Embodied intelligent perception method based on wearable mobile positioning image acquisition device

Abstract

The present invention discloses an embodied intelligent perception method based on a wearable mobile positioning image acquisition device, comprising device wear and mode setting, behavioral perception by a mobile positioning module, environmental perception by an image acquisition module, embodied perception modeling by a back-end analysis module, and integrated display on a visualization platform. Based on a wearable intelligent device, the present invention uses a mobile positioning module to accurately track the behavioral information of individuals in urban spaces over long time spans, uses an image acquisition module to record simultaneous spatial environmental information, and establishes an intelligent agent model through a back-end analysis module. The present invention can be used for embodied perception of the environment and behavior in the field of urban intelligent perception, achieving accurate identification of the correlation characteristics between the environment and behavior, and better serving urban decision-making.

Description

Technical Field

[0001] The present invention belongs to the field of urban intelligent perception, and specifically relates to an embodied intelligent perception method based on a wearable mobile positioning image acquisition device. Background Art

[0002] There is a complex spatiotemporal interaction between human behavior and the urban environment. The core goal of urban research and planning is to shape people's living environment and guide their lifestyle. Understanding the causal logic between behavior and spatial environment from an individual perspective can more accurately respond to people's diverse demands for urban space and optimize urban spatial layout. Traditional urban perception methods rely solely on spatial static image acquisition and manual spatial annotation. They are unable to process the massive spatiotemporal information generated by multidimensional urban spatial elements and multiple behaviors, and are even more difficult to effectively explain the complex characteristics and interactive relationships between behavior and environmental elements. On the other hand, the emergence of perception methods based on artificial intelligence technology provides a more efficient and scientific means of urban research.

[0003] Currently, behavioral perception methods often employ instruments such as eye trackers or cameras to monitor crowd behavior within fixed spaces. These methods tend to focus on a single dimension and struggle to accurately track individuals' spatiotemporal behavior over extended periods of time. They lack an embodied perception method that provides high-frequency feedback, precise matching, and the ability to tightly integrate and continuously track complex human activities. Furthermore, perception focuses more on identifying small movements indoors, and a perception technology system tailored to diverse behaviors across urban spaces of varying scales has yet to be established. Environmental perception methods often employ methods such as oblique photogrammetry and laser point cloud scanning to collect spatial information and perform three-dimensional reconstruction. These methods rarely incorporate behavioral information to analyze individual spatial perceptions, failing to provide effective guidance for urban planning and design. Summary of the Invention

[0004] Purpose of the invention: In response to the shortcomings of the existing technology, the present invention provides an embodied intelligent perception method based on a wearable mobile positioning image acquisition device. The present invention is based on a wearable intelligent device and can accurately track the dynamic and static behaviors of individuals in urban space, and identify synchronous spatial environmental characteristics, thereby realizing embodied intelligent perception and analytical modeling of behaviors and environments in cities.

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

[0006] The embodied intelligent perception method based on a wearable mobile positioning image acquisition device includes the following steps:

[0007] (1) Device wearing and mode setting

[0008] The subjects were fitted with the wearable components of the sensory device, including sensory glasses and a smart wristband. The mode adjustment system was used to turn the device on, enabling automatic switching of sensory modes and the acquisition and transmission of multiple types of sensory data.

[0009] (2) Mobile positioning module behavior perception

[0010] The mobile positioning module includes a mobile positioning component and a behavior perception system. It automatically switches between dynamic and static perception modes based on the individual's speed variation. In dynamic perception mode, the mobile positioning component records the individual's GPS positioning data at unit time intervals. In static perception mode, the mobile positioning component records the individual's instantaneous limb acceleration data. The behavior perception system analyzes and integrates dynamic and static behavior indicators to form an individual behavior information database, which is then transmitted in real time to the data center via the cellular network.

[0011] (3) Image acquisition module environmental perception

[0012] The image acquisition module includes an image acquisition component and an environmental perception system. In dynamic perception mode, the image acquisition component captures global spatial images. In static perception mode, the image acquisition component simultaneously captures global spatial images and images of the visible range in the direction of individual line of sight. The environmental perception system analyzes and integrates spatial and visual environmental indicators to form a spatial environmental information database, which is then transmitted in real time to the data center via a cellular network.

[0013] (4) Back-end analysis module embodied perception modeling

[0014] The back-end analysis module includes a server component, a digital sandbox system, a behavior modeling system, and an integrated analysis system. The digital sandbox system integrates spatial environment data and creates a regional 3D sandbox using 3D reconstruction methods. The behavior modeling system integrates individual behavior data, virtually reconstructing each subject's spatiotemporal behavior and mapping the reconstructed virtual individuals to their correct positions on the 3D sandbox using their GPS positioning. The integrated analysis system comprehensively analyzes individual behavior information and spatial environment parameters, and builds an agent model based on the correlation characteristics of these parameters.

[0015] (5) Visualization platform integrated display

[0016] The data results output by the back-end analysis module are displayed through a holographic digital projector, gesture recognizer and visualization system, including the regional behavior environment reconstruction sandbox and the intelligent agent deduction sandbox.

[0017] Furthermore, the mobile positioning component in step 2 is composed of a GPS locator, a timer, and an inertial measurement unit, wherein the GPS locator obtains and records the individual's real-time geographic information position, the timer records time information, and the inertial measurement unit records the acceleration of the individual's limbs.

[0018] Furthermore, the automatic switching of the dynamic sensing mode or the static sensing mode according to the individual speed change characteristics in step 2 refers to the automatic switching of the dynamic sensing mode or the static sensing mode according to the individual speed change characteristics. Calculate the individual speed standard deviation and speed change rate, and then calculate the speed fluctuation index S = α·σ v +β·△v, where α and β are weight coefficients. When the speed fluctuation index is greater than the threshold, it switches to the dynamic perception mode; when the speed fluctuation index is less than the threshold, it switches to the static perception mode.

[0019] Furthermore, the dynamic behavior index and static behavior index information in step 2 include the individual's speed and GPS coordinates, and the static behavior index information includes the posture time ratio and GPS coordinates. The posture includes standing, sitting and lying postures. The limb acceleration data recorded by the behavior perception system combined with the inertial measurement unit is obtained through Calculate the pitch and roll angles of the limbs, where (a x ,a y ,a z ) represents the acceleration components in the X, Y, and Z axis directions, and the posture type of the individual is determined according to the threshold range of the pitch angle and roll angle, and then the Calculate the proportion of time spent in each posture.

[0020] Furthermore, the image acquisition component in step three includes a panoramic camera and a local camera. The panoramic camera acquires 360° panoramic spatial images, and the local camera records spatial images of individual visual ranges at the perspective of human eyes.

[0021] Furthermore, the integration of spatial environment indicators and visual environment indicator information in step 3 refers to processing the spatial images collected by the image acquisition component through the semantic segmentation algorithm, identifying green vegetation, water bodies, buildings, roads, sky, open spaces, and obstacle facilities in the panoramic spatial image, and calculating the number of pixels N occupied by each semantic i and through Calculate the proportion of each type of pixel as a spatial environment indicator. Identify green vegetation, water bodies, buildings, roads, open spaces, and obstacles in the visible range image. Similarly, calculate the proportion of each type of semantic pixel as a visual environment indicator.

[0022] Furthermore, the fourth step comprehensively analyzes the individual behavior information and spatial environment information parameters, and establishes an intelligent perception and decision-making model based on the parameter correlation characteristics of the region, which means dividing the region into 10m*10m grid units and calculating the average behavior index B=[b1,b2,...,b n ] and the average value of spatial index E=[e1,e2,...,e m ], using a neural network to calculate the correlation model B = h(E) + ε between behavioral indicators and spatial indicators. The correlation model is embedded in the intelligent agent system, which receives spatial environment indicators as input, outputs individual behavioral indicators, and guides behavioral decision-making.

[0023] Beneficial effects:

[0024] 1. The present invention achieves precise perception and coupling of individual movements and spatial environment through wearable devices. The wearable components of the device include sensing glasses and a smart bracelet. The device is small in size, easy to wear, and suitable for all kinds of people, laying the foundation for large-scale application in urban space research and planning. It also has minimal interference with the travel behavior and movements of the subjects, ensuring effective control of experimental errors and achieving accurate collection of movement and environmental information.

[0025] 2. This invention intelligently switches perception modes based on the calculated parameters of a person's motion patterns in urban spaces, adapting to both motion and stationary states. The motion perception mode focuses on collecting information about spatiotemporal trajectory and motion speed, while the stationary perception mode focuses on collecting information about human posture. The combination of these two modes expands the scope of behavioral environmental perception, comprehensively collecting information about human behavior, and effectively improving the efficiency of behavioral perception.

[0026] 3. The present invention supports the precise recognition of an individual's diverse behavioral postures in urban spaces through an inertial measurement unit, and judges the individual's precise posture based on the individual's limb acceleration, thus avoiding the problems of limited field of view and long time consumption of traditional posture perception methods that rely on image feature recognition.

[0027] 4. This invention uses panoramic and local cameras to capture spatial images, adding embodied dimension information to spatial environment perception. The panoramic camera captures 360° panoramic images and calculates spatial environment indicators; the line-of-sight camera captures local spatial images from the subject's perspective and calculates visual environment indicators.

[0028] 5. The present invention establishes a regional behavioral environment sandbox through virtual reconstruction of multi-device perception data, forming a behavior-environment analysis database for urban areas of different scales, providing a basis for the spatial and behavioral characteristics analysis of urban areas.

[0029] 6. The present invention establishes a parameter system for behavior and environmental perception and combines the correlation characteristics between the collected data analysis parameters to establish a correlation model, which is convenient for exploring the perception mode and behavioral patterns of the public towards the urban environment.

[0030] 7. The present invention uses three-dimensional holographic projection technology to achieve visual expression and interactive feedback of perception data. The high-update frequency and high-granularity three-dimensional images enable planners and decision makers to understand the behavior-environment interaction relationship of the overall and local spaces of the area more clearly and effectively, providing scientific support for subsequent planning decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a flow chart of the method of the present invention;

[0032] Figure 2 It is a schematic diagram of individual behavior perception information;

[0033] Figure 3 It is a schematic diagram of spatial environment perception information. DETAILED DESCRIPTION

[0034] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0035] Example

[0036] like Figure 1-3 As shown, the technical solution of the present invention will be described in detail below with reference to a community in the old urban area of ​​a certain city as an example.

[0037] (1) Residents of different age groups in the community were selected as subjects. The subjects were equipped with the wearable components of the sensing device, including sensing glasses and smart bracelets. The mode adjustment system was used to adjust the device to the on state.

[0038] (2) The mobile positioning module automatically switches between dynamic and static sensing modes based on the individual's speed change characteristics. In dynamic sensing mode, the mobile positioning component records the individual's GPS positioning data at unit time intervals. In static sensing mode, the mobile positioning component records the individual's limb instantaneous acceleration data. The behavior sensing system analyzes and integrates dynamic and static behavior indicator information to form an individual behavior information database, and transmits real-time data to the data center via the cellular network.

[0039] The mobile positioning component is composed of a GPS locator, a timer, and an inertial measurement unit, wherein the GPS locator obtains and records the individual's real-time geographic information position, the timer records time information, and the inertial measurement unit records the acceleration of the individual's limbs.

[0040] The automatic switching of the dynamic perception mode or the static perception mode according to the individual speed change characteristics refers to the automatic switching of the dynamic perception mode or the static perception mode according to the individual speed change characteristics. △v=vmin max Calculate the individual speed standard deviation and speed change rate, and then calculate the speed fluctuation index S = α·σ v +β·△v, where α and β are weight coefficients. When the speed fluctuation index is greater than the threshold, it switches to the dynamic perception mode; when the speed fluctuation index is less than the threshold, it switches to the static perception mode.

[0041] The dynamic behavior index and static behavior index information, the dynamic behavior index includes the speed and GPS coordinates of the individual, the static behavior index information includes the posture time ratio and GPS coordinates, the posture includes standing, sitting and lying, the limb acceleration data recorded by the behavior perception system combined with the inertial measurement unit, Calculate the pitch and roll angles of the limbs, where (a x ,a y ,a z ) represents the acceleration components in the X, Y, and Z axis directions, and the posture type of the individual is determined according to the threshold range of the pitch angle and roll angle, and then the Calculate the proportion of time spent in each posture.

[0042] (3) Through the image acquisition module, global spatial images are collected in dynamic perception mode, and global spatial images and images of the visible range of the individual's line of sight are collected simultaneously in static perception mode. The environmental perception system analyzes and integrates spatial environmental indicators and visual environmental indicators to form a spatial environmental information database, and transmits real-time data to the data center via the cellular network.

[0043] The image acquisition component includes a panoramic camera and a local camera. The panoramic camera acquires 360° panoramic spatial images, and the local camera records spatial images within the individual field of view at the perspective of the human eye.

[0044] The integrated spatial environment index and visual environment index information refers to processing the spatial image collected by the image acquisition component through the semantic segmentation algorithm, identifying green vegetation, water bodies, buildings, roads, sky, open space, and obstacle facilities in the panoramic spatial image, and calculating the number of pixels N occupied by each semantic i and through Calculate the proportion of each type of pixel as a spatial environment indicator. Identify green vegetation, water bodies, buildings, roads, open spaces, and obstacles in the visible range image. Similarly, calculate the proportion of each type of semantic pixel as a visual environment indicator.

[0045] (4) Through the back-end analysis module, the digital sand table system integrates spatial environment data and establishes a regional three-dimensional spatial sand table through a three-dimensional reconstruction method. The behavior modeling system integrates individual behavior data, virtually reconstructs the spatiotemporal behavior of each subject, and maps the reconstructed virtual individual to the correct position on the three-dimensional spatial sand table through its GPS positioning. The integrated analysis system comprehensively analyzes the individual behavior information and spatial environment information parameters, and establishes an intelligent agent model based on the correlation characteristics of the parameters.

[0046] The comprehensive analysis of individual behavior information and spatial environment information parameters, and the establishment of an intelligent perception and decision-making model based on the parameter correlation characteristics of the region, refers to dividing the region into 10m*10m grid units, and calculating the average value of the behavior index B in each grid = [b1, b2, ..., b n ] and the average value of spatial index E=[e1,e2,...,e m ], using a neural network to calculate the correlation model B = h(E) + ε between behavioral indicators and spatial indicators. The correlation model is embedded in the intelligent agent system, which receives spatial environment indicators as input, outputs individual behavioral indicators, and guides behavioral decision-making.

[0047] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention, and such changes and modifications fall within the scope of the invention as claimed.

Claims

1. An embodied intelligent perception method based on a wearable mobile positioning image acquisition device, characterized in that: The following steps are involved: Step 1: Wear the device and set the mode The subjects are equipped with wearable components of the sensing device, including sensing glasses and a smart bracelet; the mode adjustment system is used to adjust the device to the on state, in which the device can automatically switch between sensing modes and acquire and transmit multiple types of sensing data; Step 2: Mobile positioning module behavior perception The mobile positioning module includes a mobile positioning component and a behavior perception system, which automatically switches between dynamic perception mode and static perception mode according to the individual's speed change characteristics. In dynamic perception mode, the mobile positioning component records the individual's GPS positioning data per unit time interval; in static perception mode, the mobile positioning component records the individual's limb instantaneous acceleration data. The behavior perception system analyzes and integrates dynamic behavior indicators and static behavior indicator information to form an individual behavior information database, and transmits real-time data to the data center via the cellular network. Step 3: Image acquisition module environmental perception The image acquisition module includes an image acquisition component and an environmental perception system. In dynamic perception mode, the image acquisition component collects global spatial images. In static perception mode, the image acquisition component simultaneously collects global spatial images and images of the visible range in the direction of individual line of sight. The environmental perception system analyzes and integrates spatial environmental indicators and visual environmental indicators to form a spatial environmental information database, and transmits real-time data to the data center via the cellular network. Step 4: Embodied Perception Modeling in the Backend Analysis Module The back-end analysis module includes a server component, a digital sandbox system, a behavior modeling system, and an integrated analysis system. The digital sandbox system integrates spatial environment data and creates a regional three-dimensional sandbox through three-dimensional reconstruction methods. The behavior modeling system integrates individual behavior data, virtually reconstructs the spatiotemporal behavior of each subject, and maps the reconstructed virtual individual to the correct position on the three-dimensional sandbox using its GPS positioning. The integrated analysis system comprehensively analyzes individual behavior information and spatial environment information parameters, and establishes an intelligent agent model based on the correlation characteristics of the parameters. Step 5: Visualization platform integration display The data results output by the back-end analysis module are displayed through a holographic digital projector, gesture recognizer and visualization system, including the regional behavior environment reconstruction sandbox and the intelligent agent deduction sandbox.

2. The embodied intelligent perception method based on a wearable mobile positioning image acquisition device according to claim 1 is characterized in that: The mobile positioning component in step 2 is composed of a GPS locator, a timer, and an inertial measurement unit, wherein the GPS locator obtains and records the individual's real-time geographic information position, the timer records time information, and the inertial measurement unit records the acceleration of the individual's limbs.

3. The embodied intelligent perception method based on a wearable mobile positioning image acquisition device according to claim 2 is characterized in that: In the second step, the dynamic perception mode or the static perception mode is automatically switched according to the individual's speed change characteristics. △v=v max -v min Calculate the individual speed standard deviation and speed change rate, where v i is the speed of the ith time period, n is the number of time periods, is the average velocity, v max and v min It is the maximum speed and minimum speed in a certain period of time; then calculate the speed fluctuation index S = α·σ v +β·△v, where α and β are weight coefficients. When the speed fluctuation index is greater than the threshold, it switches to the dynamic perception mode; when the speed fluctuation index is less than the threshold, it switches to the static perception mode.

4. The embodied intelligent perception method based on a wearable mobile positioning image acquisition device according to claim 3 is characterized in that: The dynamic behavior index and static behavior index information in step 2 include the individual's speed and GPS coordinates, and the static behavior index information includes the posture time ratio and GPS coordinates. The posture includes standing, sitting and lying postures. The limb acceleration data recorded by the behavior perception system combined with the inertial measurement unit is obtained through Calculate the pitch and roll angles of the limbs, where (a x ,a y ,a z ) represents the acceleration components in the X, Y, and Z axis directions, and the posture type of the individual is determined according to the threshold range of the pitch angle and roll angle, and then the Calculate the proportion of each posture time, where T total is the total time from the start of recording to the end of recording, is the sum of all different posture times.

5. The embodied intelligent perception method based on a wearable mobile positioning image acquisition device according to claim 4 is characterized in that: The image acquisition component in step three includes a panoramic camera and a local camera. The panoramic camera acquires 360° panoramic spatial images, and the local camera records spatial images of individual visual ranges at the perspective of human eyes.

6. The embodied intelligent perception method based on a wearable mobile positioning image acquisition device according to claim 5 is characterized in that: The integration of spatial environment indicators and visual environment indicator information in step 3 refers to processing the spatial images collected by the image acquisition component through the semantic segmentation algorithm, identifying green vegetation, water bodies, buildings, roads, sky, open spaces, and obstacle facilities in the panoramic spatial image, and calculating the number of pixels N occupied by each semantic i and through Calculate the proportion of each type of pixels as a spatial environment indicator, where N tatal Represents the total number of pixels in the image; identifies green vegetation, water bodies, buildings, roads, open spaces, and obstacle facilities in the visible range image, and similarly calculates the proportion of each type of semantic pixels as a visual environment indicator.

7. The embodied intelligent perception method based on a wearable mobile positioning image acquisition device according to claim 6 is characterized in that: In the fourth step, the individual behavior information and spatial environment information parameters are comprehensively analyzed, and an intelligent perception and decision-making model is established according to the parameter correlation characteristics of the region, which means dividing the region into 10m*10m grid units and calculating the average value vector B of the behavior index in each grid = [b1, b2, ..., b n ] and the spatial index average vector E=[e1,e2,...,e m ], where b1, b2, ..., b n Represents n different behavioral indicators in the first grid, e1, e2, ..., e m Representing m different spatial environmental indicators within the grid, the correlation model B = f(E) + ε between behavioral indicators and spatial indicators is calculated through a neural network, where ε is the error term; the correlation model is embedded in the intelligent agent system, which receives spatial environmental indicators as input, outputs individual behavioral indicators and guides behavioral decisions.

Images