Building evacuation method and building evacuation system
By arranging multiple sensors in the building to detect abnormal flow patterns, the problem of how to effectively monitor and evacuate people in high-rise buildings in emergencies is solved, real-time flow monitoring and rapid discovery of abnormal situations is achieved, and evacuation efficiency and safety are improved sex.
Patent Information
- Application Number
- CN201810884144.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2018-08-06
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2038-08-06
AI Technical Summary
During an emergency, how to effectively monitor and evacuate personnel in high-rise buildings, especially to grasp the location and flow direction of personnel in real time, and timely detect abnormal behaviors to speed up the evacuation process is a major issue.
By arranging multiple 2D or 3D imaging sensors or depth sensors in the building, the graph data is captured and abnormal patterns are detected in the global coordinate system, such as crowded people, wrong movement direction, too fast or too slow, generating abnormal evacuation behaviors. in order to take targeted measures.
It realizes comprehensive real-time monitoring of people in buildings, accurately tracking personnel trajectories and flow information, and can promptly detect abnormal situations, improve evacuation speed and efficiency, and ensure personal safety.
Smart Images

Figure CN110807345B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to building evacuation during emergency events. In particular, the present invention relates to a building evacuation system and evacuation method. Background Art
[0002] In the event of an emergency, how to effectively evacuate people in buildings, especially high-rise or super-high-rise buildings, is a major issue in current social security work. During the evacuation process, it is important to accurately and in real time grasp the location of people, track people and know where they are going. Once abnormal behavior is found in the tracked flow of people, reasonable measures can be taken in time based on the information to speed up the evacuation process. Summary of the invention
[0003] The technical problem to be solved by the present application is to provide an evacuation method capable of detecting abnormal situations during the evacuation process in a building, the method comprising:
[0004] 1) capturing image data in at least one area through a sensor to track the movement status of people in the area;
[0005] 2) detecting an object specified by at least one abnormal pattern in the motion state of the tracked person, and matching the object with the abnormal pattern;
[0006] 3) generating an indication indicating abnormal evacuation behavior based on the matched abnormal patterns.
[0007] In the above evacuation method, the sensor includes a 2D or 3D imaging sensor or a depth sensor.
[0008] In the above evacuation method, a plurality of the sensors are arranged at positions with different viewing angles in the area, so that the plurality of sensors have adjacent or partially overlapping continuous viewing areas.
[0009] In the above-mentioned evacuation method, the object includes at least one of the following parameters: crowd moving speed, crowd moving acceleration, crowd moving direction, crowd density, crowd width, crowd height, crowd area, crowd volume, number of people, individual moving speed, individual moving acceleration, individual moving direction, individual height, individual volume, and individual position.
[0010] In the above evacuation method, the abnormal mode includes any one of the following situations or a combination thereof:
[0011] 1) Crowded crowds;
[0012] 2) Wrong moving direction;
[0013] 3) The movement speed is too fast, too slow or stops.
[0014] In the above evacuation method, the abnormal mode is predefined or defined according to the real-time situation of the building.
[0015] In the above evacuation method, the object specified by the abnormal pattern is detected in a global coordinate system, wherein the map data is converted from the sensor coordinate system to the global coordinate system by sensor calibration.
[0016] Another aspect of the present application is to relate to a building evacuation system, comprising:
[0017] a processing unit configured to process image data captured by a sensor in an area to track the motion state of a person in the area;
[0018] a matching unit configured to match an object specified by at least one abnormal pattern detected in the motion state of the person with the abnormal pattern;
[0019] A response unit is configured to generate an indication indicating that the evacuation behavior is abnormal based on the matched abnormal pattern.
[0020] In the above evacuation system, an abnormal mode definition unit for defining the abnormal mode is further included, and the abnormal mode definition unit includes any one of the following situations or a combination thereof:
[0021] 1) Crowded crowds;
[0022] 2) Wrong moving direction;
[0023] 3) The movement speed is too fast, too slow or stops.
[0024] The above evacuation system further includes a calibration unit configured to convert the map data from a sensor coordinate system to a global coordinate system so that the object specified by the abnormal pattern can be detected in the global coordinate system.
[0025] Another aspect of the present application is to provide a controller, which includes a memory and a processor, wherein the memory stores a program, and when the program is executed by the processor, any of the methods described above is implemented.
[0026] Another aspect of the present application is to provide a building evacuation system, which includes:
[0027] A data capture module, wherein the data capture module includes a plurality of sensors;
[0028] A control module, the control module comprising the above-mentioned controller;
[0029] A terminal module is configured to receive an indication from the controller indicating abnormal evacuation behavior and has a device for reflecting the indication.
[0030] This application can fully reflect the real-time situation in the monitored area, track the trajectories of pedestrians in the area, and obtain accurate information on abnormal human movement, so that targeted measures can be taken to help people evacuate.
[0031] This application uses multi-angle pedestrian tracking technology to implement an intelligent evacuation solution. By continuously detecting and using sensors to track people, the location and flow of people can be understood during the evacuation process. Based on this information, better evacuation measures can be taken to improve the speed and efficiency of evacuation and ensure personal safety. This application can contribute to the development of intelligent building operation application technology.
[0032] Other aspects and features of the present application will become apparent from the following detailed description with reference to the accompanying drawings. It should be understood, however, that the drawings are designed for illustrative purposes only and are not intended to limit the scope of the present application, as reference should be made to the appended claims. It should also be understood that the drawings are intended only to conceptually illustrate the structures and processes described herein and, unless otherwise indicated, are not necessarily drawn to scale. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] The present application will be more fully understood by referring to the following detailed description of the specific implementation modes in conjunction with the accompanying drawings.
[0034] Figure 1 A schematic diagram of the evacuation system involved in this application;
[0035] Figure 2 An embodiment of sensor arrangement and collection in the evacuation method involved in the present application is shown;
[0036] Figure 3 A schematic diagram of a control module in an evacuation system involved in this application;
[0037] Figure 4 This is an embodiment of the evacuation method involved in this application. DETAILED DESCRIPTION
[0038] In order to help those skilled in the art to accurately understand the subject matter for which protection is sought in the present application, the specific implementation of the present application is described in detail below in conjunction with the accompanying drawings.
[0039] The evacuation method and evacuation system involved in this application are used for emergency evacuation of people in buildings. Buildings include but are not limited to high-rise buildings, large buildings, public facilities, indoor spaces with people (which can be open areas or closed areas), etc. This application can monitor the evacuation of people from the building, and can promptly detect abnormal situations during the evacuation process, and then report them to the building management staff to take targeted measures. Figure 1 As shown, the system includes at least a data capture module, a control module and a terminal module. The data capture module includes multiple sensors for collecting image data and sends the image data to the control module for data calculation, processing, analysis and judgment. Subsequently, the control module sends an indication indicating the data processing result to the terminal module, and the manager is informed of the emergency situation occurring in the building through the terminal module.
[0040] Multiple sensors capture image data about the internal conditions of a building and / or the movement of people. The sensors include 2D or 3D imaging sensors or depth sensors, including but not limited to 2D imaging sensors, 3D imaging sensors, and various depth sensors. Multiple sensors have adjacent and continuous fields of view or partially overlapping continuous fields of view in space. For a relatively large area, the field of view of a sensor is limited and it is impossible to observe the entire area, so multiple sensors are arranged at multiple viewing angles. Figure 2 As shown, the figure shows the arrangement of sensors in a room. In this area, there are five sensors, which are arranged in a circle around the area in the same area. There is a continuous spatial sensing range of adjacent fields of view or partially overlapping fields of view between two adjacent sensors, so the image data in the area can be obtained in all directions and completely. The imaging sensor can be a conventional camera. Depth sensor technologies and devices include but are not limited to structured light measurement, phase shift measurement, time of flight measurement, stereo triangulation device, light triangulation device board, light field camera, coded aperture camera, computational imaging technology, simultaneous positioning and mapping components (SLAM), imaging radar, imaging sonar, scanning LIDAR (Light Detection and Ranging), flash LIDAR, passive infrared (PIR) sensor and small focal plane array (FPA) or a combination of at least one of the foregoing. Different technologies may include active (transmitting and receiving signals) or passive (receiving signals only) and may operate under the electromagnetic or acoustic spectrum (such as vision, infrared, etc.).
[0041] The sensor scans or shoots at a certain rate to obtain multiple image data reflecting the regional conditions, where the regional conditions include visible individuals, crowds and / or backgrounds appearing in the area. The image data captured by general imaging sensors are picture data, and the image data captured by depth sensors are depth map data. These image data are sent to the control module for processing. The control module has a controller for executing the evacuation method involved in the present application. The controller includes a memory and a processor, and the memory stores a program. When the program is executed by the processor, the above-mentioned evacuation method can be implemented.
[0042] See also Figure 3 , the controller includes at least a calibration unit, a processing unit, an abnormal pattern definition unit, a matching unit and a response unit.
[0043] The calibration unit is configured to perform sensor calibration, and data related to the regional situation is recorded in the sensor coordinate system by establishing a mathematical relationship, and then the sensor coordinate data is mapped to the global coordinate transformation matrix. The sensors may not have the same sensor coordinate system, and the associated data in different coordinate systems can be transformed into a consistent coordinate system through the transformation matrix. The evacuation system involved in the present application uses a transformation matrix to obtain the position of the tracked target or the specified stationary target in the global coordinate system, so as to perform subsequent data calculation, detection, processing, etc. in the global coordinate system. The calibration unit can be a separate unit or a part of the processing unit described below.
[0044] The processing unit is configured to identify the intention of the mobile target and predict the target behavior based on the global coordinate system, thereby achieving the purpose of tracking. Tracking can be considered as a question of what the probability of a specific system state is given the previous system state, observations and uncertainties. Identifying a mobile target includes but is not limited to whether the target is approaching, leaving or passing a certain location. The graph data can be processed to provide a large number of physical quantities about the movement of the target in the global coordinate system, including but not limited to position, speed, acceleration, direction of movement, distance from a specified background, etc. The tracking process includes two steps: prediction and update. By building a model, the target state in the previous graph data is transferred to the current graph data for prediction. In the update step, the target in the current graph data is detected, and the detected target is associated with the predicted target, so that the real-time situation of the target can be grasped.
[0045] The abnormal mode definition unit is configured to define the situation where the target behavior is abnormal. Multiple abnormal modes can be set. The definition of the abnormal mode can be predefined, such as set by referring to an empirical value or manually set, or can be set according to the real-time situation of the system. Thresholds or conditions are set for each abnormal mode. The same abnormal mode can also have multiple possible threshold settings and condition requirements. Thresholds and conditions can be pre-adjusted to meet various abnormal mode situations or verified manually. Thresholds refer to physical quantities related to target movement, including numerical values and / or numerical ranges, and thresholds can be static or dynamic. Conditions can be a state composed of multiple thresholds or a state combined with thresholds and non-thresholds, for example, other information that is not related to the moving target but is related to the physical environment in the area or involves the regional layout is also referenced, such as temperature in the area, air quality, whether other sensors or alarm devices in the area are working properly, etc. When the evacuation system of the present application is part of a building security system, the abnormal mode definition module can also communicate with other systems in the security system to more thoughtfully define the mode and set thresholds and conditions corresponding to different modes. The abnormal mode definition unit can define the mode from the beginning when the controller implements the evacuation method. The threshold and condition corresponding to the mode can be reset, adjusted and / or manually set, checked or corrected. The threshold and condition can remain unchanged during the implementation process, or can be updated or generated according to the actual situation during the implementation process. New thresholds and conditions can be manually set, checked or corrected. The abnormal mode definition unit can be a separate unit or a part of other units (such as a processing unit or a matching unit).
[0046] The matching unit is configured to extract the corresponding data processing results from the processing unit based on the thresholds or conditions of each mode defined by the abnormal mode definition unit. The matching unit generates a plurality of detection subunits, each of which includes the detection object specified by the corresponding abnormal mode, and the objects include the following: crowd flow speed, crowd flow acceleration, crowd flow direction, crowd flow density, crowd flow length, crowd flow height, crowd flow area, crowd flow volume, number of pedestrians, individual movement speed, individual movement acceleration, individual movement direction, individual height, individual volume, individual position, etc. The processing unit will provide the corresponding detection object to the detection subunit based on the tracking result. Due to the dynamic nature of the moving target, the processing unit needs to continuously obtain the information of these detection objects to grasp the situation in the building in real time. A timer can be set in the processing unit to regularly update the detection object in the detection subunit. In addition, when the definition of the abnormal mode changes, the specified detection object changes immediately, and the matching unit also needs to immediately generate a new detection subunit to meet the needs of the new abnormal mode. After obtaining the detection object, a comparison subunit can be provided to compare and judge the detection object with the threshold and / or condition of the abnormal mode. The matching process is also repeated and carried out continuously, and it can be based on a preset time period or a newly generated abnormal pattern operation until a match is found. At this point, the system will believe that there may be abnormal evacuation behavior in the area. After confirmation, the system will generate a diagnostic result for executing the next step. The matching unit maintains a connection with the processing unit and the abnormal pattern definition unit. The matching unit sends a request to the processing unit to extract the detection object, and the processing unit sends the corresponding data information to the matching unit. The detection object and matching requirements determined by the matching unit are based on the pattern customization of the abnormal pattern unit. At the same time, the definition benchmark of the abnormal pattern definition unit can also be adjusted with reference to the data of the matching unit.
[0047] The response unit is configured to generate an indication indicating abnormal evacuation behavior based on the diagnosis result. The indication can be generated as a single indication, or multiple indications reflecting specific evacuation behaviors, or multiple indications corresponding to abnormal pattern types, or converted into a certified output of the system. The indication can be directly reported to a person and / or a machine and is recognizable. The indication can be a readable, visual, or audible alarm signal, and the building management personnel can take active and effective countermeasures after receiving it; the indication can also be input into the system central controller, and the central controller responds to generate new commands to control and coordinate the operation of various facilities in the building; the indication can also be a combination of the above two indications.
[0048] The indication indicating abnormal evacuation behavior is sent to the terminal module of the evacuation system for display. The terminal module can be the output part of the evacuation system or the central system of the building, including but not limited to the central control console, a display with human-computer interaction equipment, an alarm, etc., or a mobile terminal that wirelessly communicates with the aforementioned controller, etc. The report can be given to the operator of the terminal module in the form of original image data collected by the sensor, processed image data, or converted into other easily understandable text, images, audio and video, etc.
[0049] Combine the following Figure 4 The evacuation method involved in this application is explained by taking an example of a fire in a building.
[0050] Start multiple sensors to collect image data for the same area, such as an aisle. The fields of view between the multiple sensors constitute a continuous field of view that reflects the panoramic view of the area. The sensor transmits the image data to the control module to track the movement of people in the aisle. Through sensor calibration, the image data is converted from the sensor coordinate system to the global coordinate system, and the speed, direction and density of the people flow are obtained after processing.
[0051] Abnormal modes are defined as three modes. The first is the situation where the flow of people in the aisle is too crowded, in which case the flow of people is slow and the density of people is too high. The second is that the flow of people is moving in the wrong direction. The third is that the flow of people is moving too slowly. According to the first situation, the flow of people speed and density data are extracted, according to the second situation, the flow of people direction data is extracted, and according to the third situation, the flow of people speed data is extracted, and these detection objects are compared with the thresholds or conditions of the abnormal modes that have been set. If they meet, it means that the corresponding abnormal mode is established, and then the control module sends the matching result to the terminal module. These thresholds or conditions corresponding to each abnormal mode can be pre-adjusted to meet different situations, and can be accepted by system operators such as building managers for inspection and verification. Corresponding measures can be taken to deal with different abnormal modes. For the first situation of crowded crowds, firefighters can be sent into the relevant locations in the building to clear the crowds; for the second situation of incorrect crowd movement, the direction of crowd movement can be corrected through broadcast guidance or by activating the building's dynamic guidance system; for the third situation of too low crowd speeds, firefighters can be sent into the building to carry out rescue operations in consideration of the risk to life.
[0052] These three situations are not limited to the data objects specified above. For example, in the first situation, crowd congestion is largely caused by people falling, or people with limited mobility affecting the evacuation speed, or there is a collision of people. Therefore, for the first situation, you can also specify the height of the crowd to be detected or combine the direction of crowd movement or the direction of individual movement. For another example, in the second situation, if the direction of crowd movement is completely reversed, it is possible that the image data of the firefighters is collected, so the step of identifying moving pedestrians can be added. For another example, if the speed of people is too fast or even zero, it is also an abnormal evacuation behavior. The detection object can be added to the definition of the third situation, or it can be detected as the fourth abnormal mode (the speed of people is too fast) or the fifth abnormal mode (the speed of people is zero).
[0053] In addition to detecting data related to human flow, it is also possible to detect data related to individual movement or to combine data on human flow and individuals to match abnormal patterns.
[0054] The terminal module can be a screen in the central monitoring room of the building, which can output in various forms such as direct images and maps to inform management personnel which area's evacuation behavior is abnormal. Firefighters can also hold portable terminal modules, such as PDAs with real-time display and intercom functions, which are directly connected to the control module of the evacuation system or wirelessly connected to the control console in the central monitoring room of the building. Firefighters can keep abreast of the rescue situation of the building in a timely manner.
[0055] In terms of countermeasures, other targeted measures may be taken in addition to those described above. For example, by detecting the height of the crowd in the graph data, if a pedestrian who relies on a wheelchair is found among the evacuated crowd, an elevator may be temporarily started.
[0056] Although specific embodiments of the present application have been shown and described in detail to illustrate the principles of the present application, it should be understood that the present application can be implemented in other ways without departing from such principles.
Claims
1. A building evacuation method, characterized in that: 1) Arranging a plurality of sensors at positions of different viewing angles in a region so that the plurality of sensors have adjacent or partially overlapping continuous viewing areas; 2) capturing image data of the area through the multiple sensors to track the movement status of people in the area; 3) detecting an object specified by at least one abnormal pattern in the motion state of the tracked person, and matching the object with the abnormal pattern; as well as 4) generating an indication indicating abnormal evacuation behavior based on the matched abnormal patterns.
2. The method according to claim 1, characterized in that: The sensor is a 3D imaging sensor or a depth sensor.
3. The method according to claim 1, characterized in that: The object includes at least one of the following parameters: crowd moving speed, crowd moving acceleration, crowd moving direction, crowd density, crowd length, crowd height, crowd area, crowd volume, number of people, individual moving speed, individual moving acceleration, individual moving direction, individual height, individual volume, and individual position.
4. The method according to claim 1, characterized in that: The abnormal mode includes any one of the following situations or their combination: 1) Crowded crowds; 2) Wrong moving direction; 3) The movement speed is too fast, too slow or stops.
5. The method according to claim 1, characterized in that: The abnormal mode is predefined or defined according to the real-time situation of the building.
6. The method according to claim 1, characterized in that: The objects specified by the anomaly pattern are detected in a global coordinate system, wherein the map data is transformed from the sensor coordinate system to the global coordinate system by sensor calibration.
7. A building evacuation system, characterized in that include: A plurality of sensors, wherein the plurality of sensors are arranged at positions of different viewing angles in the region, so that the plurality of sensors have adjacent or partially overlapping continuous viewing areas; a processing unit configured to process the image data of the area captured by the plurality of sensors to track the motion status of people in the area; a matching unit configured to match an object specified by at least one abnormal pattern detected in the motion state of the person with the abnormal pattern; as well as A response unit is configured to generate an indication indicating that the evacuation behavior is abnormal based on the matched abnormal pattern.
8. The evacuation system according to claim 7, characterized in that: The sensor is a 3D imaging sensor or a depth sensor.
9. The evacuation system according to claim 7, characterized in that: The object includes at least one of the following parameters: crowd moving speed, crowd moving acceleration, crowd moving direction, crowd density, crowd length, crowd height, crowd area, crowd volume, number of people, individual moving speed, individual moving acceleration, individual moving direction, individual height, individual volume, and individual position.
10. The evacuation system according to claim 7, characterized in that It also includes an abnormal pattern definition unit configured to define the abnormal pattern, wherein the abnormal pattern definition unit includes any one of the following situations or a combination thereof: 1) Crowded crowds; 2) Wrong moving direction; 3) The movement speed is too fast, too slow or stops.
11. The evacuation system according to claim 10, characterized in that The abnormal pattern definition module is configured to predefine the abnormal pattern or define the abnormal pattern according to the real-time situation of the building.
12. The evacuation system according to claim 7, characterized in that: A calibration unit is also included, the calibration unit being configured to transform the map data from a sensor coordinate system to a global coordinate system so that the object specified by the anomaly pattern is detected in the global coordinate system.
13. A controller comprising a memory and a processor, wherein a program is stored in the memory, and when the program is executed by the processor, the method according to any one of claims 1 to 6 is implemented.
14. A building evacuation system, characterized in that include: A data capture module, wherein the data capture module includes a plurality of sensors; A control module, the control module comprising the controller according to claim 13; as well as A terminal module is configured to receive an indication from the controller indicating abnormal evacuation behavior and has a device for reflecting the indication.
