy and Lambert's cosine law, and considering the main light incident angle, material reflectivity, and the superposition effect of auxiliary light, the actual brightness distribution at different spatial points is accurately expressed. Simultaneously, it provides a lighting spatial benchmark for the fusion of sound field and pedestrian flow data, thereby improving the spatial consistency of light and shadow rendering and the audience's visual immersion in immersive displays.

[0042] The dynamic light and shadow analysis module also includes a dynamic calculation unit for light attenuation; The dynamic illumination attenuation calculation unit is used to dynamically calculate the illumination attenuation coefficient of each voxel in the i-th sub-region at time t, based on the change in illumination intensity at the corresponding three-dimensional spatial position of each voxel and the occlusion caused by crowds. ; The specific steps include: In real-world scenarios, light gradually weakens as it travels through the air due to factors such as scattering, absorption, and shading. In particular, the movement and gathering of people can create variable shading areas, leading to a decrease in local light intensity.

[0043] To quantify the gradual weakening of light due to scattering, absorption, and shading during its propagation through the air, and the variable shading areas formed by the movement and gathering of people, resulting in a local decrease in light intensity, the area of ​​the shading portion within the voxel region of the sub-region is divided. Total area of ​​sub-regions The ratio is calculated to obtain the occlusion percentage. ; Occlusion percentage This is used to reflect the proportion of area obscured by crowds or equipment. When the obscured area increases, it indicates that more spectators or equipment are blocking the main or auxiliary light source, resulting in a decrease in local luminous flux and a corresponding increase in the luminous attenuation coefficient. If the crowd's movement direction is the same as the main light direction, the occlusion effect is enhanced; if it is the opposite, the occlusion effect is relatively weakened. By combining the occlusion ratio with the crowd's dominant direction, a dynamically updated relative light attenuation coefficient can be obtained, which is used to correct the current lighting distribution matrix and achieve adaptive rendering of stage lighting effects.

[0044] Based on the average displacement component of the population and the average vertical displacement component of the crowd Further calculations Macro-dominant direction of the population in i sub-regions :

[0045] Identify the macro-dominant direction angle of the population This represents the macroscopic flow direction of people in the sub-region on a two-dimensional plane at time t, including: arctan is the arctangent function; if and This indicates that the crowd is moving horizontally; if For positive and This indicates that the crowd is moving vertically. Extract the cosine value of the incident angle of the light source. The dominant direction angle of the macro-flow of the population in the i-th sub-region : Calculate and obtain the occlusion angle :

[0046] like This indicates that the direction of crowd movement is the same as the direction of illumination, and there is a risk of light source attenuation effect within the sub-area due to crowd occlusion; if This indicates that the direction of crowd movement is opposite to the direction of light illumination, and there is no risk of light source attenuation effect in the sub-area due to crowd occlusion. Based on the percentage of occlusion and the angle of obstruction The coupling is performed to calculate the illumination attenuation coefficient of each voxel in the i-th sub-region at time t. :

[0047] Where q is the occlusion sensitivity coefficient. When using a high-brightness directional light source (such as a spotlight), the occlusion effect is significant, and q is set to 0.6~0.9. When using a soft or diffused light source (such as a background fill light), the occlusion effect is weaker, and q is set to 0.2~0.5. In scenes with mixed light source types or complex light fields, an intermediate value of q of 0.4~0.7 can be used, and dynamic self-calibration can be performed through measured illumination feedback data, updating the illumination attenuation coefficient of each voxel. .

[0048] The dynamic light and shadow analysis module also includes a first correction unit, which is used to set the light attenuation threshold. If the illumination attenuation coefficient of each voxel in the i-th sub-region at time t This indicates abnormal light attenuation, which may cause localized shadows on actors or stage objects, potentially affecting the audience's visual focus. In this case, the light enhancement ratio is calculated for the voxel with abnormal light attenuation. :

[0049] in, This indicates the maximum amplitude of a single enhanced illumination, set to 0.5. This represents the first illumination gain factor, set to 1~1.5; and sets the illumination enhancement ratio. Illumination intensity at each voxel's corresponding three-dimensional spatial location Make corrections to obtain the corrected light intensity. :

[0050] During the adjustment process, the main light source, auxiliary light source, and light direction are adjusted so that the light intensity of each voxel at its corresponding 3D spatial position is updated after the light is projected onto the shadow area, until the corrected light intensity is achieved. Until the value is reached.

[0051] If a voxel is originally 200 lx, has a light attenuation coefficient of 0.4, a first light gain factor of 1.2, and a light attenuation threshold of 0.25: K v =min((0.4-0.25)×1.2,0.5)=min(0.18,0.5)=0.18; The corrected illuminance is 200 × 1.18 ≈ 236 lx; local shadows are compensated, and the visual focus is restored to a level close to the design illuminance.

[0052] In this embodiment, the dynamic calculation unit for light attenuation is combined with the first correction unit to achieve intelligent identification and adaptive correction of lighting anomalies in the stage area. The principle is to establish a dynamic attenuation model coupled with the occlusion ratio and light direction by real-time acquisition of light intensity changes and crowd movement data. When audiences move or gather within the stage or display area, the crowd forms variable occlusion zones, obstructing the illumination of the main or auxiliary light source. This invention calculates the occlusion ratio and the macroscopic dominant direction angle of the crowd, combined with the cosine relationship of the incident angle of the light source, to obtain the occlusion angle, thereby dynamically evaluating the light attenuation coefficient. Furthermore, it uses the occlusion sensitivity coefficient k to perform weighted correction for different types of light sources, achieving quantitative modeling and real-time response to light attenuation. When the light attenuation coefficient of a voxel exceeds a threshold, the system automatically triggers the first correction unit, calculating the gain using the light enhancement ratio formula and adjusting the brightness and direction of the main and auxiliary light sources to compensate for local shadow areas. The beneficial effect is that this embodiment can identify lighting attenuation anomalies caused by crowd movement, device occlusion, or changes in the incident angle of the light source in real time, achieving adaptive adjustment of the spatial distribution of light. By dynamically controlling the illumination enhancement ratio, problems such as excessively low local illumination or focus shift are avoided, which promotes and enhances the audience's visual focus and the sense of stage layering. Based on the self-calibration of light source type, the accuracy and robustness of illumination correction are improved, enabling the system to maintain a stable immersive visual experience in complex lighting environments.

[0053] Example 5 This embodiment is an explanation based on Embodiment 1. Please refer to it. Figure 1 Specifically, the acoustic environment analysis module includes a sound field parameter acquisition unit and a sound field modeling unit; The sound field parameter acquisition unit is used to acquire the total number of sound sources on the stage within the sub-area. The sound power of the a-th sound source The propagation path length from the a-th sound source to the corresponding three-dimensional spatial position (x, y, z) of each voxel. Establish a sound field dataset and convert the collected sound field dataset into a spatial raster data format so that it can be fused and analyzed with the illumination distribution matrix in the same coordinate system; The sound field modeling unit is used to establish a three-dimensional sound field distribution model based on the Sound-Energy-Propagation Model. It divides the stage scene in the i-th sub-region into m1×n1×h1 voxel units according to the actual three-dimensional coordinates. The coordinates (x,y,z) of each voxel correspond to a set of lighting datasets. Corresponding to the three-dimensional coordinates of the lighting distribution modeling unit, the coordinates (x,y,z) of each voxel correspond to a set of acoustic parameters. The sound pressure distribution matrix is ​​obtained using the inverse ray-tracing algorithm. The sound pressure distribution matrix consists of the sound pressure intensity at the three-dimensional spatial position (x, y, z) corresponding to each voxel. Composition, expressed as follows:

[0054] in, This indicates the total number of sound sources on the stage within the sub-area. This represents the sound power of the a-th sound source. Let represent the propagation path length from the a-th sound source to the corresponding three-dimensional spatial position (x, y, z) of each voxel; during propagation, sound energy diffuses onto a sphere centered on the sound source, representing free propagation, with a sphere area of ​​. .

[0055] The sound field modeling unit further obtains the sound pressure distribution matrix through inverse ray-tracing. The sound pressure distribution matrix consists of the sound pressure intensity values ​​at the corresponding three-dimensional spatial positions of each voxel, used to characterize the uniformity and local focusing characteristics of sound energy distribution in the stage space. Sound field modeling and lighting modeling use the same voxel coordinate system, allowing the sound pressure field and lighting field to be fused and analyzed in the same spatiotemporal domain, laying the foundation for multimodal stage environment modeling. Through inverse ray-tracing, complex reflection, diffusion, and interference effects within the stage can be accurately simulated. Compared to the traditional forward ray model, computational efficiency is improved by approximately 30%, and the error is reduced to within ±1.5 dB.

[0056] Example 6 This embodiment is an explanation based on Embodiment 5. Please refer to it. Figure 1 Specifically, the acoustic environment analysis module also includes a sound energy attenuation calculation unit and a second correction unit; The sound energy attenuation calculation unit is used to identify the r-th reflecting surface in the path from each voxel to each sound source. The reflecting surfaces include different building materials. After identifying the movement state of the crowd, it calculates the sound absorption attenuation coefficient of each voxel. :

[0057] in, This represents the sound absorption coefficient of the r-th reflecting surface along the propagation path. Indicates the number of reflective surfaces of building materials along the path; This represents the calculated proportion of sound energy remaining after reflection from all surfaces. This represents the calculated proportion of sound energy remaining after passing through the crowd within the voxel; multiplying these two values ​​gives the total proportion of remaining energy from the sound source, through path reflection and absorption by the crowd at the destination, finally reaching the listener (or voxel center); subtracting this total proportion of remaining energy from 1 gives the total energy loss proportion, defined as the sound absorption attenuation coefficient for each voxel. Sound absorption attenuation coefficient per voxel The value ranges from [0,1]. The larger the value, the more severe the total attenuation of sound energy from the sound source to the voxel. The r-th reflecting surface along the propagation path, if made of smooth concrete wall, has a sound absorption coefficient of 0.02~0.05. If the material is gypsum board, the sound absorption coefficient is 0.1~0.2; If the material is fabric, the sound absorption coefficient is 0.3~0.5; If the material is carpet, the sound absorption coefficient is 0.2~0.4; Constructing the sound absorption coefficient of the crowd The identification process involves recognizing the population state settings within each voxel, and the steps are as follows: Calculate the ratio of the number of people within a voxel to the voxel's area to obtain the population density; When the crowd density is in the range of 0 people / m² to 0.5 people / m², it is identified as a dispersed state of single people with no mutual obstruction, and the crowd sound absorption coefficient is set to 0.1 to 0.15; When the crowd density is in the range of 0.5 people / m² to 2.0 people / m², it is identified as a dense standing state, where the crowd is close together, the bodies are close together, and the clothing absorbs sound significantly. The crowd sound absorption coefficient is set to 0.2 to 0.3. When the crowd density is higher than 2.0 people / m², the audience is identified as extremely dense or seated, the seating arrangement is fixed, the crowd is highly concentrated, and the crowd sound absorption coefficient is set to 0.4~0.7.

[0058] Standards such as ISO 3382 (Acoustic Measurement Methods in Buildings) and ISO 18233 (Application of Acoustic Measurement Methods in Buildings) address the impact of human populations on the indoor acoustic environment and involve the determination of sound absorption coefficients. International standards (such as ISO 354, "Methods for measuring sound absorption of materials") provide measurement results for the sound absorption coefficients of various building materials. For example: smooth concrete wall: 0.02~0.05 (strong reflection, weak sound absorption); gypsum board: 0.1~0.2 (slight sound absorption); fabric curtain: 0.3~0.5 (medium sound absorption); carpet: 0.2~0.4 (good sound absorption for mid-to-low frequency range). For complex environments (such as crowds of audience members), it is difficult to accurately measure the sound absorption of each person, so empirical ranges are used: single dispersed people (0~0.5 people / m²): human body sound absorption is limited, sound absorption coefficient ≈0.1~0.15; densely standing people (0.5~2 people / m²): increased obstruction by people and clothing, sound absorption is improved, crowd sound absorption coefficient ≈0.2~0.3; extremely dense crowd / seat audience (>2 people / m²): more severe obstruction, clothing and seats contribute more to sound absorption, crowd sound absorption coefficient ≈0.4~0.7. In engineering design, software such as EASE and CATT~Acoustic also have built-in sound absorption coefficients for different materials and crowds, which can be used for acoustic simulation.

[0059] The second correction unit is used to preset the sound absorption attenuation threshold. If the sound absorption attenuation coefficient of each voxel Exceeding the sound absorption attenuation threshold This indicates that the sound absorption effect on the voxel is abnormal, which may lead to a significant decrease in sound pressure intensity, insufficient local sound energy or imbalance of the sound field, and requires correction of the sound pressure intensity. For voxels with abnormal sound absorption effects, the sound pressure correction gain ratio is calculated based on the difference between the sound absorption attenuation coefficient of each voxel and the sound absorption attenuation threshold. :

[0060] in, This represents the second sound pressure level gain factor, set to 1~2; To control the maximum amplitude of sound pressure during a single enhancement and avoid overcompensation of sound pressure, it is set to 0.5~0.6; And the sound pressure correction gain ratio Sound pressure intensity at each voxel corresponding to a three-dimensional spatial location Make corrections to obtain the corrected sound pressure level. :

[0061] During the adjustment process, the sound energy is effectively replenished to the abnormal voxel by adjusting the corresponding sound source power or sound source direction until the sound pressure intensity reaches the corrected sound pressure intensity. until.

[0062] A voxel has an initial sound pressure level of 85 dB, a sound absorption attenuation coefficient of 0.6, a threshold of 0.4, a second sound pressure gain factor Gs = 1.5, and a maximum enhancement G. max =0.5, then the corrected sound pressure level is 85×(1+0.3)≈110.5dB.

[0063] In this embodiment, sound energy propagating between air and structural surfaces undergoes various influencing factors, including distance attenuation, sound absorption attenuation, and reflection loss. According to the principle of energy conservation, sound pressure intensity is inversely proportional to the square of the propagation distance, and energy attenuation occurs at each reflection according to the sound absorption coefficient of the reflecting surface. Sound pressure intensity decreases significantly as the sound absorption attenuation coefficient increases. When the sound absorption attenuation exceeds the system-set sound absorption attenuation threshold, it indicates excessive sound energy loss in that voxel, leading to an unbalanced sound field.

[0064] By identifying the sound absorption characteristics of each reflecting surface along the propagation path and the changes in crowd density, a multi-factor coupled sound absorption attenuation coefficient is established. This system enables precise calculation of sound energy attenuation in complex scenarios. When local voxel absorption anomalies cause a drop in sound pressure, the system can automatically calculate and correct the gain, compensating by adjusting the sound source power and direction to ensure a uniform sound pressure distribution in the audience area, significantly improving the auditory experience. The system can dynamically update attenuation parameters based on real-time collected crowd density and material sound absorption status, forming a "sensing-feedback-adjustment" closed loop to improve the system's robustness and adaptability in different scenarios. Through a maximum enhancement amplitude control mechanism, it can prevent excessive sound pressure increase from causing howling, interference, or sound field instability, ensuring a smooth and controllable acoustic adjustment process.

[0065] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value, it is acceptable.

[0066] The above formulas are all derived from software simulation using a large amount of data and are selected to be close to the actual values. The coefficients in the formulas are set by those skilled in the art according to the actual situation. The above description is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any equivalent substitutions or changes made by those skilled in the art within the technical scope disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the protection scope of the present invention.

Claims

1. A multi-source data fusion-based immersive collaborative design system for environmental art, characterized in that: include: The pedestrian interaction behavior recognition module is used to divide the scenic area or open exhibition space into several stage sub-areas according to the terrain, traffic flow, and viewing path. Each sub-area corresponds to an independent light and shadow control domain and sound field control domain. It also collects the real-time location set of the audience in the i-th sub-area at time t and sets the entrance boundary for each sub-area. and export border Construct the number of new entrants entering the entrance boundary within time t. and the speed of the flow of people This is used to identify whether a command to activate the lighting and sound field devices has been triggered, and to construct the number of people leaving the exit boundary within time t. and leaving ratio To identify whether a command to shut down lighting and sound field devices has been triggered; The dynamic lighting analysis module is used to collect lighting datasets from sub-regions and establish a three-dimensional lighting matrix. The three-dimensional lighting matrix consists of the lighting intensity at the three-dimensional spatial position (x, y, z) corresponding to each voxel. Composition; and construction of the illumination attenuation coefficient of each voxel in the i-th sub-region at time t. To assess and identify voxels with anomalous light attenuation, and to evaluate the light intensity of voxels with anomalous light attenuation. Make corrections; The acoustic environment analysis module is used to collect sound field datasets in sub-regions, establish a three-dimensional sound field distribution model based on the sound energy propagation model, and construct the sound pressure intensity at the three-dimensional spatial location corresponding to each voxel. And combined with the sound absorption coefficient of the r-th reflecting surface in the path from the sound source to each voxel. and crowd sound absorption coefficient The sound absorption attenuation coefficient of each voxel is jointly constructed. To assess and identify anomalous voxels affecting sound absorption, and to evaluate the sound pressure intensity of these anomalous voxels. Make corrections.

2. The multi-source data fusion-based immersive collaborative design system for environmental art according to claim 1, characterized in that, The human interaction behavior recognition module includes a partitioning unit, a video frame acquisition unit, and a dynamic trigger sub-region control unit; The partitioning unit is used to divide the scenic area or open exhibition space into several stage sub-areas according to the terrain, circulation routes and viewing paths, i=1,2,...,Q; Q represents the total number of sub-areas; each sub-area i corresponds to an independent light and shadow control domain and sound field control domain; The video frame acquisition unit is used to deploy visual recognition devices in each sub-region to acquire video frame sequences of the audience and identify the individual positions of the audience using YOLO or OpenPose object detection algorithms, thereby obtaining the set of real-time positions of the audience in the i-th sub-region at time t. in, Real-time number of users in the sub-region; Let be the two-dimensional coordinates of the k-th viewer in the i-th sub-region; These represent the positions of the k-th viewer in the i-th sub-region in the horizontal x-direction and the vertical y-direction, respectively.

3. The multi-source data fusion-based immersive collaborative design system for environmental art according to claim 2, characterized in that, The dynamic triggering sub-zone control unit is used to set the entry boundary in each sub-zone. and export border This is used for triggering decisions in the light and shadow control domain and the sound field control domain. The specific steps are as follows: Let the boundary of the entrance to the i-th sub-region be... Identify the number of new entrants who enter the entrance boundary within time t. ; The sliding window method is used, by sliding the time window. Temporal difference is performed on the positions of adjacent video frames to calculate the average displacement vector of the pedestrian flow in the i-th sub-region. ; Decomposed into average displacement components of the population and the average vertical displacement component of the crowd ; The vector synthesis of the two displacement components yields the magnitude of the pedestrian velocity in this sub-region. ; Set threshold for new hires and movement speed threshold ,when or At that time, the command to start the lighting and sound field devices is triggered.

4. The multi-source data fusion-based immersive collaborative design system for environmental art according to claim 1, characterized in that, The pedestrian interaction behavior recognition module also includes a second trigger sub-area control unit and a flow path sequence generation unit; The second trigger sub-zone control unit is used to record the audience dwell time from the start time of the light and sound field equipment to the current time within the i-th sub-zone, calculate the occupancy time ratio of the current sub-zone, and obtain the occupancy time ratio. ; Let the exit boundary of the i-th sub-region be... Identify the number of people who leave the exit boundary within time t. ; Calculate the number of people leaving the exit boundary within time t. The departure ratio is obtained by comparing the number of people who enter with the total number of people who leave. ; Preset stay threshold With departure ratio threshold ,when or The second trigger sub-zone control unit outputs a command to shut down the light and sound field devices; The unit for generating flow path sequences is used to generate flow path sequences at the center point of each sub-region. As path nodes, sub-regions with a "closed trigger" status are identified and marked as flow endpoint nodes, arranged according to the order of their closed trigger status to form a flow path sequence: If multiple sub-regions are closed within the same time window, they are arranged from highest to lowest according to their departure ratios → the main flow path takes priority.

5. The immersive collaborative design system for environmental art based on multi-source data fusion according to claim 1, characterized in that, The dynamic light and shadow analysis module includes a light parameter acquisition unit and a light distribution modeling unit; The illumination parameter acquisition unit is used to acquire multi-source illumination parameters of the stage scene in the sub-area in real time through a scene sensor array, photometering equipment, and ambient light sensor, including: the position of the main light source. Light intensity at the location Auxiliary light source in position intensity Angle between auxiliary light source and surface normal and material reflectivity The multi-source illumination parameters are discretized into a spatial raster format to form an illumination dataset; The illumination distribution modeling unit is used to establish and train an illumination energy balance model based on the illumination radiation transfer model. The stage scene in the i-th sub-region is divided into m1×n1×h1 voxel units according to the actual three-dimensional coordinates. The coordinates (x,y,z) of each voxel correspond to a set of illumination datasets. m1 represents the number of voxels in the horizontal direction; n1 represents the number of voxels in the vertical direction; h1 represents the number of voxels in the vertical direction, i.e., the z-axis direction; Lighting datasets from different spatial points are acquired using multi-angle lighting sensors, cameras, and brightness probes installed at the top, sides, and front of the stage. These non-uniformly distributed lighting datasets are then mapped to corresponding voxel coordinate systems. Bilinear or Gaussian interpolation algorithms are used to spatially smooth and complete the lighting values ​​in the missing areas, forming a three-dimensional lighting matrix. This three-dimensional lighting matrix consists of the lighting intensity at each voxel's corresponding three-dimensional spatial position (x, y, z). The composition is the result of the fusion of main light source, auxiliary light, and reflected brightness factors.

6. The immersive collaborative design system for environmental art based on multi-source data fusion according to claim 5, characterized in that, The dynamic light and shadow analysis module also includes a dynamic light attenuation calculation unit; The dynamic calculation unit for illumination attenuation is used to dynamically calculate the illumination attenuation coefficient of each voxel in the i-th sub-region at time t, based on the change in illumination intensity at the corresponding three-dimensional spatial position of each voxel and the occlusion caused by crowds. ; The specific steps include: To quantify the gradual weakening of light due to scattering, absorption, and shading during its propagation through the air, and the variable shading areas formed by the movement and gathering of people, resulting in a local decrease in light intensity, the area of ​​the shading portion within the voxel region of the sub-region is divided. Total area of ​​sub-regions The ratio is calculated to obtain the occlusion percentage. ; Based on the average displacement component of the population and the average vertical displacement component of the crowd Further calculations Macro-dominant direction of the population in i sub-regions ; Identify the macro-dominant direction angle of the population This represents the macroscopic flow direction of people in the sub-region on a two-dimensional plane at time t, including: if and This indicates that the crowd is moving horizontally; if and This indicates that the crowd is moving vertically. Extract the cosine value of the incident angle of the light source. The dominant direction angle of the macro-flow of the population in the i-th sub-region : Calculate and obtain the occlusion angle ; like This indicates that the direction of crowd movement is the same as the direction of illumination, and there is a risk of light source attenuation effect within the sub-area due to crowd occlusion; if This indicates that the direction of crowd movement is opposite to the direction of light illumination, and there is no risk of light source attenuation effect in the sub-area due to crowd occlusion. Based on the percentage of occlusion and the angle of obstruction The coupling is performed to calculate the illumination attenuation coefficient of each voxel in the i-th sub-region at time t. .

7. The multi-source data fusion-based immersive collaborative design system for environmental art according to claim 1, characterized in that, The light and shadow dynamic analysis module further includes a first correction unit, which is used to set the light attenuation threshold. If the illumination attenuation coefficient of each voxel in the i-th sub-region at time t This indicates abnormal light attenuation, which may cause localized shadows on actors or stage objects, potentially affecting the audience's visual focus. In this case, the light enhancement ratio is calculated for the voxel with abnormal light attenuation. and increase the proportion of light. Illumination intensity applied to each voxel at its corresponding three-dimensional spatial location Make corrections to obtain the corrected light intensity. ; During the adjustment process, the main light source, auxiliary light source, and light direction are adjusted so that the light intensity of each voxel at its corresponding 3D spatial position is updated after the light is projected onto the shadow area, until the corrected light intensity is achieved. Until the value is reached.

8. The multi-source data fusion-based immersive collaborative design system for environmental art according to claim 1, characterized in that, The acoustic environment analysis module includes a sound field parameter acquisition unit and a sound field modeling unit; The sound field parameter acquisition unit is used to acquire the total number of sound sources on the stage within the sub-area. The sound power of the a-th sound source The propagation path length from the a-th sound source to the corresponding three-dimensional spatial position (x, y, z) of each voxel. Establish a sound field dataset and convert the collected sound field dataset into a spatial raster data format so that it can be fused and analyzed with the illumination distribution matrix in the same coordinate system; The sound field modeling unit is used to establish a three-dimensional sound field distribution model based on the sound energy propagation model. It corresponds to the three-dimensional coordinates of the illumination distribution modeling unit. The coordinates (x, y, z) of each voxel correspond to a set of acoustic parameters. The sound pressure distribution matrix is ​​obtained by using an inverse acoustic ray tracing algorithm. The sound pressure distribution matrix consists of the sound pressure intensity at the three-dimensional spatial position corresponding to each voxel. .

9. The multi-source data fusion-based immersive collaborative design system for environmental art according to claim 1, characterized in that, The acoustic environment analysis module also includes a sound energy attenuation calculation unit and a second correction unit; The sound energy attenuation calculation unit is used to identify the r-th reflecting surface in the path from each voxel to each sound source. The reflecting surfaces include different building materials. After identifying the movement state of the crowd, it calculates the sound absorption attenuation coefficient of each voxel. ; Constructing the sound absorption coefficient of the crowd The identification process involves recognizing the population state settings within each voxel, and the steps are as follows: Calculate the ratio of the number of people within a voxel to the voxel's area to obtain the population density; When the crowd density is in the range of 0 people / m² to 0.5 people / m², it is identified as a dispersed state of single people with no mutual obstruction, and the crowd sound absorption coefficient is set to 0.1 to 0.15; When the crowd density is in the range of 0.5 people / m² to 2.0 people / m², it is identified as a dense standing state with mutual occlusion, and the crowd sound absorption coefficient is set to 0.2 to 0.

3. When the crowd density is higher than 2.0 people / m², it is identified as a densely packed group of people sitting in their seats, with mutual obstruction. The crowd sound absorption coefficient is set to 0.4~0.

7.

10. The multi-source data fusion-based immersive collaborative design system for environmental art according to claim 9, characterized in that, The second correction unit is used to preset the sound absorption attenuation threshold. If the sound absorption attenuation coefficient of each voxel Exceeding the sound absorption attenuation threshold This indicates that the sound absorption effect on the voxel is abnormal, which may lead to a significant decrease in sound pressure intensity, insufficient local sound energy or imbalance of the sound field, and requires correction of the sound pressure intensity. For voxels with abnormal sound absorption effects, the sound pressure correction gain ratio is calculated based on the difference between the sound absorption attenuation coefficient of each voxel and the sound absorption attenuation threshold. ; And the sound pressure correction gain ratio Sound pressure intensity at each voxel corresponding to a three-dimensional spatial location Make corrections to obtain the corrected sound pressure level. ; During the adjustment process, the sound energy is effectively replenished to the abnormal voxel by adjusting the corresponding sound source power or sound source direction until the sound pressure intensity reaches the corrected sound pressure intensity. until.

Images