Indoor personnel identity and activity track identification method and system

Abstract

The invention relates to the technical field of intelligent sensing and Internet of Things technologies, in particular to an indoor personnel identity and activity track recognition method and system, and the system comprises a pressure sensing module, an infrared monitoring module, a data queue, a pressure feature registration module, a data processing and identity recognition module, and a track construction and behavior analysis module. The pressure sensing module is composed of a plurality of pressure detection ground mats integrated with high-precision pressure sensors, and the pressure detection ground mats are fixedly laid on the inner side of an entrance door, a bedroom doorway, a bathroom doorway and a kitchen doorway which must pass through; the pressure detection ground mat is used for measuring a characteristic pressure value generated by a person stepping on the pressure detection ground mat and a sole characteristic pressure distribution profile; key identity features are obtained through the arranged pressure ground mat, seamless trajectory tracking is carried out in combination with the widely-deployed infrared sensor, and therefore non-inductive, privacy-violation-free and high-reliability indoor person recognition and positioning are achieved.

Description

Technical Field

[0001] This invention relates to the technical field of intelligent sensing and Internet of Things technology, and in particular to a method and system for identifying the identity and activity trajectory of people indoors. Background Technology

[0002] In fields such as smart homes and elderly care monitoring, accurately identifying indoor occupants and tracking their activities is crucial for providing personalized services (such as smart lighting and temperature control), safety monitoring (such as fall detection for the elderly), and energy management. Currently, mainstream technical solutions have the following limitations: 1. Video vision solutions (such as cameras and facial recognition): This is the most intuitive method, but it poses a serious risk of privacy leakage, making users (especially in the private environment of the home) feel uncomfortable and resistant. At the same time, it has high computational complexity and is greatly affected by lighting conditions. 2. Wireless signal solutions (such as Bluetooth beacons and Wi-Fi fingerprints): These methods locate devices by using the signal strength of the terminal device. However, the accuracy of these methods is greatly affected by multipath effects and signal interference, resulting in poor stability. More importantly, they require users to carry specific devices (such as mobile phones or smartwatches). For elderly people and children who do not often carry mobile phones or are prone to forgetting their devices, these methods have low reliability and create blind spots in the monitoring process. 3. Single sensor solution (e.g., using only infrared or pressure): A single passive infrared sensor can only detect movement within a certain area, but cannot distinguish which specific person it is; A single pressure sensor can detect stampede events, but if deployed in a dispersed manner, it is difficult to form a coherent trajectory; if deployed over a large area, the cost is high, and it is difficult to uniquely identify the perpetrator based solely on instantaneous pressure data.

[0003] Therefore, existing technologies lack an effective means to distinguish different indoor occupants and map their activity trajectories in a low-cost and highly reliable manner while strictly protecting privacy and without requiring users to wear any devices. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides an indoor personnel identification and activity trajectory recognition method and system that acquires key identity features through the placement of pressure mats and combines them with widely deployed infrared sensors for seamless trajectory tracking, thereby achieving seamless, privacy-free, and highly reliable indoor personnel identification and positioning.

[0005] The present invention provides an indoor personnel identification and activity trajectory recognition system, comprising a pressure sensing module, an infrared monitoring module, a data queue, a pressure feature registration module, a data processing and identification module, and a trajectory construction and behavior analysis module; Pressure sensing module: Composed of multiple pressure detection mats with integrated high-precision pressure sensors, the pressure detection mats are fixedly laid on the inside of the entrance door, bedroom door, bathroom door, and kitchen door of the indoor space. The pressure detection mats are used to measure the characteristic pressure value and foot pressure distribution profile generated by people stepping on them. Infrared monitoring module: It consists of multiple human infrared sensors deployed in various functional spaces indoors. Each human infrared sensor is responsible for monitoring human movement events within its coverage area. Data queue: Data from the pressure sensing module and the infrared monitoring module forms a data stream and is sent to the data queue. The data queue then distributes the data to the pressure feature registration module and the data processing and identification module. Stress Feature Registration Module: Used for the establishment and updating of the identity feature database. Through pre-entry or learning over a period of time, it establishes stress feature profiles of indoor resident personnel, including stress amplitude and duration curves. Data processing and identification module: Communicates with the pressure sensing module and infrared monitoring module, including identification module and spatiotemporal information marking module; Identity Recognition Module: When someone steps on the pressure mat, this module receives pressure data and determines the identity of the person passing through by matching the real-time pressure value with the feature profiles in the identity feature database and the update module. Spatiotemporal information tagging module: used to record the timestamp and pressure detection mat location number when identity recognition is successful; The trajectory construction and behavior analysis module includes an identity information transmission and association unit and an activity trajectory construction unit; Identity information transmission and association unit: It associates the identity identifier from the data processing and identity recognition module with the detection data from the infrared monitoring module and the pressure sensing module. When the pressure detection mat at the entrance door identifies person A, the human infrared sensor in the living room is triggered. The system then determines that person A is currently active in the living room, thereby realizing the determination of the person entering the space and the determination of the person entering and leaving the space. Activity trajectory construction unit: Based on the spatiotemporal sequence of infrared sensor triggers and the result of the last successful identity recognition, dynamically construct and update the continuous activity trajectory of the person indoors.

[0006] Preferably, it also includes an abnormal behavior judgment unit; Abnormal Behavior Judgment Unit: Based on the activity trajectory constructed by the activity trajectory construction unit, further analyze the behavior patterns of personnel, including judging whether the elderly linger in the bathroom for too long or wander around abnormally at night.

[0007] Preferably, it also includes a trajectory visualization and output module; Trajectory visualization and output module: The system builds and displays the activity trajectories of people with different identities in the room in real time in the background, and can provide data interfaces to smart home platforms and monitoring applications.

[0008] Preferably, it also includes an OIT gateway; It also includes an OIT gateway: the pressure sensing module and the infrared monitoring module send the data to the data queue after forming a data stream through the OIT gateway.

[0009] Preferably, a method for identifying the identity and activity trajectory of people indoors is characterized by comprising the following steps: S1. Deployment and initialization: Lay pressure detection mats on critical indoor paths, install human infrared sensors in various functional spaces, and establish or learn pressure characteristic profiles of resident personnel in the system. S2. Identity recognition trigger: When a person steps on any pressure detection mat, the pressure sensing module collects pressure data and uploads it. S3. Identity Matching: The data processing module compares the collected stress data with the pre-stored feature profiles to identify the current person's identity and generate an identity event with a timestamp and location. S4. Trajectory Tracking and Correlation: S4.1 The system will assign the identified identity ID to the person; S4.2 By monitoring the triggering sequence and time of each person's infrared sensor, the system tracks the person's movement path. The system logic is as follows: after the most recent identity recognition, movement events occurring indoors are attributed to that identity by default until a person is identified as a new identity on any pressure mat. In this way, continuous, identity-based trajectory filling is achieved from one recognition point to the next.

[0010] Preferably, determining whether a person has entered the space includes the following steps: S11. Real-time monitoring of relevant data within the space; S12. When someone enters the infrared sensing range, a 30-second timer is activated, and then the process returns to S11 to continue monitoring. S13. Determine if there is space pressure data. If no pressure data is detected, the process returns directly to S11. If pressure data is detected, it indicates that the pressure mat has detected a pressure change, and the process proceeds to determine if there is a 30-second space marker. S14. Determine if a 30-second spatial marker exists. If a 30-second spatial marker exists, it means that a 30-second timed marker has been initiated and is still in the timed period. The process returns to S11 to continue monitoring. If no 30-second spatial marker exists, the process proceeds to the step of matching the identity from the stress feature database. The system will call the pre-stored feature profiles and match the personnel identity according to the stress data. After the identity matching is completed, a 10-second timed marker is initiated. S1.5 After entering the judgment node of whether infrared data is generated in the space within 10 seconds, if no, that is, no infrared data is detected within 10 seconds, the process returns to S11 to continue monitoring. If yes, it means that infrared data is detected within 10 seconds. Combined with pressure data and identity matching results, it is determined that the designated person has entered this space. Then the process enters the termination step, and this judgment process ends. By using infrared sensors and pressure mats to monitor spatial data in real time, combined with 30-second and 10-second timing mechanisms and identity matching using a pressure feature database, the system can accurately determine whether an elderly person has entered a specific space. This design effectively reduces false alarms and improves the accuracy and reliability of space entry detection.

[0011] Preferably, determining whether a person has left the space includes the following steps: S21. Real-time monitoring of relevant data within the space; S22. Determine if there is spatial infrared data. If so, it means that someone is active in the space. At this time, start a 10-second timer. Then proceed to the next step to determine if spatial pressure data is generated within 10 seconds. If not, it means that no infrared data was detected. Clear this spatial timer and return to S21 to continue monitoring. S23. Determine whether spatial pressure data is generated within 10 seconds. If yes, it means that the pressure detection mat detected a pressure change within 10 seconds, indicating that someone is moving or staying in the space. At this time, start the 10-second timer again, and then proceed to the next judgment to determine whether infrared light is generated in the space within 30 seconds. If no, that is, no pressure data is detected within 10 seconds, clear this space timer and return to S21 to continue monitoring. S24: If infrared radiation is detected in the space within 30 seconds, and if so, it indicates continuous detection of infrared data within 30 seconds. The system then proceeds to the identity matching step from the pressure feature database. This involves calling the pre-stored pressure feature database, matching the person's identity based on the pressure data, and then determining if the designated person has left the space. The process then terminates. If not, meaning no infrared data was detected within 30 seconds, the space is cleared after a timer and the system returns to S21 to continue monitoring space data. By using infrared sensors and pressure mats to monitor space data in real time, combined with 10-second and 30-second timer mechanisms and identity matching from the pressure feature database, the system accurately determines whether an elderly person has left a specific space. Through judgments under different conditions and a timer clearing mechanism, the system effectively avoids misjudgments and invalid judgments, improving the accuracy and reliability of determining whether an elderly person has left the space.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. Ultimate privacy protection: No cameras or recording devices are used throughout the process. Only non-visual and non-auditory sensors such as pressure and infrared are used, which fundamentally eliminates the risk of privacy leakage and provides a user-friendly experience. 2. No device required: Personnel do not need to carry any terminal or tag, achieving truly contactless identification and tracking, solving the problems of users forgetting or charging, and is especially suitable for the elderly and children; 3. High recognition accuracy and high reliability: It uses pressure characteristics, a stable biometric feature, for differentiation, which is more stable and accurate than wireless signals. By setting "identity recognition points" along the necessary paths, it ensures that the system can periodically and with high confidence confirm the identity of personnel and correct any cumulative errors that may occur in trajectory association. 4. Low cost and easy deployment: The pressure and infrared sensors used are mature, low-cost IoT hardware. The pressure mats do not need to be laid throughout the house, but only deployed at key nodes, which greatly reduces system costs and installation complexity. 5. The system has good robustness: the pressure and infrared sensors complement each other. The pressure sensor provides high-confidence identification information but covers a point, while the infrared sensor provides continuous spatial coverage but does not provide identification information. The fusion of the two realizes the combination of "point (identity) and line (trajectory)" to build a complete and reliable monitoring network. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the system structure of the present invention; Figure 2 This is a schematic diagram of the installation of the pressure detection mat and the human infrared sensor; Figure 3 This is a flowchart illustrating the process of determining when people enter a space; Figure 4This is a flowchart illustrating the process of determining when people leave a space; Figure 5 This is a schematic diagram of the spatial entry determination timing; Figure 6 This is a schematic diagram of the spatial entry determination timing; Figure 7 This is a schematic diagram of the spatial exit determination timing; Figure 8 This is a schematic diagram of the timing for determining when a space exits.

[0014] The attached diagram is labeled as follows: 101, pressure detection mat; 102, human infrared sensor. Detailed Implementation

[0015] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. The present invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete. Example

[0016] like Figures 1 to 8 As shown, an indoor personnel identification and activity trajectory recognition system of the present invention includes a pressure sensing module, an infrared monitoring module, a data queue, a pressure feature registration module, a data processing and identification module, and a trajectory construction and behavior analysis module; Pressure sensing module: Composed of multiple pressure detection mats 101 with integrated high-precision pressure sensors. The pressure detection mats 101 are fixedly laid on the inside of the entrance door, bedroom door, bathroom door, and kitchen door of the indoor space. The pressure detection mats 101 are used to measure the characteristic pressure value and the characteristic pressure distribution profile of the foot sole generated by people stepping on them. Infrared monitoring module: It consists of multiple human infrared sensors 102 deployed in various functional spaces indoors. Each human infrared sensor 102 is responsible for monitoring human movement events within its coverage area. Data queue: Data from the pressure sensing module and the infrared monitoring module forms a data stream and is sent to the data queue. The data queue then distributes the data to the pressure feature registration module and the data processing and identification module. Stress Feature Registration Module: Used for the establishment and updating of the identity feature database. Through pre-entry or learning over a period of time, it establishes stress feature profiles of indoor resident personnel, including stress amplitude and duration curves. Data processing and identification module: Communicates with the pressure sensing module and infrared monitoring module, including identification module and spatiotemporal information marking module; Identity Recognition Module: When someone steps on the pressure mat, this module receives pressure data and determines the identity of the person passing through by matching the real-time pressure value with the feature profiles in the identity feature database and the update module. Spatiotemporal information marking module: used to record the timestamp and pressure detection mat 101 location number when identity recognition is successful; The trajectory construction and behavior analysis module includes an identity information transmission and association unit and an activity trajectory construction unit; Identity information transmission and association unit: The identity identifier from the data processing and identity recognition module is associated with the detection data from the infrared monitoring module and the pressure sensing module. When the pressure detection mat 101 of the entrance door identifies person A, the human infrared sensor 102 in the living room is triggered. The system then determines that the person currently active in the living room is person A, thereby realizing the determination of the person entering the space and the determination of the person entering and leaving the space. Activity trajectory construction unit: Based on the spatiotemporal sequence of infrared sensor triggers and the result of the last successful identity recognition, dynamically construct and update the continuous activity trajectory of the person indoors; It also includes an abnormal behavior judgment unit; Abnormal Behavior Judgment Unit: Based on the activity trajectory constructed by the activity trajectory construction unit, further analyze the behavior patterns of personnel, including judging whether the elderly linger in the bathroom for too long or wander around abnormally at night; In this embodiment, 1. Ultimate privacy protection: No cameras or recording devices are used throughout the process. Only non-visual and non-auditory sensors such as pressure and infrared are used, which fundamentally eliminates the risk of privacy leakage and provides a user-friendly experience. 2. No device required: Personnel do not need to carry any terminal or tag, achieving truly contactless identification and tracking, solving the problems of users forgetting or charging, and is especially suitable for the elderly and children; 3. High recognition accuracy and high reliability: It uses pressure as a stable biometric feature for differentiation, which is more stable and accurate than wireless signals. By setting "identity recognition points" on the necessary paths, it ensures that the system can periodically and with high confidence confirm the identity of personnel and correct the cumulative errors that may occur in trajectory association. 4. Low cost and easy deployment: The pressure and infrared sensors used are mature, low-cost IoT hardware. The pressure mats do not need to be laid throughout the house, but only deployed at key nodes, which greatly reduces system costs and installation complexity. 5. The system has good robustness: the pressure and infrared sensors complement each other. The pressure sensor provides high-confidence identification information but covers a point, while the infrared sensor provides continuous spatial coverage but does not provide identification information. The fusion of the two realizes the combination of "point (identity) and line (trajectory)" to build a complete and reliable monitoring network. Example

[0017] Based on Embodiment 1, the indoor personnel identification and activity trajectory recognition system of the present invention further includes a trajectory visualization and output module; Trajectory visualization and output module: The system builds and displays the activity trajectories of people with different identities indoors in real time in the background, and can provide data interfaces to smart home platforms and monitoring applications; It also includes OIT gateways; It also includes an OIT gateway: the pressure sensing module and the infrared monitoring module send the data to the data queue after forming a data stream through the OIT gateway. Example

[0018] A method for identifying the identity and activity trajectory of people indoors, characterized by comprising the following steps: S1. Deployment and initialization: Lay pressure detection mats 101 on critical indoor paths, install human infrared sensors 102 in each functional space, and establish or learn pressure characteristic profiles of resident personnel in the system. S2, Identity Recognition Trigger: When a person steps on any pressure detection mat 101, the pressure sensing module collects pressure data and uploads it; S3. Identity Matching: The data processing module compares the collected stress data with the pre-stored feature profiles to identify the current person's identity and generate an identity event with a timestamp and location. S4. Trajectory Tracking and Correlation: S4.1 The system will assign the identified identity ID to the person; S4.2 By monitoring the triggering sequence and time of each person's infrared sensor 102, the system tracks the person's movement path. The system logic is as follows: after the most recent identity recognition, indoor movement events are attributed to that identity by default until a person is identified as a new identity on any pressure mat. In this way, continuous, identity-based trajectory filling is achieved from one recognition point to the next. The determination of whether a person has entered the space includes the following steps: S11. Real-time monitoring of relevant data within the space; S12. When someone enters the infrared sensing range, a 30-second timer is activated, and then the process returns to S11 to continue monitoring. S13. Determine if there is space pressure data. If no pressure data is detected, the process returns directly to S11. If pressure data is detected, it indicates that the pressure mat has detected a pressure change, and the process proceeds to determine if there is a 30-second space marker. S14. Determine if a 30-second spatial marker exists. If a 30-second spatial marker exists, it means that a 30-second timed marker has been initiated and is still in the timed period. The process returns to S11 to continue monitoring. If no 30-second spatial marker exists, the process proceeds to the step of matching the identity from the stress feature database. The system will call the pre-stored feature profiles and match the personnel identity according to the stress data. After the identity matching is completed, a 10-second timed marker is initiated. S1.5 After entering the judgment node of whether infrared data is generated in the space within 10 seconds, if no, that is, no infrared data is detected within 10 seconds, the process returns to S11 to continue monitoring. If yes, it means that infrared data is detected within 10 seconds. Combined with pressure data and identity matching results, it is determined that the designated person has entered this space. Then the process enters the termination step, and this judgment process ends. The determination of whether a person has left the space includes the following steps: S21. Real-time monitoring of relevant data within the space; S22. Determine if there is spatial infrared data. If so, it means that someone is active in the space. At this time, start a 10-second timer. Then proceed to the next step to determine if spatial pressure data is generated within 10 seconds. If not, it means that no infrared data was detected. Clear this spatial timer and return to S21 to continue monitoring. S23. Determine whether spatial pressure data is generated within 10 seconds. If yes, it means that the pressure detection mat 101 detected a pressure change within 10 seconds, indicating that someone is moving or staying in the space. At this time, start the 10-second timer again, and then proceed to the next step of determining whether infrared light is generated in the space within 30 seconds. If no, that is, no pressure data is detected within 10 seconds, clear this space timer and return to S21 to continue monitoring. S24. Is there infrared radiation in the space within 30 seconds? If yes, it means that infrared data has been continuously detected within 30 seconds. Proceed to the step of matching the identity from the pressure feature library. Call the pre-stored pressure feature library, match the identity of the person according to the pressure data, and then determine that the person with the specified identity has left this space. The process enters the termination step and the current determination process ends. If no, that is, no infrared data has been detected within 30 seconds, clear this space timer and return to S21 to continue monitoring space data. By using infrared sensors and pressure mats to monitor spatial data in real time, combined with 30-second and 10-second timing mechanisms and identity matching using a pressure feature database, the system can accurately determine whether an elderly person has entered a specific space. This design can effectively reduce false alarms and improve the accuracy and reliability of space entry detection. By using infrared sensors and pressure mats to monitor spatial data in real time, combined with 10-second and 30-second timer mechanisms and identity matching from a pressure feature database, the system accurately determines whether an elderly person has left a specific space. Through judgments under different conditions and a timed clearing mechanism, the system can effectively avoid false and invalid judgments, improving the accuracy and reliability of determining whether an elderly person has left the space. Example

[0019] Implement this invention in a household where elderly people and children live together: 1. Deployment: A pressure testing mat was placed inside the front door, at the entrance to the elderly person's bedroom, and at the entrance to the bathroom. Human infrared sensors are installed in the living room, kitchen, hallway, elderly bedroom, and bathroom. All sensors connect to the home gateway via Zigbee / Wi-Fi, and the gateway has a built-in or connected data processing and identification module. 2. Initialization: During system initialization, the elderly and their children step on each pressure detection mat multiple times, record and store their average pressure values, and establish a feature profile (including pressure amplitude and change duration curves, and the characteristic duration of the curves also varies from person to person). 3. Workflow: Scenario 1 (Distinguishing Members): In the morning, the mother steps on the pressure detection mat at the front door and enters the living room. The system identifies her as "Mother". Subsequently, the infrared human body sensor in the living room is triggered, and the system determines that "Mother is active in the living room". A few minutes later, the father steps on the pressure detection mat at the bedroom door and enters the bathroom. The system identifies him as "Father" and updates the status to "Father is in the bathroom". At this time, the system can clearly know the location of each member. Scenario 2 (Elderly Monitoring): At night, the system detects that the elderly person steps on the pressure detection mat in the bedroom and enters the bathroom. If the human infrared sensor in the bathroom does not trigger any movement within the next 30 minutes, the system can determine that the elderly person may be "abnormally staying" and then send an alarm message to the children's mobile phones, informing them that "father has stayed in the bathroom for too long". Trajectory reconstruction: By recording the complete sequence of the "father" from the "bedroom pressure detection mat" to the "corridor human infrared sensor" to the "bathroom human infrared sensor", the system can reconstruct the complete path of the elderly man from getting up to going to the bathroom; Through the above methods, this invention achieves accurate differentiation of the identities of different people indoors and reliable tracking of their activity trajectories while perfectly protecting privacy.

[0020] The pressure detection mat 101 and human infrared sensor 102 of the indoor personnel identification and activity trajectory recognition method and system of the present invention are commercially available. Technical personnel in this industry only need to install and operate them according to the accompanying instruction manual, without requiring any creative work from those skilled in the art.

[0021] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An indoor personnel identification and activity trajectory recognition system, characterized in that, It includes a pressure sensing module, an infrared monitoring module, a data queue, a pressure feature registration module, a data processing and identity recognition module, and a trajectory construction and behavior analysis module; Pressure sensing module: Composed of multiple pressure detection mats (101) with integrated high-precision pressure sensors. The pressure detection mats (101) are fixedly laid on the inside of the entrance door, bedroom door, bathroom door, and kitchen door of the indoor space. The pressure detection mats (101) are used to measure the characteristic pressure value and foot characteristic pressure distribution profile generated by people stepping on them. Infrared monitoring module: It consists of multiple human infrared sensors (102) deployed in various functional spaces indoors. Each human infrared sensor (102) is responsible for monitoring human movement events within its coverage area. Data queue: Data from the pressure sensing module and the infrared monitoring module forms a data stream and is sent to the data queue. The data queue then distributes the data to the pressure feature registration module and the data processing and identification module. Stress Feature Registration Module: Used for the establishment and updating of the identity feature database. Through pre-entry or a period of learning, it establishes a stress feature profile of indoor resident personnel, including baseline stress amplitude and duration curves of stress variation. Data processing and identification module: Communicates with the pressure sensing module and infrared monitoring module, including identification module and spatiotemporal information marking module; Identity Recognition Module: When someone steps on the pressure mat, this module receives pressure data and determines the identity of the person passing through by matching the real-time pressure value with the feature profiles in the identity feature database and the update module. Spatiotemporal information marking module: used to record the timestamp of successful identity recognition and the location number of the pressure detection mat (101); The trajectory construction and behavior analysis module includes an identity information transmission and association unit and an activity trajectory construction unit; Identity information transmission and association unit: The identity identifier from the data processing and identity recognition module is associated with the detection data from the infrared monitoring module and the pressure sensing module. When the pressure detection mat (101) of the entrance door identifies person A, the human infrared sensor (102) in the living room is triggered. The system then determines that the person currently active in the living room is person A, thereby realizing the determination of the person entering the space and the determination of the person entering and leaving the space. Activity trajectory construction unit: Based on the spatiotemporal sequence of infrared sensor triggers and the result of the last successful identity recognition, dynamically construct and update the continuous activity trajectory of the person indoors.

2. The indoor personnel identification and activity trajectory recognition system as described in claim 1, characterized in that, It also includes an abnormal behavior judgment unit; Abnormal Behavior Judgment Unit: Based on the activity trajectory constructed by the activity trajectory construction unit, further analyze the behavior patterns of personnel, including judging whether the elderly linger in the bathroom for too long or wander around abnormally at night.

3. The method and system for indoor personnel identification and activity trajectory recognition as described in claim 1, characterized in that, It also includes a trajectory visualization and output module; Trajectory visualization and output module: The system builds and displays the activity trajectories of people with different identities indoors in real time in the background, and can provide data interfaces to smart home platforms and monitoring applications.

4. The indoor personnel identification and activity trajectory recognition system as described in claim 1, characterized in that, It also includes OIT gateways; It also includes an OIT gateway: the pressure sensing module and the infrared monitoring module send the data to the data queue after forming a data stream through the OIT gateway.

5. A method for identifying the identity and activity trajectory of people indoors, characterized in that, Includes the following steps: S1. Deployment and initialization: Lay pressure detection mats (101) on critical indoor paths, install human infrared sensors (102) in each functional space, and establish or learn the pressure characteristic profiles of resident personnel in the system. S2, Identity Recognition Trigger: When a person steps on any pressure detection mat (101), the pressure sensing module collects pressure data and uploads it; S3. Identity Matching: The data processing module compares the collected stress data with the pre-stored feature profiles to identify the current person's identity and generate an identity event with a timestamp and location. S4. Trajectory Tracking and Correlation: S4.1 The system will assign the identified identity ID to the person; S4.2 By monitoring the triggering sequence and time of each person's infrared sensor (102), the system tracks the person's movement path. The system logic is as follows: after the most recent identity recognition, the movement event that occurs indoors is attributed to that identity by default until the next time a person is identified as a new identity on any pressure mat. In this way, continuous, identity-based trajectory filling is achieved from one recognition point to the next recognition point.

6. The indoor personnel identification and activity trajectory recognition system as described in claim 1, characterized in that, The determination of whether a person has entered the space includes the following steps: S11. Real-time monitoring of relevant data within the space; S12. When someone enters the infrared sensing range, a 30-second timer is activated, and then the process returns to S11 to continue monitoring. S13. Determine if there is space pressure data. If no pressure data is detected, the process returns directly to S11. If pressure data is detected, it indicates that the pressure mat has detected a pressure change, and the process proceeds to determine if there is a 30-second space marker. S14. Determine if a 30-second spatial marker exists. If a 30-second spatial marker exists, it means that a 30-second timed marker has been initiated and is still in the timed period. The process returns to S11 to continue monitoring. If no 30-second spatial marker exists, the process proceeds to the step of matching the identity from the stress feature database. The system will call the pre-stored feature profiles and match the personnel identity according to the stress data. After the identity matching is completed, a 10-second timed marker is initiated. S1.5 After entering the judgment node for whether infrared data is generated in the space within 10 seconds, if no infrared data is detected within 10 seconds, the process returns to S11 to continue monitoring. If infrared data is detected within 10 seconds, it means that infrared data was detected within 10 seconds. Combining the pressure data and identity matching results, it is determined that the designated person has entered this space, and then the process enters the termination step, and this judgment process ends.

7. The indoor personnel identification and activity trajectory recognition system as described in claim 1, characterized in that, The determination of whether a person has left the space includes the following steps: S21. Real-time monitoring of relevant data within the space; S22. Determine if there is spatial infrared data. If so, it means that someone is active in the space. At this time, start a 10-second timer. Then proceed to the next step to determine if spatial pressure data is generated within 10 seconds. If not, it means that no infrared data was detected. Clear this spatial timer and return to S21 to continue monitoring. S23. Determine whether spatial pressure data is generated within 10 seconds. If yes, it indicates that the pressure detection mat (101) detected a pressure change within 10 seconds, and someone is moving or staying in the space. At this time, start the 10-second timer again, and then proceed to the next judgment to determine whether there is infrared in the space within 30 seconds. If no, that is, no pressure data is detected within 10 seconds, clear this space timer and return to S21 to continue monitoring. S24. Is there infrared radiation in the space within 30 seconds? If yes, it means that infrared data has been continuously detected within 30 seconds. Proceed to the step of matching the identity from the pressure feature library. Call the pre-stored pressure feature library, match the identity of the person according to the pressure data, and then determine that the person with the specified identity has left this space. The process enters the termination step, and this determination process ends. If no, that is, no infrared data has been detected within 30 seconds, clear this space after the timer, and return to S21 to continue monitoring the space data.

Images